Imported outline

PRECIPICE OF POWER
The Data Hidden in Your Migraines
How One Billion Migraine Sufferers Can Fight Back with Their Own Data

Roger Mishoe
Founder, Precipice Health, LLC
U.S. Navy and Marine Corps Veteran
Chronic Migraine Patient

Working manuscript — Full manuscript — Preface through Epilogue
 
CONTENTS

Preface
Chapter One
The Invisible Disease
Chapter Two
What They Never See
Interlude
Kitchen Table
Chapter Three
The Collapse Between Experience and Evidence
Interlude
Nineteenth
Chapter Four
The Seven Dimensions
Chapter Five
The Leaky Boat
Chapter Six
The Threshold Model
Chapter Seven
The Parallel Pipe
Interlude
Sunday Evening
Chapter Eight
The Clinician One-Page
Chapter Nine
The Disability Evidence Chain
Chapter Ten
Living With the Instrument
Chapter Eleven
What the Instrument Cannot Do
Chapter Twelve
What You Owe
Chapter Thirteen
The Wider Field
Chapter Fourteen
The Precipice
Epilogue
 

Preface
I am not a doctor. I have no clinical training. I have no medical degree. I am not a neurologist, not a pain specialist, not a researcher in the pathophysiology of chronic disease. If you have picked up this book hoping I am going to explain what is happening in your brain during a migraine attack — the electrochemistry of it, the neuroanatomy of it, the cellular biology of cortical spreading depression — I am not the author who can do that for you. There are better books than this one for that, written by doctors who have spent their careers on it, and I would send you to them without hesitation.
What I can speak to, confidently, is how migraines have impacted me and what I have done to understand them. That is the entire authority behind this book. I am a chronic migraine patient who happens, by the accident of my professional life, to have spent twenty-five years building evidence architectures for the federal government — and who eventually, in the collapse of the career that skill was supposed to serve, turned that same discipline inward. This book is what I built. It is not a cure. There is no cure. I have to say that plainly and early, because the world of chronic illness is full of people who will offer you cures and take your money for them, and I refuse to be one more voice on that pile. What I offer you instead is an instrument. A specific set of frameworks that, if you use them, will let you see what your body has been telling you for years.
I also want to be honest about who else this book is for.
I have written it, primarily, for people who share my disease. I know migraine from the inside. I have logged eight thousand rows of my own data across a period during which the disease was taking my career from me, and I built the frameworks in the pages ahead in the language of migraine because that is the language I have lived in. But — and I have thought about this carefully, and I have been careful not to overreach — the same frameworks apply, with modest translation, to a whole set of sister conditions that share migraine's specific structural cruelty. The cruelty of not being able to be seen.
Fibromyalgia. Endometriosis. Myalgic encephalomyelitis and chronic fatigue syndrome. Long COVID. Postural orthostatic tachycardia. Interstitial cystitis. Chronic Lyme. Ehlers-Danlos. The list is longer than that, and I am not qualified to give you a complete taxonomy of it. What I can tell you is that the people I have talked to who live with these conditions describe an experience that structurally resembles mine — the invisibility, the disbelief, the failed evidentiary chain between what their body is doing and what the medical, insurance, and disability systems are able to read about it. If any of that resonates, then this book is also for you. Not because I am an authority on your disease. I am not. But because the instrument I built for mine may, with translation, help you build one for yours.
I want you to know that this book sees you. I want you to know that it walks with you. Not as an expert. As a partner. Hand in hand to your own precipice — because there is a moment, if you do the work I am going to teach you to do, when the accumulated weight of the evidence you have built shifts something. A neurologist's tone. An adjudicator's decision. A spouse's willingness to stop asking whether you are doing enough. Your own quiet permission to trust what your body has been saying to you all along. That is the moment. That is what I am calling the precipice of power. It is not power over the disease. It is not a cure. It is the specific, hard-earned power of being finally seen — first by yourself, and then, because you have built the record, by the world around you.
The battlefield in the chapters ahead is migraine, because that is where I know the terrain. The instrument is generalizable, because the enemy is the same. I hope, if you have any invisible illness at all, you will read the migraine chapters with translation in your hand. I have tried, at the end of each of the framework chapters, to leave you a short note showing how the framework travels to some of the sister conditions I named above. I am not qualified to write those notes with the depth I have written the migraine material. I am qualified to say: the instrument does not care what your disease is called. The instrument cares only whether you build it.
Now Chapter One.
 
Chapter One
The Invisible Disease
The conference room lights were fluorescent, which meant they were weapons.
I remember calculating the distance to the door — three steps, maybe four, past a program manager and a contract officer I could not afford to insult — and understanding, with the clarity that only comes when your visual field is beginning to fracture at the edges, that I was not going to make it through the meeting. I was going to make it through the next thirteen minutes of the meeting. And then I was going to leave the room in a way that looked deliberate, because Marines do not stagger out of conference rooms, and even fifteen years after taking the uniform off I still owed the version of myself I had been an accounting for how I carried myself under load.
I did not know, sitting there, that I was watching my own neurology issue a warning order.
I did not know that the fluorescent tubes above me were only the acute stimulus in a system that had been drifting toward failure for days — since a bad night of sleep sometime that week, since the barometric pressure had dropped four millibars while I slept, since I had missed the small window of morning water because a call went long and a shower went short. I did not know that migraine is not an event. I only knew that I had a briefing to give and I could no longer read the words on my own slide.
I made it out. Marines make it out.
I sat on the floor of a federal bathroom stall with my forehead against the tile for forty minutes, and when I stood up my career had not ended and my day had not ended and the tile was still there and the fluorescent tubes were still there and I understood, in the still-thinking part of my brain that had not yet gone dark, that I was going to have to solve this problem the way I had been trained to solve every other hard problem in my life.
I was going to have to measure it.
I did not know, that afternoon, that I had roughly two years left in the career I was standing inside of. I did not know that in a small number of months I would begin to keep a log — a bad one at first, thirty-seven columns wide, most of them abandoned by the third week — and that the log would eventually become an instrument, and that the instrument would eventually become the reason I could speak clearly about a disease that had been trying, for the better part of a decade, to erase me from my own life.
I did not know that a woman named Christa was going to walk into that story about a year later.
I did not know very much of anything, sitting on that bathroom floor, except that the round had already left the tube and I was going to have to update the model and prepare for the next one.
I was an artillery officer.
That means the tools of my trade were angles, atmospherics, and load-outs. Wind speed. Powder temperature. Charge selection. Time of flight. The math a howitzer battery does in ninety seconds is the same math I was taught in five years of mechanical engineering at Southern University A&M College — thermodynamics, statics, dynamics — except at Southern nobody died if I miscalculated, and in the Marine Corps, they could.
I never practiced mechanical engineering as a civilian trade. I went from Navy boot camp in 1993 through a program called BOOST — Broadened Opportunity for Officer Selection and Training, a Navy officer-preparation pipeline for enlisted sailors — into a full ride at Southern, and out the other side in June of 1999 as a second lieutenant of Marines. Regular commission. Active duty. My primary military occupational specialty was 0802: field artillery officer.
I deployed to Japan in 2001. I deployed to Luzon in 2002 for Operation Balikatan — shoulder-to-shoulder — where the 3rd Battalion, 3rd Marines was pushing alongside the Armed Forces of the Philippines against a group called Abu Sayyaf. We ran a screening mission and helped train Filipino Marines while other elements of our battalion landing team took the fight into the jungle. In October of that year I reported to the Naval Postgraduate School in Monterey. I earned a master's degree in operations research. I left the Marine Corps as a captain on 31 March 2006.
That is where a lot of memoirs would place a clean line. Marine deploys, Marine comes home, Marine moves on. I would like to write that sentence. It is not the sentence I have.
I left the Marine Corps in 2006 because my knees would no longer let me stay. Four surgeries in eight years — the kind of quiet, cumulative damage that lives inside the training calendar of any 0802 who has spent a career keeping pace with the infantry he was there to support. As Artillery Liaison Officers we did not stay behind the guns. We went forward with the grunts — to learn the ground they would move across, to be close enough to hear the call for fire when they needed indirect support, and to prove that the King of Battle had earned the dignity Frederick the Great once said artillery gives to what would otherwise be an ugly brawl. I loved that work. I loved the branch. I followed my brother into it, and he set a standard I did not intend to fall short of — not for my sake, and not for his.
I woke up one morning and I could not run. The Marine Corps offered me permanent light duty, and I turned it down. Not out of pride. Out of the plainest fact of the job description. Marines lead by example. You do not stand in front of a formation, clipboard in hand, and assess the physical fitness of the Marines you are supposed to lead if you cannot do the fitness with them. That is not what we do. That is not who we are.
I did not choose to leave. I was made to leave, gently, by a body that had used up its margin. I did what I was taught to do. I came home. I moved on.
That was Act I of my working life. An act that survived four surgeries and eight years of trying to keep pace with a career I loved.
This book is about Act II.
This time, I am not moving on. This time, I am fighting back. This time, I have found a weapon my knees never gave me, and I intend to teach you how to use it.
I came out of the Marine Corps in the spring of 2006 and went to work as an operations research analyst in a small consulting firm in Stafford, Virginia. That is a job title, not a story. I sat in a windowless room and turned raw data — logistics data, readiness data, budget data — into decision-quality intelligence for people who had to move billion-dollar programs on the strength of what I put in front of them. It suited me. Operations research is applied artillery for civilians. You look at a noisy environment, model it, calculate what you can control and what you cannot, and produce a number a decision-maker can defend.
Over the next six years I rose from analyst to senior director inside that firm, running analytics and logistics work in support of the Marine Corps I had just left. In 2012 I moved back inside the government, this time as a Department of Defense civilian — Director of Studies and Analysis for Marine Corps Logistics Command in Albany, Georgia. Three years later I moved again, into the Treasury Department, as Director of Enterprise Data Management. In 2021 I became the Acting Chief Data Officer for the Department. Later that year the acting was removed, and I sat in a Senior Executive Service seat as the Chief Data Officer of the United States Department of the Treasury, responsible for setting the department's data strategy across every one of its bureaus, each of which had its own CDO reporting into the enterprise structure I was building.
I mention that career trajectory not because I am proud of it, though I am, but because I need you to feel the shape of what was about to be taken from me. The disease I want to tell you about did not take an ordinary career. It took a Senior Executive of the United States government whose entire professional life had been organized around the discipline of turning ambiguous inputs into decision-quality intelligence.
It took, in other words, the exact person best equipped in the world to see it coming.
And I did not see it coming.
The migraines began after service. I want to be careful about that, because a great deal of chronic-illness writing wants to locate the beginning in a specific event — a fall, a blast, a fever, a grief. Mine has no such story. I came out of the Marine Corps intact. The migraines began later, in the second act of my working life, in a series of meetings and airports and hotel rooms that resembled each other so completely that I could not tell you now where the first bad one hit me.
I remember only that at some point I began to notice a pattern: I was losing days. Not weeks — days. A day here, a day there, sometimes two consecutive days, blurred by pain and by the pharmacology I had been given to manage it, edited out of my calendar with the same clean stroke I had once used to edit low-priority training events off a training schedule when a higher priority came in.
There were more of them each year.
The two years I want to tell you about ran from January of 2023 to December of 2024. In those twenty-four months I lost eleven hundred and forty-seven work hours to migraine disease.
Eleven hundred and forty-seven work hours. Roughly the equivalent of thirty full weeks of the professional life of a Senior Executive of the federal government, gone. Not vacation. Not travel. Not administrative leave. Not the ordinary attrition of an executive schedule. Gone. Hours during which I was supposed to be present at a briefing, on a call, in a room, running a meeting, chairing a governance board, drafting an authorship line on the Department's Strategic Plan — and was instead in a dark room, or a bathroom, or a bed, or on the floor of my office with the lights off, holding still against a pain that had, by then, ceased to be interesting to describe.
I say the number, in full, because numbers are how I think. Numbers are how you get an executive to believe you. Numbers are how you build a case to yourself when the case to yourself has been getting harder to make.
But I also say the number because those twenty-four months were not only when the disease took my career. They were the twenty-four months during which I built the instrument.
Both things happened at the same time.
That is not a narrative device. It is the fact of the matter. From January 2023 forward, every day the migraines were taking hours from me, I was pouring hours back into the instrument — the spreadsheet that eventually became a data model, that eventually became a framework, that eventually became seven of them: the Seven Dimensions, the Leaky Boat, the Threshold Model, the Evidence Funnel, the Clinician One-Page, the Disability Evidence Chain, and a set of documentation protocols I now teach patients to use inside twenty minutes.
I built those instruments during the collapse of the career they could not save.
In February of 2024 I went on FMLA.
On the fifth of July, 2024, I left work for what I told my Department would be at least a month — long enough, I thought, to get the migraines under control at last. I had committed to seeing a holistic provider that summer, and for a stretch I saw real progress. The attacks softened. The gaps between them lengthened. There were days when I looked at myself in the mirror and thought I might make it back.
The progress minimized. The body returned to its usual signals. Yawning. Stiff necks. Auras. Blurred vision. The migraines came back stronger than they had ever been.
In November of 2024 I understood, with the same clarity I had once understood the fluorescent lights would take me down at the meeting, that I was not going to work again. Not in that role. Probably not in that career.
I did not fall apart. I doubled down on the instrument.
I want you to understand what that November looked like from the inside. The career I had built for twenty-five years was over. The marriage I had lived inside was over — I had separated from my former spouse in 2022 and the divorce had been final for exactly one year that same month. The person I had been about to become — Chief Data Officer of the United States Department of the Treasury for another three, five, ten years, however long the mission held — was no longer available to me. Everything the disease had been trying to do since the first bad meeting, it had finally succeeded in doing.
And in the middle of that ruin, one thing was still upright.
The tool.
The tool, which had cost me two years of my life to build, was working. It was catching prodromes forty-eight hours out. It was telling me, in the language of numbers, what my body had been trying to tell me for a decade in a language none of my doctors could read. The disease had won every other engagement it had picked with me. But the instrument I had been building against it — quietly, on nights and weekends and stolen executive hours — was doing the one thing the disease had never expected any of us to do.
It was seeing the disease.
On the nineteenth of December, 2024, I received notice of termination from my position as the Chief Data Officer of the United States Department of the Treasury for medical inability to perform the duties. I signed the paperwork. I turned in my badge. I walked out of the Department for the last time on a cold afternoon in Washington, D.C., with a small cardboard box of the kind you have seen in every layoff scene ever written, and in that box, on top of a coffee mug and a framed photograph of my sons and a challenge coin from the Marine Corps, was a laptop containing eight thousand rows of the most granular, longitudinal, framework-encoded personal migraine data ever recorded by a single person, so far as I have been able to determine.
The Treasury had let me go. But I had not left empty-handed.
Christa was there.
She had been there since January of that year — since eleven months before the termination, since six weeks before the FMLA, since the entire acute phase of what would eventually take me. She met me nearly fully broken. She met me while my divorce from my former spouse was still recent, while my career was still unsteady on its feet, while I was still convincing myself I could go back to work if I could just find the right doctor, the right supplement, the right neurologist. She is an athletic trainer by profession, and most recently an adaptive physical education teacher — which is to say, her life's work is teaching bodies with limitations to move again. I do not think it is an accident that she is the woman who is presently my wife.
Christa did not save me. Christa was with me as it happened. She did not pull me out of the ruin — the ruin had its own gravity, and there was a period during which nothing was going to keep me from going into it. What she did — and I say this in full awareness that it is the smallest and largest thing anyone can do for anyone else — was refuse to look away.
That is not a spouse's job. It is a witness's job.
You will see her name a few more times in this book. You will not see much of our life together, because our life together is ours and not the reader's. But you should know she is in the room. She has been in the room since the room began to be a room worth being in.
This book is not what you may think it is.
It is not about a cure. There is no cure for migraine disease, and if there were I would not be qualified to sell it to you. It is not about a startup, though a company has grown up around the work described in these pages. It is not about my career, though my career is the reason I know what an executive-grade data instrument looks like, and it is the reason this book has any teeth. It is not a spiritual memoir, though I have prayed a great deal about this over the last several years and I would not want to be here without prayer.
It is a book about what I built with my hands during two years when everything else was being taken from me.
It is a book about a specific set of frameworks that took twenty-four months of some of the hardest work of my life to develop, and that I want you to have because I did not have them and I would very much have wanted them. The Seven Dimensions — because migraine is not one thing, and a scale of one to ten cannot hold it. The Leaky Boat — because migraine is not an event, it is a threshold system, and the water rises before you feel it. The Threshold Model — because triggers do not act alone; they interact, they accumulate, they carry the boat closer to the waterline. The Evidence Funnel — because the system reads what is written down, not what happened. The Clinician One-Page — because a neurologist has twelve minutes to read your last three months of life. The Disability Evidence Chain — because a single artifact will not save you; a sequence of linked evidence will.
You will learn each of these in the chapters ahead. Not as theory. As instruments. Same way I learned my howitzer.
It is a book about a specific set of institutions — insurers, disability adjudicators, primary-care physicians who have twelve minutes with you, specialists who see thirty patients before lunch — that cannot read what you are trying to tell them, because the language they read in was not written by patients like you and me. And it is about what you can do to translate.
It is a book about the fact that data, at long enough time horizons and high enough resolutions, becomes something more than data. It becomes evidence. And evidence, in a world of adversarial systems and busy clinicians and skeptical carriers, is the only currency of transparency that any of us has. It is the gateway to evidence. It is, for the chronically ill, the precipice of power.
If that phrase sounds grand, I would ask you to sit with it. I did not write it to sound grand. I wrote it in a hotel room in Falls Church, Virginia, on a cold night in the fall of 2024, thirty-one hours into an attack that would not break, after two failed abortives and a failed nap, staring at a spreadsheet that had, for the first time in nearly two years of collection, told me — thirty hours before the attack began — that this attack was coming.
The system had failed. My body had failed. But the tool had not failed.
That night, for the first time, I understood what I had built. I built it for me.
I built it because I could no longer trust my own memory across an attack — because migraine attacks eat memory, especially the memory of what preceded them, and if I could not trust my memory then I needed an instrument that could not forget. I built it because the doctors were doing their best inside a system that did not give them the inputs they needed to help me. I built it because I had roles to hold: as a father, as a leader, as a Senior Executive of the government, and eventually simply as a husband to a woman named Christa who deserved to marry a whole man.
I did not build it to sell. I did not build it as a business. I did not have a co-founder or a deck or a name for a company. I had a spreadsheet, and then I had a data model, and then I had a set of frameworks, and then — long after the frameworks were doing their work — I had the growing understanding that if I did not put these instruments into the hands of the forty million Americans who share this disease, I would be guilty of a specific kind of unfaithfulness to what the disease had cost me to learn.
The tool eventually took a name. Migraine Data Detective. The company that grew, quietly and reluctantly, to house it — because I could not distribute a tool of that kind from a personal Google Drive folder to the people who needed it, and because the ethical and legal shape of what I had built required a container — took a name too. Precipice Health.
The company is not the point of the book.
The company is a consequence of the book being true.
You are going to learn, in the chapters ahead, how to see what your body has been telling you for years. You are going to learn how to build a record that a doctor can read in ninety seconds. You are going to learn how to write an appeal that an insurance carrier cannot dismiss. You are going to learn how to think about triggers the way an artilleryman thinks about wind — as an input to a calculation, not as an enemy.
You are going to be given, in short, the instrument I built for myself during the two years the disease was taking everything else.
I hope you find it useful. I hope you find it in time.
The fluorescent lights, in the end, were not the problem. The problem was that I had no instrument to measure the room before I walked into it.
By the end of this book, you will.
 
Chapter Two
What They Never See
There are, by the World Health Organization's most recent accounting, roughly one billion people alive on earth today who live with a primary headache disorder.
Headache disorders are the second-leading cause of disability in the world. Not the tenth. Not the fifth. The second. Ahead of depression. Ahead of every cancer combined. Ahead of stroke, of diabetes, of every one of the diseases that occupies the ordinary imagination when the word disability is spoken. The only condition that outranks it is low-back pain — a category so vast and so uncontested it functions as the ocean of human musculoskeletal suffering. Just behind that ocean, in the shallow water where we live and work and try to raise our children, sits migraine.
One in seven of us has it.
Forty million in the United States. Roughly a billion in the world.
And ninety percent of us are, at any given moment, being asked to explain to somebody why we cannot come to the meeting today, why we cannot pick up the children today, why we cannot drive today, why we cannot answer the question that has been asked of us, because a disease that no one in the room can see is, at that moment, taking us apart.
This is the argument of this chapter.
You cannot see it. Nobody can. That single fact — the invisibility of migraine disease — is the reason forty million Americans live inside a suffering the world has decided, by structural default, is not real enough to organize itself around. It is why the neurologists have twelve minutes and a photocopied grid. It is why the insurance carrier writes denied on the claim. It is why the boss says, Are you feeling better?, in the tone that means You have been out again. It is why the person you love more than anyone else on earth — the person you married — will one day, in the front seat of a car returning home from a good vacation, tell you gently that she thinks you could be doing more.
Not because she does not love you.
Because she cannot see it. That is what this chapter is about. We were coming home from Myrtle Beach.
Christa was driving. That is the first thing you should know, because it is the smallest, most quietly load-bearing detail I can offer you about my current life: I do not drive anymore. Not on trips. Not through the parts of the day when the light on the road has a certain slant to it. Not through the last hour of a long haul home when what is left of the visual system is starting to buckle. My wife drives me. She has been driving me since we started building a life together, and she does it without commentary, because that is who she is, and because the alternative is that we do not go anywhere.
It was late afternoon, tipping into early evening. The kind of thin, low, honeyed South Carolina light that used to be one of my favorite things about coming out of a good week at the beach. Four of our five children were in the car with us. So was a family friend. The two youngest — the nine-year-old girl and the eleven-year-old boy — were in the very back, negotiating something small and sacred that had happened at the pool three days earlier. My son is non-verbal and uses an AAC device, but he does most of his real arguing the way he does most of his real living — through grunts and moans and the exact shoulder-set of an eleven-year-old boy who is not going to be talked out of a position he has already claimed. Do not mistake the sound for weakness. He knows how to hold his own. His nine-year-old sister was making her case in full sentences and losing it anyway, which is roughly how the negotiation between them tends to go. The fourteen-year-old had her nose in a book, which is where she usually is. She reads the way some people breathe. She was not with us in the ordinary sense — she was somewhere else that a book had taken her — and she was giving the family the dignity of doing that quietly. My twenty-three-year-old was riding up front behind Christa, earbuds in, which is where the calmest person in the car generally ends up sitting.
I was in the passenger seat.
I had a good day. That mattered. I had gone through most of it with a pain level I could hold in the background — a three, maybe a four, which for a chronic migraineur is a good day, because a chronic migraineur has calibrated the meaning of the word good against a scale nobody else in the car uses. I had eaten dinner. I had held a conversation. I had, at one point in the afternoon, felt something almost like a full range of feeling — I had noticed the honeyed light. That does not always happen anymore.
We were maybe forty-five minutes out from home when Christa said the thing.
She did not say it as an accusation. She said it as an observation — the way an athletic trainer says something to a runner who has been favoring the wrong knee for too long. She has spent her adult life helping people move again after injury; her most recent work has been as an adaptive PE teacher, which means she teaches bodies that other bodies have written off. When she says something about a body, she says it with the authority of somebody who has watched people rebuild themselves for a living.
She said she did not think I was doing all I could do to manage the migraines, or to prevent them.
She said she had seen so many people come back — from ACL tears, from concussion, from spinal fusion, from things that would sink a person who was not doing the work. She had helped them come back. And she said she wondered, out loud, whether I was doing that work.
What happened next, in the space of three or four seconds inside my chest, was two things at once.
The first thing was the flare of a very old anger. It is not a good anger. It is the anger of a chronically ill person who has spent, by that point in the afternoon, seven years defending the medical reality of his own body against the ordinary and forgivable skepticism of everyone he loves. The anger is not directed at the person in front of you. The anger is directed at the accumulated weight of every conversation that came before this one, and it lives underneath your ribs like a coiled thing, and it does not know how to distinguish the conversation you are in from the other three hundred you have already had.
The second thing was the very familiar collapse of self-trust that follows.
Is it, though? the collapse says. Is it really as bad as you're making it out to be? Maybe you're not doing everything. Maybe she has a point. Maybe you're the one who is not seeing this clearly. Maybe you have been so inside this thing for so long that you have lost the ability to look at yourself and know what you are actually looking at.
That collapse is the reason I am writing this chapter, and it is the reason this book has a chance of being useful to you.
Because that collapse — maybe she has a point, maybe I am not really that sick, maybe I should be doing more — is the exact mechanism through which invisibility completes its work on a chronically ill person. The invisibility of the disease means the outside world cannot verify your suffering. And when the outside world cannot verify your suffering, you eventually stop being able to verify it yourself. That doubt is not weakness on the sufferer's part. It is the design of the system.
I sat with the anger and the collapse for a few seconds. The nine-year-old and the eleven-year-old were still negotiating in the back. Christa was watching the road, which is what you do when you have said something difficult in a car full of children and are giving the other person the dignity of choosing his response.
I said, as gently as I could, that she was not a chronic-pain patient.
I did not say it to hurt her. I said it because it is true, and because the difference between the person who has been in daily pain for years and the person who has helped hundreds of people recover from injury is not a difference of intelligence or of love or of moral seriousness. It is a difference of access. She has watched pain from the outside for a very long time. She is one of the best in the world at doing that. But she has never been the one carrying the pain across a decade, and there is a set of things you cannot know about pain until you have carried it that long.
I told her, in the front seat of that minivan, that when you are inside pain like this — the daily, drilling, layered pain of chronic migraine — your access to your own tools changes. You may own the tools. You may have built the tools. You may have spent two years of your life engineering the frameworks that are, in theory, at your disposal in that exact moment. And you still may not be able to reach them. Because the pain itself takes reasoning offline. Because what the pain wants, more than anything else, is for you to walk to a dark room and be quiet. And in the moment the pain is winning, silence and darkness begin to look like the only two available goods in the entire world.
She was quiet for a moment.
She said, I hadn't thought about it that way.
That is what she said, exactly. She is a fair person, and a listener, and she said it the way people say things when they are turning a piece of new information over in their hands and finding that it fits.
We drove a little further. I looked at the honeyed light on the road for as long as it lasted. The nine-year-old and the eleven-year-old finally worked out whatever had gone wrong at the pool. The twenty-three-year-old had his music on. Christa reached over — one hand on the wheel, one hand on my leg — and left it there.
That is the scene.
I am going to tell you what it means. The most hurtful part of living with an invisible disease is not the disease itself.
That is a strong sentence. I mean it exactly. The disease is worse in the acute moment — nothing outside a burning building beats the acute moment of a severe migraine — but the acute moment ends. What does not end is the requirement to explain yourself. Chronic invisible disease means that every day of your life you are asked, in some small or large way, to convince somebody of a fact about your body that they cannot see. Sometimes the person you are convincing is a doctor. Sometimes it is a claims and ratings evaluator. Sometimes it is your boss. Sometimes it is your children. Sometimes it is your wife in the front seat of a car returning from a good vacation. Sometimes it is yourself.
Every single one of those transactions costs you something.
I am going to call it — for the rest of this book, and possibly for the rest of my public life around this disease — The Convincing Cost.
The Convincing Cost is the emotional, cognitive, and physical price a chronically ill person pays for the daily labor of convincing other people that a real thing is real when that real thing cannot be seen. It shows up in three ways, and I want you to know all three, because knowing them is the first step in refusing to pay them.
The first form is The Performance Tax.
The Performance Tax is what you pay when you show up looking okay because the alternative is being treated as unwell in a world that has no framework for hearing you. You wear the clean shirt. You shave. You go into the meeting. You hold the smile. You make the joke. You perform the appearance of a person who is not in pain, because you have learned that the appearance is what earns you the right to be treated as a full participant in the room. The tax is real. It costs you sleep, it costs you recovery time, it costs you the honest self-report you need in order to survive.
You know you are paying it when you get home from a workday and you cannot speak in full sentences to the people you love, because the last three units of energy you had were spent performing wellness for people who did not know you were performing.
The second form is The Cost of Convincing.
The Cost of Convincing is what you pay in each individual conversation in which you have to defend the reality of your own body. It is what you pay when you sit across from a physician who has twelve minutes and is trying to decide whether you are a sincere patient or a difficult patient. It is what you pay when you sit across from a claims and ratings evaluator who is trying to decide whether the words you are writing add up to enough of a story to justify a check. It is what you pay when you sit in the front seat of a car returning from Myrtle Beach and are asked, in love, whether you are really doing everything you could. Every time you tell the story, some of it comes out. Every time you tell the story, you have less of the story to give to somebody else later.
The Cost of Convincing is why so many chronically ill people eventually go silent. Not because their pain stopped. Because they ran out of story.
The third form — and this is the one I care about the most — is The Cognitive Transference.
The Cognitive Transference is what happens when the Convincing Cost has been running long enough that you begin to doubt yourself. This is where invisibility does its most complete work. Because the outside world cannot see your disease, and you cannot make them see it, you begin — quietly, without noticing — to internalize their doubt as your own. You begin to ask yourself the questions they were asking:
Is it really that bad? Am I really that sick? What if they have a point? What if I have been dramatizing this? What if the doctor is right that this is anxiety? What if my wife is right that I could be doing more? What if the truth is that I am the malingerer everyone is quietly wondering whether I am?
I want to be careful here, because self-examination is a virtue in the chronically ill patient. You should be interrogating your own experience with rigor. That is the whole premise of the frameworks in this book. But there is a difference between the self-examination of a scientist studying his own condition and the self-abandonment of a patient who has been asked to defend himself so often that he has begun to volunteer the prosecution.
The Cognitive Transference is the moment you start prosecuting yourself.
It is the deadliest of the three costs, because it does something the other two cannot do. It cuts you off from the honest report of your own body. And the honest report of your own body is the only signal you actually have.
You cannot build a framework on data you have already talked yourself out of trusting.
I know the Cognitive Transference personally, and not just from the migraine years. I know it from my knees.
I have already told you, in the last chapter, that I lost my Marine Corps career in 2006. What I did not tell you is what it cost me to lose it slowly. The knees did not fail all at once. They failed the way any load-bearing structure fails when you refuse to report the load — a little at a time, then a lot at a time, then all at once on a morning I could not run.
I had four surgeries in eight years. Between those surgeries, I performed wellness. I performed it for my Marines, for my commanders, for the physical fitness tests I was supposed to lead by example, and — most of all — for myself. Every time a knee gave me an early warning, I told it to be quiet. Every time it gave me a middle warning, I told it that Marines run through worse than this. Every time it gave me a late warning, I told it, in the voice I had learned to use inside my own skull, that this was not the day I was going to admit to weakness.
The knees, being knees, kept their own record.
They did not care that I was performing. They did not care that I was refusing to report. They did not care that I had a battalion evaluation coming up or a promotion board coming up or a fitness report to write. The load was the load. The tissue that was going to fail was going to fail. The morning I could not run was going to come whether or not I had done the psychological work of preparing for it.
That morning was Act I of what I now understand, in a way I could not have understood then, to be a whole life spent underneath the same pattern.
The migraines are Act II. Same pattern. Same disease of invisibility. Same Convincing Cost. Same Performance Tax. Same Cognitive Transference sitting at the wheel of my inner voice, telling me — during a bad week, during a bad month, during a bad year — that if I would only do a little more, be a little tougher, exercise a little more, sleep a little differently, be a little more disciplined, I could get on top of this thing.
The knees taught me a lesson that took eight years and four surgeries to learn.
You cannot outrun a body you refuse to listen to.
I did not listen to my knees. I paid for it. I want to be clear that I am not sorry I served — I loved that work, I loved the branch, I loved the men and women I served alongside, and I would do it again. But I did not have to lose the career the way I lost it. If someone had handed me, in 1999, a framework for reporting the honest load on a chronic soft-tissue injury, I might have had a longer career. I might have kept my knees. I might have deployed one more time.
Nobody handed me that framework.
Twenty-five years later, when the migraines came, I made myself a promise. I was not going to lose this career the way I lost the last one. I was not going to pay the Convincing Cost until my body did the reporting for me. I was going to build the honest reporter myself — an instrument that could see what I could not, that would not lie for anyone, that would remember what I could not remember, and that would speak, when the time came, in a language the world could actually hear.
That is what the rest of this book teaches you to build.
I want to say the last thing directly, because it is the whole book, and it is the reason I am writing it.
If you are reading this, you may be a person with migraine disease. Or you may be a person with fibromyalgia, or endometriosis, or ME/CFS, or Long COVID, or postural orthostatic tachycardia, or interstitial cystitis, or one of the hundred other conditions that share migraine's defining structural cruelty — the impossibility of being seen. You may be a person whose body is telling you something the world cannot verify. You may be a person who has been paying the Convincing Cost, in one form or another, for a decade or for a lifetime.
If you are, I want to tell you the same thing I told Christa in the front seat of that minivan, and I want to tell it to you with more kindness than I told it to her, because the pool argument was not going on in the back of your car, and you are not driving through the last light of a South Carolina evening trying to get five children home to their beds.
You are the one carrying it.
The people who love you may not be able to see it. That is not their failure. It is the disease's most efficient work. What we are going to do, over the next twelve chapters, is give you an instrument that does see it — an instrument that will keep its own record, that cannot be talked out of what it knows, that will remember on the days you cannot remember, and that will speak, when you take it into the doctor's office or the claims and ratings evaluator's file or the disability hearing room, in the currency of transparency the whole system is designed around: evidence.
Your body generates data all the time. It always has. It is generating data right now, while you read this.
Learning to recognize that data, learning to collect it faithfully, learning to interpret it with rigor, and learning to hand it to the people who need to see it — this is the precipice moment this book is designed to bring you to.
You are not powerless.
You are not required to look like the people around you who are not sick. You do not have to keep paying the Convincing Cost. There is another way to do this, and the rest of this book is the map.
 
Interlude
Kitchen Table
The Tuesday after Myrtle Beach, Christa was on the road with the fourteen-year-old. I was at home with the two youngest.
I remember that evening better than I remember most evenings, which is not because anything happened. Nothing happened. That is precisely the point of this small chapter, and it is why I have chosen to place it here between the emotional weight of the car ride home and the technical work I am about to ask of you in the chapters that follow. The book has been asking a lot of you. It has been asking a lot of me. I would like to give both of us — reader and writer — a Tuesday evening to breathe in.
I was at the kitchen table with M. and A. The eleven-year-old and the nine-year-old. I have been careful in this book about how I use my kids' names — some of them will read this someday, and their friends' parents will read this, and I do not know yet what shape the internet will make of any of it — so I will call them M. and A. for now. M. is my non-verbal eleven-year-old stepson. A. is my nine-year-old stepdaughter. Christa's from before. Ours now, in the ordinary meaning of that word — the blended household meaning, the got-married-in-August-of-2025 meaning, the one that eventually stops needing the modifier "step."
M. was working his AAC device. If you have not sat with an eleven-year-old boy using augmentative and alternative communication before, I want to tell you a small thing about it that is easy to miss. The device is not slower than speaking. It is different. Speech has a rhythm. AAC has a series of decisions. Each screen tap is a small verdict about which pre-programmed tile most closely matches what the child means, and between the taps there is a stillness that people who have never used AAC misread as the child having nothing to say. What is actually happening in the stillness is that the child is choosing between three tiles that all sort of mean it, none of which mean exactly what he means, and he is deciding which compromise is closest to the truth. It is a version of the Convincing Cost I described in the last chapter, run at a smaller scale, inside a smaller body, over a smaller stake.
I do not know that everyone at that table understood that. I understood it because I have spent the last several years learning to translate my own body into tiles the medical system can read. M. and I are, in that specific structural sense, working the same problem.
He wanted juice. He tapped through to it — three taps, one course correction, two more taps — and then handed me the device and looked at me. His look, when he wants something, is not the look of an eleven-year-old asking. It is the look of an eleven-year-old confirming. Confirming that I saw. Confirming that I got it. Confirming that the sentence he had just paid for in taps had, in fact, been received on the other end.
I got up. I got him the juice. Apple juice, cold, the kind of small transaction that is the fabric of parenting and that I have, in the two years the disease was taking me, sometimes had to sit down for.
A. was working on a math worksheet. She likes math. She is nine, and she likes math the way some nine-year-olds like horses, which is to say she likes it enough to bring it to the kitchen table even when nobody is asking her to. She had a pencil with a specific kind of eraser on the top that she has been particular about since kindergarten. She was humming. I do not remember what. Something small.
Let me tell you what I was, in that moment, holding.
I was holding the residue of an attack that had ended a day and a half earlier — postdromal fog, the specific after-effect that a bad migraine leaves in the body, the one that convinces everyone around you that you are fine now because the pain has ended and does not tell them, because you would sound crazy saying it, that your brain is still running on eight cylinders instead of twelve. I was holding the paperwork the FMLA people had emailed that morning, which was going to require another form. I was holding the memory of the car ride home from Myrtle Beach, which was still sitting somewhere in my chest even though Christa and I had already talked about it twice and were fine. I was holding, if I am honest, a small hard piece of fear that I do not usually let out on the page.
I do not know if I am going to be a good father to M. and A.
I say that carefully, because it is not a self-pitying sentence. It is a factual one. I did not get here through the door most stepparents come through. I came here through a door that I am not entirely inside of yet. I have my own two boys, adults now, who I raised in an earlier life and who are their own men. I know how to be their father. I have been rehearsing that role for thirty years. M. is not asking me to be their father. A. is not asking me to be their father. They are asking me to be their kitchen-table adult on a Tuesday evening when their mother is on the road with their sister and the man at the head of the table is a man who came into their story two years ago and who, on the wrong week, is not fully present because his brain is running on eight cylinders instead of twelve. I do not know how to do this yet. I am doing it. Those are not the same sentence.
I got A. a glass of water. She said thank you the way a nine-year-old says thank you when she is not looking up from her worksheet — with the pitch of gratitude but without the eye contact of it. M. tapped through to something and held the screen up. It said, in the flat computer voice of the app: I love you. He looked at me. He looked back at the screen. He looked back at me. He was not asking. He was reporting.
I do not know how long I sat there. Not long. Not longer than a healthy man would have sat there. Long enough that A. eventually looked up from her math and said "Roger?" in the specific tone a nine-year-old uses when the adult in the room has gone quiet for slightly longer than the adult in the room usually goes quiet, and she wanted to make sure I was still there.
I was still there.
I said "I love you too" — to M. first, then to A., who had not said anything but who was in the room and who was watching me. I said it in my own voice, without the tiles, and I meant it in a way that I have not always been able to mean it. Some evenings the words come out clean. Some evenings they come out through the fog. This one was clean. I want to give you that, because I have not given you very many clean evenings in this book, and I do not want you to think that a chronic illness memoir is only about the hard ones.
Christa came home an hour later with the fourteen-year-old. The fourteen-year-old had her book. She said hello in the specific way a fourteen-year-old says hello when she wants you to know she has arrived but does not want to be pulled out of whatever her book has given her — efficient and courteous and slightly medieval — and it made me smile. M. finished his juice. A. finished her math. Christa put her hand on the back of my neck when she walked by, the way she does when she can tell I have been holding something during the hour she was gone.
She did not ask what.
I did not tell her.
That is a small thing about our marriage that is difficult to write down without making it sound like more than it is, so let me say it plainly. Some things I hold, I hold. She lets me. She knows I will say them when I am ready or I will not say them, and either one is a thing she has decided she can live with, because she married a chronically ill man who does not always have the energy to translate his interior into a language she can read, and she has decided she does not need every translation. She has decided that the man being present at the kitchen table on a Tuesday evening — with M. and with A. and with a small hard piece of fear that he did not put out on the table — is worth marrying.
That, more than any framework in this book, is what I want you to know about how a chronically ill life gets lived. Some of it is data. Some of it is the taps on a device. Some of it is a hand on the back of a neck. All of it is the record.
The seven dimensions are the next chapter. This one was just a Tuesday.
 
Chapter Three
The Collapse Between Experience and Evidence
Here is a fact about medicine that took me longer to accept than it should have:
Nothing that happens to you gets to the doctor.
That is not a slight against doctors. It is not a slight against medicine. It is a description of the plumbing. The plumbing between what your body does in the middle of the night and what a clinician writes in your chart at 10:47 the following Tuesday morning is a very long piece of pipe, and every joint in that pipe leaks. By the time your experience reaches the record, most of your experience is gone.
Picture the pipe.
I want you to see it before we build anything, because everything we are going to build in the rest of this book is designed to survive it. If you understand where the pipe leaks, you can seal the joints. If you don't understand where the pipe leaks, you will spend a lifetime pouring your suffering into one end of it and wondering why the doctor on the other end keeps saying she cannot see anything wrong.
I am going to give this piece of plumbing a name. In the field of migraine — and, I am coming to believe, in the field of every invisible chronic disease — the pipe has a shape, and the shape has stages, and each stage strips signal from your story. I call it The Evidence Funnel.
You have been living inside The Evidence Funnel your whole medical life. You just haven't been given the map.
Let me put a real day inside it.
Say you had a bad attack on a Wednesday night. It started at 8:47 PM. You know the time because you were on the couch and your daughter was in her room doing homework and the television was showing something you were only half-watching. You noticed, at 8:47 PM, that the small print at the bottom of the screen — the ticker of scores from a sport you don't follow — had started to shimmer. That is your prodrome. Your body is telling you, in the only way it knows how, that a threshold is about to be crossed.
You did what most of us do. You did not write down the time. You did not photograph the ticker. You did not open a note on your phone and type shimmer, 8:47 PM, no known trigger, ate normally, slept badly last night, weather clear. You did not open a spreadsheet. You did not, because you were tired, and because Wednesday nights are for sitting on the couch, and because the whole reason we sit on the couch is that we are not, in that moment, supposed to be working against our own disease.
You noticed the shimmer. You closed your eyes. You told your daughter you were going to lie down for a little while. You went to your room. You did not turn on the light. You lay there for the next several hours as the attack came in — the aura, the pressure behind the right eye, the nausea that arrives about forty minutes into what your body already knew was coming. At some point in the night — you do not remember exactly when — you took the abortive. You do not remember whether you took one dose or two. You do not remember whether you drank enough water with it. You woke up at 3:14 AM with the pain in a different location than it had been when you went to bed, and the pain woke you up, and then you were awake for the next hour and a half in that particular gray country the middle of a bad night puts you in.
You slept, eventually. When you woke up in the morning, you had lost most of the specifics of what had happened between 8:47 PM and 3:14 AM. You retained the general shape — bad night, headache started around eight, took the meds, tossed all night — but the specific times were gone, the specific quality of the pain was gone, the specific effect of the abortive was gone, and even the specific room the attack had begun in was, by then, blurring gently around the edges.
This is stage one of the funnel: The Memory Layer.
Your own experience of your own attack is not fully available to you the next morning. It never is. This is not a personal failure. This is a specific feature of how migraine attacks and human memory interact. The migraine attack itself is doing damage to the very cognitive systems you would need in order to encode the attack accurately for later recall. You lose data at the source. Every migraineur does. Whatever you remember on Thursday morning about Wednesday night is a fraction of what actually happened, and the fraction that survived was chosen not by your reason but by whatever parts of your memory happened to be online during the specific hours the attack was underway.
By the time you sit down at your kitchen table on Thursday morning with your coffee, your Wednesday night has been compressed into perhaps three sentences and one adjective.
That compression is the first joint in the pipe. It leaks.
Now say you have an appointment with your neurologist on the following Tuesday. Not this Thursday, when the attack is still nearly fresh. Tuesday. Five days later. This is important because on the day you actually sit in front of the doctor, Wednesday night's attack is competing for shelf space in your head with Friday's mild attack, Saturday's poor sleep, Monday morning's stiff neck, and the low-grade dull pressure you had at your desk yesterday afternoon that you did not officially count as a migraine because it did not hit the threshold at which you feel entitled to use the word.
The neurologist has twelve minutes. She has, by the design of the American healthcare payment system, been given exactly twelve minutes to see you today, and inside those twelve minutes she has to review your chart, ask how you have been, listen to your answer, perform whatever short physical evaluation her clinical judgment requires, make a therapeutic recommendation, and document all of it in a note the insurance carrier will accept as adequate care.
She asks how you have been.
You are now standing at the second joint of the pipe. This one leaks worse than the first. Stage two of the funnel: The Translation Layer.
You have five days of imperfectly remembered experience to translate into perhaps ninety seconds of speech that will fit inside the shape of a clinical interview. You have to choose what to tell her. You have to choose in what order. You have to choose what to emphasize. You have to do all of this while managing the fact that you are, at this moment, also performing the Convincing Cost — you are trying to look like a sincere patient, not a difficult one, not a drug-seeking one, not one who exaggerates. You are aware, even if you have never said it out loud, that your credibility is being evaluated as you speak. So you edit. You leave out the moments that sound crazy. You leave out the moments you are not sure of. You leave out the pool argument in the back seat and the shimmering ticker at 8:47 PM because those details do not fit inside the shape of a clinical answer.
You give her the three sentences and the one adjective.
You add, if you are lucky and if you have prepared, a rough count: maybe four bad ones this month. She writes it down. The third joint of the pipe is the note. Stage three of the funnel: The Documentation Layer.
The three sentences and the one adjective and the rough count that you just gave the neurologist are now translated a second time — this time into the language of the electronic health record. The EHR does not care about your Wednesday night. The EHR cares about ICD-10 codes and CPT codes and about producing a note structured such that Medicare, Medicaid, and your commercial carrier will each be willing to pay the practice for the visit. What your neurologist actually writes in your chart, in the section of the note that will follow your body around the American healthcare system for the next thirty years, is something like:
Patient reports approximately 4 headache days in the past month. Denies changes in aura pattern. Continues on daily preventive with reported benefit. No new triggers identified.
That is the entire content, in the record, of the Wednesday night that woke you up at 3:14 AM.
Read the note again. Look at what has been lost.
The shimmer at 8:47 PM is gone. The time it started is gone. The location of the pain is gone. The failure of the abortive to hold you through the night is gone. The specific quality of the 3:14 AM wake-up — the fact that the pain was in a different place than it had been at bedtime, which is diagnostically important information about how your particular attacks behave — is gone. Your sleep quality across the week is gone. The dull pressure at your desk on Monday afternoon, which almost certainly was a low-grade attack even though you refused to call it one, is gone.
What is in the note is not a lie. It is not even inaccurate, by the standards of a documentation system that was never designed to hold what you have.
It is simply not your experience.
It is your experience after four rounds of compression, and each round of compression is happening in a plumbing joint that leaks.
The fourth joint is the specialist referral.
You get sent to a headache specialist. She reads the note before you walk in the door. What she reads is not your Wednesday night. What she reads is the eight-line summary the neurologist wrote after your twelve-minute visit last Tuesday. She sees approximately 4 headache days in the past month, continues on preventive with reported benefit. She forms an impression. Her impression is this patient is stable and responding to preventive therapy.
She was going to be your best chance.
She walks into the exam room with an expectation that has already been shaped by the compressed summary of a compressed conversation about a compressed memory of a night whose actual detail was already gone from you by Thursday morning.
You spend fifteen minutes of your appointment trying to un-compress her expectation.
You will not fully succeed. Nobody ever does. What she came into the room believing about you is stronger than what you can build in fifteen minutes on top of the note. This is not because she is a bad doctor. It is because she is a busy doctor with a hundred patients, and the way she stays afloat is by reading the notes and trusting them, and if the note said you were stable, then you are stable until proven otherwise, and the burden of proof is on you.
Stage four of the funnel: The Referral Layer.
Now you are three layers removed from Wednesday night, and the specialist is being asked to make a therapeutic decision about a person she has met once, based on a note written about a conversation that was itself a compressed translation of a memory that had already been decimated by the attack it was trying to describe.
Nothing about this is malicious. Everything about it is lossy.
The fifth joint is the insurance carrier.
Say you need a new therapy — a CGRP monoclonal antibody, or an intranasal abortive, or a class of injections that runs several thousand dollars a month. The carrier will require documentation of medical necessity. What the carrier's utilization-management reviewer will read is the same eight-line note. What the utilization-management reviewer will not read is your Wednesday night. What the utilization-management reviewer will use to decide whether you get the therapy is not your experience of the disease. It is the fifth-generation compressed derivative of a very small subset of your experience.
The carrier writes denied.
The letter arrives at your house. It says the criteria for approval were not met. It cites the same eight-line note, which was itself derived from the ninety-second conversation, which was itself derived from three sentences and one adjective, which were themselves derived from a memory that had already been eaten by the disease.
Stage five of the funnel: The Coverage Layer.
By this point, if you have not been living inside The Evidence Funnel long enough to recognize what is happening, it is very easy to conclude that the problem is you. That you are not communicating well. That you are not sick enough. That you should be doing more, taking better care of yourself, trying harder. This is where The Cognitive Transference we named in the last chapter completes its work. The system has just told you, in the coldly bureaucratic language of a denial letter, that what you said was not enough. You begin to wonder if what you said was true.
The sixth joint is the one I know best, and the one this book was in part written to address.
Stage six of the funnel: The Adjudication Layer.
If your disease progresses to the point that you can no longer sustain the work you do, and if you need to make a claim to a disability system — Social Security, the Department of Veterans Affairs, a private long-term disability policy, or an employer's short-term leave policy that ultimately becomes long-term — you will meet a new class of professionals.
At the VA, they are called claims and ratings evaluators. Elsewhere they go by other titles: disability examiners, medical reviewers, benefit specialists, adjudicators. The name changes. The function is the same. They read files. They render decisions. They do not meet you. They do not see the shimmering ticker at 8:47 PM. They do not sit next to you in the front seat of the minivan while your wife tells you she wonders if you're doing all you can. They see a file. And the file is the sixth-generation compressed derivative of your Wednesday night.
I have watched myself hand a file to a system like this. I have watched what that file contains after it has come out the far end of the funnel. And I have watched the decision that comes back.
The decision is not, at that point, being made about you. The decision is being made about a shadow of you — a shadow cast by a small and imperfectly-preserved artifact of the smallest, most cliché-worn piece of your actual life inside this disease.
If that shadow is thin, the decision goes against you.
If the shadow is thick, well-supported, and internally consistent — if it has survived the funnel with enough of your original signal intact — the decision may go in your favor.
This book is about how to keep the shadow thick.
That is the whole game.
Let me put it into a sentence I want you to be able to repeat to yourself the next time you get a denial letter, the next time a doctor writes something in your chart that doesn't match what you told her, the next time your spouse says she wonders if you're doing enough.
The system does not read what happened to you. The system reads what survived the pipe.
You cannot fix the pipe. The pipe was not built by you and it is not going to be rebuilt because you asked it to be. What you can do — what the rest of this book is about — is build a parallel pipe. A pipe you own. A pipe whose joints do not leak because you built the joints. A pipe that carries your Wednesday night from 8:47 PM to the desk of the claims and ratings evaluator without losing the shimmering ticker, the 3:14 AM wake-up, the failed abortive, the sleep deficit, the barometric pressure, the location of the pain, the color of the aura, and the ninety-second conversation you would have had with the doctor if the doctor had been given ninety minutes instead of ninety seconds.
The parallel pipe is not a diary. It is not a journal. It is not a headache tracker app. It is not a printout you hand the doctor. Those are useful tools, some of them, but none of them are what I mean. What I mean is an instrument that survives the funnel. An instrument that arrives at the sixth layer with the specific information the sixth layer needs to make the right decision — and that arrives already translated into the language of the sixth layer, so the sixth-layer reviewer does not have to do the translation herself and get it wrong.
That instrument does not exist in any drawer of the healthcare system.
I have looked. You have looked, or you would not be reading this. Nobody is going to build it for us.
That is what the next ten chapters are about. That is the whole reason I spent two years of a collapsing career building the frameworks I am about to hand you. The Seven Dimensions — a structured description of what an attack actually contains, so that no future funnel can strip the dimensions out. The Leaky Boat — a model of how attacks accumulate, so no future funnel can pretend your disease is a series of isolated events. The Threshold Model — a way of understanding trigger interactions, so no future funnel can dismiss you with the sentence no new triggers identified. The Clinician One-Page — a document that gives your neurologist twelve months of high-fidelity data in the ninety seconds she has to read it. The Disability Evidence Chain — a sequence of linked artifacts that arrives at the claims and ratings evaluator's desk pre-translated into the language she is required to use.
Each one is a joint you get to keep. Each one is a place the pipe will not leak.
By the end of this book you are not going to be a person who has better luck with doctors. You are going to be a person who has built a better pipe.
You are going to be, in short, an evidence-generating patient.
That is a category of patient the American healthcare system does not know how to refuse. I want to close this chapter with one more thing.
I have been an operations research analyst. I have been a data management director. I have been a Chief Data Officer of the United States Department of the Treasury. I have spent a full working life thinking about how signal moves through systems and where systems lose it. I built the frameworks in this book with the full weight of that career applied to a single question:
What is the minimum viable evidence architecture a chronically ill patient can build for themselves, alone, at their kitchen table, with tools they can afford, in a life whose energy budget is already taken by the disease?
The answer to that question turns out to be a very small number of things done very consistently.
You do not need a data science background to build it. You do not need a spreadsheet with a hundred columns. You do not need a wearable that costs six hundred dollars. You do not need a physician who is willing to spend forty-five minutes with you at every visit. You do not need any of that.
What you need is the discipline to keep the pipe from leaking on your end.
I am going to teach it to you, one joint at a time.
 
Interlude
Nineteenth
I need to take you back further than either nineteenth to make sense of them. I need to take you to a Tuesday in October of 2024 when I got up out of a bed I had been in for the entire day, in the middle of a crash migraine that had ended sometime that morning and had left me — the word fuzzy is too kind for what I was — worse than fuzzy. Emptied. The kind of postdromal fog where you are functionally awake but you cannot trust your judgment about anything, and you know it, and there is a specific loneliness that comes with being lucid enough to know you are not lucid enough.
I checked my email. I do not know why. Nervousness, probably. I had missed enough time from work by then that the arithmetic in the back of my head — the executive-grade arithmetic I had spent a career doing — was telling me the account was overdrawn in a way that was going to catch up with me any day. There was a note from my boss's secretary. She was inviting me to a meeting in the next day or so to discuss my status. She used the word status.
Later that week we met. I do not remember the room. I remember what she told me, and I remember it because it landed with the specific ordering of consequences that a career federal employee never wants to hear delivered in one sentence. I was out of leave. I was also out of leave donations — the small transfers from colleagues that keep an officer on the books during a long medical crisis. I would no longer be receiving a paycheck.
I want to tell you exactly what I felt sitting on the other side of that news, because it is one of the specific costs of an invisible chronic illness that gets talked about less than the physical ones and that hits, in some ways, harder. I felt humiliated. I felt scared. Those are two different things at the same time, and if you have not been through it you may not know that they are not mutually exclusive. I had spent my entire career being the person the numbers added up for. I had never been the person the numbers added up against. I did not know how to be that person yet.
We also discussed, that afternoon, that sooner or later I was going to have to make a decision about whether or not I was going to be returning to work. You have to understand what my role was, structurally, for this to land right. The Chief Data Officer position at the Department of the Treasury was established by the Foundations for Evidence-Based Policymaking Act of 2018 — the Evidence Act, most of us called it — and formalized through implementation guidance in 2019. It is a position created by Congress. It is the law. Every Cabinet department is required to have one. The CDO is not a favor granted by the Secretary. The CDO is a statutory obligation. When the work of the office of the CDO is not being performed — when the person in the chair is not there to sign the data governance memoranda, chair the Data Governance Board, deliver on the strategic plan the department has committed to Congress — Congress has to know about it. My absence, at that point, was not merely a personal medical crisis. It was a statutory gap. It was putting the Department at real risk in ways I understood better than most of the people talking to me about it, because I had built pieces of the machinery that measured that risk.
My boss told me, gently and clearly, that sooner or later they were going to have to issue a Notice of Proposed Removal. Not that I was being removed at that moment. That they were proposing to remove me because I could not show up. At the time I still had the ability to work remotely as liberally as I wanted. The disease had taken that away from me too.
Then the nineteenth of November showed up.
It was an email. My boss and HR. Notice of Proposed Removal. It stated the reasons. It stated my right to respond, and the timeframe for that response — thirty calendar days — and the process by which my response, if I chose to file one, would work its way through to a final decision. I had known the letter was coming for the better part of a month. I had, in the way of a person who has spent his career reading these kinds of documents, already rehearsed how I would read it. None of that mattered when it landed. My heart nearly beat out of my chest. For the first time in my entire working adult life, I was on the cusp of being fired. I want to say the sentences that ran through my head, because they are the sentences. A Marine. Fired from his post. A professional. Taken out of his job.
That was where I sat for the next two weeks. I want to be honest about what I did in those two weeks, because it is not what a Senior Executive is supposed to do when he receives a proposed removal. What he is supposed to do is respond to the letter. What I did was interview. I accepted interviews for other jobs. I was, in the middle of that stretch, actually offered a new position. I will tell you more about that later in the book, because it matters, and because the story of what happened with that offer is a story I want to save for its own place. What I will tell you now is that being honest with myself about what I could and could not do — the same honesty this book is going to ask you to apply to your own body — brought me, by the end of the two weeks, to a conclusion I did not want to reach. No amount of the Performance Tax I described three chapters ago was going to get me back into that job. I was no closer to returning to work in early December than I had been on the fifth of July when I first left.
I elected not to respond to the letter. I chose to leave it in God's hands.
That is not a rhetorical sentence. I want you to hear it as the plain fact of the theology I brought to the moment. I had, by then, done everything I knew how to do. I had run the arithmetic. I had rehearsed the counter-arguments. I had listed the interventions I could still try. None of them added up. When the arithmetic runs out and the account you are trying to defend is your own life, there is a particular kind of surrender that becomes available to a person of faith. It is not passive. It is not giving up. It is the specific decision to stop trying to steer a rowboat through a hurricane and to trust that the God who has kept you alive for forty-nine years is going to keep being the God who does that. That is what I chose.
Then, on a Tuesday evening in late November — after hours, after most of the building had gone home — I drove into the office one more time.
I want to be careful about what I say here, because it is the specific piece of the story that took me the longest to write down and that I want to give you cleanly. I did not wait for someone to ask me for the equipment. I knew the process. I had signed off on pieces of the process during my four years in the Department. I knew, from the moment the November nineteenth letter arrived and I knew I was not going to respond to it, that the equipment was going to be asked for. So I went in that evening and I turned it in myself. My badge. My laptop. My key fob. The two backup drives I had used for classified work in a life that no longer applied to me. I cleared my desk. I put what was mine into a cardboard box — the same kind of cardboard box you have seen in every layoff scene ever written, and yes, the mug from my older son and the challenge coin from the Marine Corps and the photograph of both boys were in it. I carried the box out on my own. I did not want anyone to have to ask me for the badge. I did not want anyone to have to walk me to a security desk. I wanted the last thing I did as the Chief Data Officer of the United States Department of the Treasury to be a thing I did on my own initiative, with my own hands, in the specific dignity that is available to a Marine who understands what is coming and refuses to be surprised by it. I cleared the desk in November. Nothing was going to prepare me for the months ahead. But I was going to give myself, on the way out, the smallest possible piece of self-directed dignity that the situation still permitted.
I prayed on the drive home that night. I asked God to show me the way.
The nineteenth of December was a Thursday. I woke into a bad one — a major attack that had run through the night and that was still, when the sky began to grey around six in the morning, refusing to break. I spent the first several hours of that morning trying to get it to calm down. You know the routine as well as I do if you have this disease. Dark room. Cool cloth. The abortive that has never quite been the abortive. The strategic small sips of water, timed to avoid the nausea. The specific chair, the specific angle of the neck. It broke, sort of, sometime after nine.
I got up. I made coffee. I want to say something about the coffee, because it is a small honest detail I would rather have on the page than off it. I do not like coffee. I have never liked coffee. I drink it, when I drink it, exclusively because the caffeine in it will sometimes buy me twenty minutes of tolerability against a migraine when nothing else will. That morning I needed the twenty minutes. I made the coffee. I hated it. I drank it anyway.
I did not know it at the time, but the coffee that morning was Part I of a bitter pill I was going to have to swallow that day. Part II was going to arrive in my inbox. I want to give you that in retrospect, not in foreshadowing, because in the moment it was just a cup of a drink I have never liked, and I did not have the benefit of knowing what was coming ninety minutes later. The body registered the bitterness first. The mind registered it second. There would be, before the day was over, a specific symmetry to that sequence.
The email came at some point after that. It was from the Department. The subject line was formal. The body was formal. The decision was formal. The charge — the exact phrase, which I have since read a great many times — was that I was being removed for medical inability to perform the duties of my position. The letter explained that there was no ascertainable date on which I could reasonably be expected to resume the performance of those duties.
Let me tell you how it had been signed, because the institutional plumbing of a federal removal is not obvious to most readers and it is the plumbing that made the day the specific day it was. The office with oversight of the Chief Data Officer function inside the Department was the Office of the Assistant Secretary for Management. A removal action of this kind was not a decision the Assistant Secretary made personally. It was a decision the Assistant Secretary's office assigned to an independent arbiter — a senior officer sitting elsewhere in the Department, in this case in Financial Transparency, whose institutional distance from the CDO function was the point. That officer was charged with evaluating the medical inability case on its merits, working alongside Human Resources, and rendering an independent decision on whether the proposed removal should be sustained. The design of the process was deliberate. Independence was the whole idea.
I do not want to say very much about the officer who signed the letter, and I want to explain why. The design of the process — the deliberate institutional distance, the independence, the fact that the person on the other end of the pen was chosen precisely because she was not in my chain of command — is exactly the design a chronically ill executive should want. The process worked as it was designed to work. A federal Department cannot run its statutory obligations on the goodwill of the people who happen to like each other. It has to run them on procedure. The procedure was fair. It was applied fairly. The letter that came out of it was procedurally correct. The letter was, in the narrow sense that a personnel letter can be, kind. It was also the letter that ended my career. All of those things were true at the same time. Holding them all at once is one of the specific tests that being a chronically ill executive requires you to pass, over and over, in the last months of a job that is being taken from you. You have to learn not to attach your grief to the person who is administering it, because the person is not the source of the grief. The disease is the source of the grief. The system is the vehicle. Everyone inside the vehicle is doing their job. I hope I passed that test. I did not always pass it in the moment. I have passed it, mostly, in retrospect. That is what I have to offer on the subject.
I sank in the chair.
I had grabbed the coffee. I hate coffee.
I called the labor relations person at Treasury who had sent the email, and I confirmed with her, in a voice that I do not remember but that I hope was steady, that being removed from the position also meant being terminated from the Department. She said yes. I thanked her. I hung up.
I texted Christa.
I said: It's done. I've been fired from Treasury.
I want to give you the truth of what happened in the minutes and hours after that text, because I owe it to you and because I owe it to myself, and because I have spent enough of this book putting the disciplined executive on the page and not enough of it putting the man on the page. For the first time in my adult life, I was unemployed. As much as I had known that letter was coming — for a month, for two months, for the whole trajectory of the year — it still hit like a freight train with no brakes ramming into a wall. I was undone. I was scared. I sat in the chair and I could not stand up out of it for a while. I stared at the ceiling. The ceiling did not have any answers. I stared at the phone. The phone did not have any answers either. I needed some kind of positive encouragement in that moment, and none was to be found. I felt as though my world had ended. I had no idea what was going to happen next.
I do not know how long I sat there. I know that Christa answered the text. I know that at some point that day I ate something, though I do not remember what. I know that at some point that day I went to bed. I know that I did not sleep well.
I want to tell you what I did the next morning, because it is the pivot on which everything else in this book turns.
The next morning I opened the migraine journal.
I did not decide to. I did not sit down and think about it and choose to. I opened it the way a person opens a bible when there is nothing else to open. I opened it because it was the only thing I had left in the house that had been built during the year I had lost my career. It was the only artifact of that year that had not been taken. The badge was gone. The laptop was gone. The office was gone. The Strategic Plan I had authored the last full section of was gone. The Data Governance Board I had chaired was gone. The Deputy who had held the line for me during the July-through-December stretch was going to be reassigned within a month. Everything was gone. The instrument was not gone. The instrument was on my kitchen table, in the form of a spreadsheet with eight thousand rows in it, and it did not know the career had ended, and it was not going to care when I told it.
I opened the file. I started scrolling. I started, as I like to say now, feverishly looking for answers.
I was in denial. I know I was in denial. I want to be honest about that. I was telling myself, in the specific rationalization that a chronic illness patient does when the disease has taken something enormous and he is trying to hold on to the possibility that he can still fix it, that if I could just maneuver this thing — the log, the tool, the eight thousand rows — into a shape that would let me identify the right treatments, I could get back into a role like the one I had lost within a year. I did not know, that morning, that I was not going to get back into that role. I did not know that the job offer I had already been given was going to fall through in ways I will tell you about in a later chapter. I did not know that Precipice Health did not yet exist, and that Migraine Data Detective did not yet exist, and that the tool I was scrolling through was in fact the beginning of both.
What I knew was that I was going to keep working on it. I was going to work on it hard. I was going to put December nineteenth into the rearview mirror by putting my hands on the one thing the disease had not been able to take from me.
The discoveries that led to the book you are reading came out of that morning. Not the emergency, not the letter, not the sinking in the chair. The morning after. When I opened the log. When I began, in the specific denial and the specific hope and the specific faith that a man can bring to a spreadsheet on the day after everything else is gone, to look for the answers I would eventually find.
The seven dimensions are the next chapter.
This one was the day the career ended, and the day after, when the work began.
 
Chapter Four
The Seven Dimensions
The first joint of the parallel pipe is a question.
It is not a spreadsheet. It is not an app. It is a question — and until you can answer it, no spreadsheet on earth will help you.
What actually happened in your attack?
I mean that literally. The next time you have a migraine, and the next time you have to describe it to somebody — your doctor, your spouse, a claims and ratings evaluator, yourself — what are the facts of the thing that happened? What was the shape of it? Not the sentence version. Not the four-word summary. Not bad one last night. The actual shape.
If you are like every migraine patient I have ever met, including me — including the version of me who was Chief Data Officer of the United States Department of the Treasury and who had spent his entire career turning fuzzy inputs into decision-quality intelligence — the honest answer is that you do not know. You know pieces of it. You know the general vibe of it. You know how bad it felt on some interior scale that you have never actually calibrated. But if I asked you to describe the last five attacks with any precision, you could not. Nobody can. This is not a moral failing. It is the direct product of a medical culture that has been asking one single question about migraine attacks for the last hundred years:
On a scale of one to ten, how bad is the pain?
I want to spend a little bit of time on that scale, because it is the reason we are here. The pain scale is not a bad instrument. It is a misapplied instrument.
The pain scale was designed for acute pain — the pain of surgery, the pain of a broken bone, the pain of the twenty-year-old in the emergency room whose kidney stone is passing right now and needs the anesthesiologist to know how much morphine to start with. In that context, the pain scale is a wonderful tool. It captures a single dimension of a single event that is going to be resolved within hours. It is fast, cheap, and clinically useful.
The pain scale was not designed for chronic disease.
It certainly was not designed for a disease whose defining feature is that pain is only one of the things that is wrong.
Ask a migraine patient — any migraine patient — what a bad attack contains, and if she is honest, and if you have given her time to think, she will tell you that pain is roughly the fifth thing on the list. Above pain, she will list nausea, or aura, or sensory hypersensitivity, or the specific cognitive fog that eats her ability to read a sentence twice and understand it either time. Beside pain, she will list the exhaustion that a real attack leaves behind, the postdromal fog that can last two days after the pain itself resolves, the dread of the next one, the effect on her sleep, and the interlocking web of medications she may or may not have taken correctly during the hours when the attack was actually happening and her judgment was compromised.
The one-to-ten pain scale asks her about none of this.
She circles a seven and hands the paper back.
The clinician sees a seven. The chart records a seven. The claims and ratings evaluator, three layers down the funnel, sees a seven. Nothing else about the attack is preserved.
You cannot build a medical case on a seven.
What I built, over the two years between January of 2023 and December of 2024, is the alternative.
I want to be careful about how I introduce it, because I do not want to hand you a framework the way a textbook hands you a framework. A framework is not a chapter heading. A framework is a tool you pick up. My hope is that by the end of this chapter, the Seven Dimensions will not be seven bullet points on your notes app. They will be seven things you actually reach for the next time your body starts to shimmer at 8:47 PM on a Wednesday.
Here they are.
The Pain Profile. The Medication Response. The Sleep Signal. The Autonomic Load. The Environmental Field. The Hormonal Context. The Functional Impact.
Seven dimensions. Every migraine attack you have ever had — good, bad, aborted, breakthrough, prodromal, postdromal — happened along all seven of them at the same time. Every attack you will have going forward will happen along the same seven. The medical culture that reduces your attack to a single number on the pain scale is capturing one-seventh of the shape of what happened to you. Six-sevenths is being discarded. Six-sevenths is the ocean of information your body has been generating for you your entire adult life, and nobody has been teaching you how to see it.
Let me walk you through them, one at a time. I want you to notice, as we go, that none of the seven is difficult. None of them requires a medical degree. None of them requires a device you do not already own. What they require is the discipline of a structured description, which is different from the discipline of remembering.
You cannot remember your attacks accurately. Nobody can. The disease itself will not let you.
You can describe them accurately, if you are given the frame. The Seven Dimensions is the frame.
DIMENSION 1 — THE PAIN PROFILE
Notice the word. It is not the pain. It is not the pain scale. It is the pain profile.
A profile has shape. A profile has parts.
For every attack, four small facts belong to the profile. Location. Where does the pain live in your head today? Behind the right eye? Along the temporal artery on the left side? A band across the forehead? The base of the skull where it meets the neck? The location is not decorative. The location is diagnostic. Migraines in different locations behave differently, respond to different medications, and correlate with different prodromes. Recording the location every time turns your log from a headache diary into an early neurological map of your own disease.
Quality. What kind of pain is it? Throbbing? Stabbing? Pressure? A dull ache with electric pulses inside it? Migraine pain has qualitatively different presentations, and the presentation itself is data. A stabbing attack behind the right eye is not the same as a throbbing pressure across both temples, even if they both hit a seven on the scale you were handed. Write down the quality in your own words. Three or four adjectives is enough. Later, when you have logged twenty or fifty or two hundred attacks, you will begin to see the patterns your body actually runs — not the average, but the shapes.
Intensity. Now you may use the scale. But use it here, as one of four parts of the profile, not as the whole. And use it honestly. A seven this week is not necessarily a seven next week — your calibration will drift as the disease progresses — but honest self-report inside a stable framework beats no self-report at all.
Trajectory. This is the one nobody asks about, and it is possibly the most important. Did the pain build slowly and peak at hour three? Did it start at peak and stay there? Did it come in waves? Did the intensity swap sides? Did the location move? A migraine attack has a shape through time, and that shape carries most of the useful signal about what is happening to you neurologically. A trajectory that moves — pain migrating from right occipital to left frontal over the course of five hours — is a different disease from a trajectory that stays put. Your neurologist would want to know that, if she had time to ask. She doesn't. You have to tell her.
Four small facts. Location, quality, intensity, trajectory. Under sixty seconds to record. Six-sevenths more information than the one-to-ten scale delivered.
DIMENSION 2 — THE MEDICATION RESPONSE
You took something. You almost always take something. What did you take, when did you take it, and what did it do?
I want to be honest here about what most patients record about their medications, because it is what I recorded for years myself, and it is worthless: took a Maxalt. Two words. No time. No dose. No effect. That entry, multiplied across ninety attacks over two years, is why your specialist could not tell you last April whether your triptan was actually working for you. Because took a Maxalt is not information. It is a shrug.
The medication response has three parts.
What. The full name and the full dose. Not the Maxalt — the 10 mg rizatriptan ODT. Not the nasal spray — the 8 mg zavegepant intranasal spray. This matters because you may be prescribed several formulations of the same molecule over the course of your care, and the formulation changes the pharmacokinetics enough to affect outcome. The nasal spray hits differently than the tablet. The dissolvable hits differently than the swallowed. Record the actual formulation.
When. Not sometime after it started. The clock time you took it, and — this is the one nobody records — how long after the onset of the attack. Was it fifteen minutes in? Ninety minutes in? Four hours in? Triptans in particular have a dose-response curve that is shockingly time-sensitive. A dose at fifteen minutes may abort. The same dose at ninety minutes may take the edge off but leave the tail. The same dose at four hours may do nothing at all. Your medication is not failing you. Your timing is failing you. But you cannot know that unless you are recording the timing.
Effect. What did it do? Not helped or didn't help. Something more specific. Pain reduction — from what to what, over what interval? Nausea reduction? Aura resolution? A partial abortion followed by a recurrence six hours later? A full abortion but with a two-day postdromal fog that cost you Saturday? These are all different responses, and they mean different things for what you should try next.
If you record these three things for every dose of every medication, in six months you will know more about your own medication response than your prescriber does. Not because she is bad at her job. Because she has never had access to this data. Nobody has ever collected it for a single patient across enough attacks to see the pattern. You will be the first.
DIMENSION 3 — THE SLEEP SIGNAL
Sleep is not a trigger. I want to say that clearly, because a great deal of migraine literature treats sleep as a trigger — one more item on the almanac list of aged cheese, red wine, poor sleep, stress. That framing is wrong. Sleep is not a trigger. Sleep is a substrate. Sleep is the thing your nervous system does at night to preserve the capacity you will draw against tomorrow, and when the substrate is bad, everything downstream of it — every threshold, every response, every recovery — moves closer to failure.
There are two things worth recording, and only two.
The night before. How much did you sleep? Not how much you were in bed. How much you actually slept, best guess. This is where a wearable earns its keep, if you have one — an inexpensive one is fine, an Apple Watch or a Whoop or a Garmin or an Oura ring will give you a rough total. If you don't have a wearable, guess. Ninety percent of what you need to know can be captured in around six hours, restless or close to eight, felt solid.
The quality. Restless, deep, fragmented, dream-heavy, or unusually shallow. One or two words. This is not a research paper. This is a signal you are preserving so that when you have logged sixty attacks, you can go back and see whether your bad attacks disproportionately follow certain kinds of nights. In many cases they do. In some cases they do not. The fact of knowing which — for your particular body — is more useful than a shelf of migraine textbooks written about the average patient's average sleep.
DIMENSION 4 — THE AUTONOMIC LOAD
This is the one I want you to spend the most time with.
The autonomic nervous system is the system that runs your body without your conscious involvement — heart rate, digestion, blood pressure, temperature regulation, sweat, pupillary response. It has two branches: sympathetic (the fight-or-flight branch that revs the system up) and parasympathetic (the rest-and-digest branch that cools it down). A well-regulated autonomic system moves gracefully between the two across the day. A poorly-regulated one gets stuck.
Migraine disease and autonomic dysregulation are entangled in ways the research community is only beginning to map. What I can tell you, from watching my own logs across two years, is this: the twenty-four to forty-eight hours before a bad attack almost always contain autonomic signals I can now see plainly on the page. Elevated resting heart rate. Depressed heart rate variability. A body temperature that is running half a degree warm. Sweat that shows up in the wrong places. A blood pressure reading that has drifted up. A resting respiratory rate that is a beat faster than my week's average.
None of this is subtle if you are watching. All of it is invisible if you are not.
What to record. If you have a wearable, log the resting heart rate and — if it offers one — the heart rate variability. Ten seconds of copying, morning of the attack. If you have a blood pressure cuff at home (they cost thirty dollars at a drugstore), take a reading during a bad attack and one on a good day, and note both. If you notice unusual sweating, temperature, or GI response — record it. Two adjectives. Hands unusually cold at onset. GI turned over at hour two.
The autonomic dimension is the one that will eventually let you predict attacks instead of merely describe them. That's a promise the rest of the book will help you cash. But you can't predict them if you never recorded them.
DIMENSION 5 — THE ENVIRONMENTAL FIELD
Your migraine did not happen in a vacuum. It happened inside a room, on a day, in a week, in a season, at a moment when your body was interacting with an environment that carried its own load. Weather. Light. Sound. Air quality. Barometric pressure. Ambient noise. Screen exposure. Travel. Air pressure changes from flying. Altitude changes from driving through the mountains. That entire field is data — some of it easy to capture, some of it requiring you to notice things you were not previously noticing.
The three fields that give the highest signal per unit of effort are:
Barometric pressure. Free. Available in any weather app. Record the reading at attack onset, and — if you can — the change over the preceding twenty-four hours. Barometric drops of more than four to six millibars in a rolling twenty-four-hour window are, for many migraineurs, a specific risk signature. You will not know if you are one of them until you have logged the pressures.
Light. Fluorescent, sunlight through a window, prolonged screen time, headlights on a night drive. One or two words. Been in fluorescent conference rooms most of the day. Long screen day. Bright sunlight without sunglasses on the walk home. These will not always correlate with an attack — but they will correlate more often than you expect once you have thirty attacks logged.
Sound and vibration. Not always. But sometimes. Loud restaurants, road trips, extended calls with a headset on, an unusually noisy office day. Record it if it stood out.
The environmental field is where most naïve headache diaries fail — they ask you to pick from a checklist of predefined triggers, which is exactly the wrong tool. The right tool is a small honest field where you describe, in your own words, what your environment was doing to you today. That field, across sixty attacks, will teach you your environmental sensitivities in a way no checklist ever could.
DIMENSION 6 — THE HORMONAL CONTEXT
For roughly two-thirds of the migraine population — the two-thirds who menstruate, or who are perimenopausal, or who are on hormonal contraception, or who are pregnant, or postpartum — the hormonal context is the single largest predictor of attack timing across the month. For the other third — cisgender men, primarily — the hormonal context is nearly silent as a variable but still worth noting when unusual life events (cortisol spikes, testosterone shifts, thyroid changes) are in play.
If you menstruate, the single most useful piece of information you can capture is day of cycle at attack onset. Just a number. Day 3. Day 12. Day 27. Nothing else. That number, logged for sixty attacks, will tell you within a year whether you have a menstrual migraine pattern, a mid-cycle pattern, an ovulatory pattern, or no cycle-linked pattern at all. Any one of those answers meaningfully changes what treatment approaches your prescriber should be considering.
If you do not menstruate, this dimension is often empty. That's fine. Leave it empty. An honestly empty field is more useful than a made-up field. When something does happen — a thyroid diagnosis, a hormone therapy start, a significant stressor known to shift your cortisol — log it and note the effect.
DIMENSION 7 — THE FUNCTIONAL IMPACT
The last dimension is the one the disability system will care about the most, and it is the one migraine patients record the worst.
The functional impact is not how bad it felt. The functional impact is what you could not do because of it.
I want to make that distinction absolutely clear, because it is the distinction on which every claims-and-ratings-evaluator decision I have ever seen has turned. The system does not compensate you for suffering. The system compensates you for lost function. If your attack ruined a weekend but you were still technically capable of doing your job on Monday, that's one number. If your attack meant you missed three consecutive days of a work week during which specific projects went undelivered, that's a completely different number. The system needs to know which.
Four small facts capture it.
Duration. How long, in hours, was the attack acutely disabling? Not how long you had a headache. How long you could not function at the level you would ordinarily function at.
Missed activities. Specific. Missed the morning meeting. Missed dinner with the family. Missed the client presentation and it had to be rescheduled. One or two lines. These are the artifacts a disability review actually uses.
Caregiver burden. Did somebody else have to step in? Did your spouse have to drive? Did you cancel the babysitter and reschedule the date? Did somebody at work cover a meeting for you? Every hour of caregiver burden is an hour the system wants documented, and no one is going to document it for you.
Recovery time. How long after the pain ended did it take for you to return to baseline function? A two-hour recovery is a different disability than a two-day recovery, and they should not be recorded as the same event.
READING THEM TOGETHER
Seven dimensions. Every attack. In three to five minutes.
I want you to see what a properly-recorded attack looks like on the page. Here is an entry from my own log, from June of 2024 — three weeks before I went out on leave. I have edited it lightly for legibility, but the content is unchanged.
Attack. Started ~14:20, Thursday, 6 June 2024.
Pain profile: Right retro-orbital → migrated to right temporal by hour 2. Throbbing at onset, pressure-and-stabbing by hour 3. Intensity 6 → 8 → 6 → 4 across roughly six hours. Trajectory: single peak at hour 3, slow decay.
Medications: 10 mg rizatriptan ODT at 14:47, 27 min after onset. Partial abort — pain dropped 8 → 4 within 90 min, held for ~4 hours, mild recurrence overnight. No second dose.
Sleep: 5h 40m the night before, restless. Second night in a row under six.
Autonomic: Resting HR 74 (7-day average 62). HRV depressed morning of. Hands cold at onset. Mild nausea sustained through peak.
Environmental: Barometric pressure dropped 7 mb overnight. Long screen day. Two hours in a fluorescent conference room mid-morning.
Hormonal: N/A.
Functional: Left the office at 15:00. Missed a 15:30 briefing (delegated). Missed dinner. Christa handled evening carpool. Recovery to baseline: next morning, ~09:00. Total disabling duration: approximately 19 hours.
Read that entry once. Then read the eight-line note your neurologist would have written from the same event: Patient reports one moderate-severity attack this week, partial response to rizatriptan, no new triggers.
Now imagine handing the first entry — sixty of them, across a year — to a claims and ratings evaluator.
That is what the parallel pipe carries. That is joint one of six. I want to close this chapter the way I want you to close it in your own body.
You are going to feel, when you first start doing this, that it is too much. That you don't have the energy. That the record itself is another thing the disease has taken from you. I am not going to pretend otherwise. The first month is the hardest month. It gets easier once the template lives in your muscle memory — once you have logged twenty attacks and the seven dimensions come to you as fast as brushing your teeth.
Do not try to be complete. Do not try to be perfect. Do not skip an attack because you can't do all seven dimensions right now. Log what you can. Two dimensions honestly logged beats seven dimensions half-remembered a week later. The instrument gets better as you get better. Nothing here has to be built at full resolution on day one.
The record does not have to be beautiful. It only has to be true.
The truth, over enough time, becomes the shape. The shape, over enough time, becomes the shadow.
The shadow, if you have kept the joints of your parallel pipe tight, becomes something even the claims and ratings evaluator — who has never met you, who reads a file, who has been given ninety seconds to make a decision about your life — cannot ignore.
§
A NOTE FOR NON-MIGRAINEURS
I have been writing this chapter in the language of migraine, because migraine is the disease I know from the inside. I want to pause here — briefly — to tell you what the Seven Dimensions look like if the disease you are reading this book about is something else.
If you have myalgic encephalomyelitis or Long COVID, the seven dimensions still exist. Your Pain Profile is not always a headache — it is the specific shape of your fatigue, your post-exertional malaise, the ache in your limbs on a bad day. Your Medication Response is still the four-part record I described. Your Sleep Signal is even more diagnostic in ME/CFS than in migraine. Your Autonomic Load is often the single most important variable in your case. Your Environmental Field is broader, and includes the specific physical and cognitive exertion of the previous few days. The Hormonal Context matters the same or more. Your Functional Impact is the dimension your disability adjudicator will read the closest.
If you have fibromyalgia, the Pain Profile is the map of your tender points across time. If you have endometriosis, the dimension I called Hormonal Context becomes the spine of the whole log. If you have POTS, the Autonomic Load is the entire chapter. If you have interstitial cystitis, the Environmental Field expands to include foods and beverages that would not matter to a migraine patient.
The Seven Dimensions are not seven dimensions of migraine. They are seven dimensions of the honest description of a chronic-illness event. The event varies. The architecture does not. Translate the vocabulary, keep the framework. You are not the wrong reader for this book. You are one of the readers this book was quietly written for the whole time.
The next chapter zooms out. Attacks do not happen one at a time — they accumulate against a body whose capacity is itself moving. That is the Leaky Boat, and it is where a log of individual attacks becomes a map of a chronic disease.
 
Chapter Five
The Leaky Boat
There is a moment on the water I think about all the time.
I was in my mid-twenties, on a small boat with people I loved, out on a river in South Carolina. It was late afternoon in July — the kind of afternoon that stays with you because the air is that specific, still, weighty green that only shows up in the South in mid-summer when the sun has been on the water for eight straight hours. We were coming back to the launch. The river was low. Somebody at the wheel — it may have been me, it may have been someone else, memory has been kind about that — took a line that was six or eight feet outside of where the channel actually was, and the hull came up on a gravel bar with the specific soft crunch that is the sound of a bad afternoon starting.
We did not know, standing on the boat, that we had punched a small hole below the waterline.
Nobody heard it. Nobody saw it. The boat looked fine. We backed off the bar, restarted the engine, and continued down the river. For twenty minutes we told ourselves the story that nothing important had happened. And for twenty minutes the boat, without telling us anything, was quietly taking on water.
I want you to sit with what happened next, because it is exactly what happens to a chronically ill body, and the way it happened on the water is the way it happens in your life.
We noticed the boat felt heavy first. Then we noticed the bilge was full — much fuller than it should have been. Then somebody checked the transom and found the wet stain on the deck that meant the water was over the floorboards. By the time we understood we were in trouble, we had a lot of water in the boat and a much shorter distance to shore than the situation required. We made it back. It was fine. Nothing bad happened.
But something happened to me on that boat that has stayed with me for twenty-five years.
We had not been caught by a storm. We had been undone by accumulation.
There was no wave. There was no lightning strike. There was no clean, dramatic, story-worthy event that anyone could point to and say there — that is what caused the emergency. What caused the emergency was a small hole that we had put in the boat twenty-two minutes earlier and then chosen to ignore, plus the ordinary fact that water is going to enter any hole that is under a waterline, plus the ordinary passage of time, which is the substance in which all accumulation happens.
The gravel-bar hole did not sink us. The gravel-bar hole plus twenty minutes of unnoticed inflow nearly sank us.
That is the model I want to give you for migraine disease.
That is the reason your neurologist's checklist of triggers has been failing you for as long as you have been reading it.
The story medicine has been telling you about your attacks is the wrong story.
The story medicine tells you is the trigger story. It goes like this: you have an attack because something set it off. A glass of red wine. A late night. A skipped meal. A stressful meeting. A change in the weather. Whatever the item was — you look for it, you find it, you cross it off next time, and you have fewer attacks.
If that model worked for you, you would not be reading this book.
It does not work because it is trying to describe accumulation with the vocabulary of an event. A trigger is not what caused the attack. A trigger is what happened to be sitting closest to the waterline when the last inch of accumulated capacity ran out. The wine did not cause the migraine. The wine was the last drop of water that entered a boat that had been taking on water for the entire preceding week, and there is no version of your health in which crossing red wine off a list is going to fix what actually went wrong.
The problem is not the drop. The problem is the accumulation.
The problem is the leaky boat. I want to describe the boat.
The boat is you. Not you as an idea. You as a physiological system with a finite capacity to absorb load. Every day of your life, your nervous system operates inside a budget. On some days the budget is generous — you slept eight hours, you ate well, you exercised gently, you had a good conversation with somebody who loves you, your hormones were quiet, the weather was stable, you were not asking your body to translate an invisible disease into somebody else's comfort for ninety of the previous ninety-six hours. On those days you have a lot of budget. You can absorb load. You can drink a glass of wine at dinner, sit through a stressful meeting, sleep six hours, and wake up fine.
On other days the budget is very small.
The budget is small when you have already been running against a deficit. Bad sleep two nights ago. A stressful week. A cold that took a lot out of you. A cross-country flight. A fight with someone you love. A month in the wrong part of the hormonal cycle. Prolonged screen time. Prolonged fluorescent light. Prolonged noise. Prolonged everything. On a day when the budget is small, the same glass of wine that did nothing to you last month, at dinner, on a Saturday, becomes the last inch of inflow into a boat that was already sitting low in the water. You have the attack. You blame the wine. You cross wine off your list.
And you did not learn anything.
The wine has not been the problem in most of the cases where the wine was the visible thing at the moment of attack. The problem was the capacity. The problem was that you were, without knowing it, already sinking.
The boat has three components, and if you can hold all three of them in your head at once, you will understand your disease better than most of the specialists you have ever seen.
Component one is the water level. The water level is your accumulated physiological load — every deficit you are currently carrying, expressed as inches above the floorboards. Sleep deficit is inches. Chronic dehydration is inches. Autonomic dysregulation is inches. Ongoing hormonal shifts are inches. Untreated anxiety is inches. Prolonged inflammation from anything else your body is doing is inches. All of these are cumulative and all of them are invisible to you unless you are recording them. They accumulate silently.
Component two is the holes. The holes are not triggers. The holes are the specific things that make water come in faster. A short night. A skipped meal. A weather system that has dropped barometric pressure by nine millibars overnight. A long screen day. An emotionally hard conversation. Each hole is a rate — how fast water enters the boat per unit of time. You have some holes you were born with and some holes you have opened over years. You cannot patch all of your holes. You can choose, on any given day, not to sail across the gravel bar that will open a new one.
Component three is your bail capacity. The bail capacity is your body's mechanisms for removing water. Sleep. Real hydration. The rest that comes after a good meal. Time in silence. Time in prayer. Time outdoors. Time with people who are not asking you to convince them of anything. Restorative movement — the walk, the stretch, the yoga session, the swim in a quiet pool. Medications that support the underlying nervous system — the preventatives, the magnesium, the beta-blocker, the CGRP inhibitor. Each of these is a rate at which you can take water out of the boat. If your bail capacity exceeds your inflow rate, the boat is stable. If your inflow rate exceeds your bail capacity for long enough, you cross the threshold. You have an attack. The boat swamps.
I want to give you the sentence that took me two years to earn, and that will change how you think about your disease forever if you take it seriously.
Migraine is not an event. Migraine is a state change.
An event has a cause. A state change has a history. When you have a migraine attack, what has happened is not that something caused the attack in the last fifteen minutes. What has happened is that your system, having accumulated enough load and having lost enough bail capacity across a window of hours or days that you had not been tracking, has crossed the specific threshold at which it can no longer maintain the state it was in and must move into a new state — the state we call an attack. The trigger you noticed at the moment of onset was not the cause. The trigger was the last inch of water. The cause was every hour of the preceding week.
That is the whole framework in one thought.
The trigger you noticed at the moment of onset was not the cause. The trigger was the last inch of water. The cause was every hour of the preceding week. Once you have the boat in your head, the practical implications begin to reorganize your life.
The most important implication is that the intervention window is not the fifteen minutes before the attack. The intervention window is the seventy-two hours before the attack. That is when your capacity was bleeding. That is when the boat was going down. That is when it was still recoverable. By the time the aura arrives, you have entered the state change, and while your abortive medications may reduce the duration or the intensity, the attack itself is already in progress. If you want to reduce how many attacks you have — not just how long they last — you have to intervene in the accumulation phase.
Which means you have to see the accumulation phase.
Which means you have to be recording the seven dimensions from the last chapter, every day, even on days when you feel fine.
I want to say that as clearly as I can, because it is where most migraine patients stop reading. Recording data on days you feel fine is boring. Recording data on days you feel fine feels pointless. Recording data on days you feel fine feels like more work being demanded of a person whose disease has already demanded too much. I understand. I resisted it for the first six months of my own build.
Here is why you do it anyway.
The record of your good days is what tells you what a good day looks like for you. It is your baseline. Without it, you cannot detect a bad-day-in-progress until the bad day has already arrived. With it, you can look at Tuesday morning's heart rate variability and say my HRV has been depressed for three days now, my resting heart rate is up four beats over baseline, I slept under six hours two of the last three nights, and the barometric pressure is dropping. I am not currently in an attack. But my boat is taking on water and my bail capacity is low. This is not the day to drink a glass of wine at dinner. This is the day to eat early, hydrate seriously, cancel the after-dinner call, and sleep by ten.
That paragraph is what health looks like, at the operational level, for a person with migraine disease.
That paragraph is not achievable without the data. The data is only achievable if you are recording it before you need it. I want to give you one more piece of the boat, because it will save you from a specific kind of despair. You are going to fail at this.
I mean you specifically. You are going to fail at recording. You are going to have weeks where you do not open the log. You are going to have attacks that go uncaptured. You are going to have three-day stretches where you meant to write things down and did not. When that happens — and it will, and it did to me, dozens of times, across the two years I was building this — you must not throw the whole system away.
The boat metaphor helps here too. You do not need every drop of water to be accounted for. You need the shape to be visible. Sixty attacks recorded across a year, even if seven attacks that year went completely unlogged, is a shape that gives you and your clinician and your claims-and-ratings-evaluator a real picture of your disease. Sixty attacks is a data set. Sixty attacks is a story. Sixty attacks is a shadow the system cannot ignore.
Perfect logging is not the goal. Consistent-enough logging is the goal.
The record does not have to be beautiful. It only has to be true, and it only has to be true across enough of the year that the true shape shows through.
Now let me tie the boat back to the seven dimensions, because they are not separate frameworks. They are the same framework, seen from two different distances.
The Seven Dimensions describe a single attack. They are the anatomy of one event.
The Leaky Boat describes what has been happening to you across the days and weeks between attacks. It is the physiology of your capacity.
If you record the Seven Dimensions for every attack, and if you record two or three key indicators every day even when you are not having an attack — a resting heart rate, an approximate sleep duration, a note about environmental load, a subjective one-word capacity rating — you will have, within six months, everything you need to see both the anatomy and the physiology at the same time. You will be able to look at any attack in your log and see the boat that was already going down before it happened.
That is not a small thing. That is what medicine, for a hundred years, has been unable to give a single migraine patient at their kitchen table.
You will be giving it to yourself.
I want to close this chapter the way I want you to close the day you first understand it.
The next time you have an attack, do not ask what triggered this. You will have been trained by every neurologist you have ever seen and every migraine article you have ever read to ask that question, and the question does not open the door you need opened. The question opens a door that leads to a checklist and a shrug.
Ask instead: what did the last seventy-two hours look like?
Ask what your sleep was doing on Sunday night. Ask what your heart rate was doing on Monday morning. Ask what the weather did on Tuesday. Ask whether you sat under fluorescent lights all afternoon. Ask whether you spent Wednesday emotionally translating your invisible disease into somebody else's comfort. Ask whether there is a pattern of accumulation across those three days that, in retrospect, was going to end with an attack no matter what you had done in the last fifteen minutes.
The answer will not always be clean. Some attacks come in without visible antecedents, because your body is a complicated system and no framework will explain every event. But when you ask the question about the seventy-two hours, and when you ask it with the seven dimensions of your last several attacks in front of you, and when you ask it with a running record of your daily capacity indicators next to you, you will begin — for the first time in your medical life — to see the water rising.
Seeing the water rising is what health looks like for a migraineur.
Not curing the disease. There is no cure. Seeing the water rising.
Once you can see it, you can bail. Once you can bail, you can preserve capacity. Once you can preserve capacity, you can go weeks and sometimes months in a range of stability the checklist model would have told you was impossible.
That range of stability is what the rest of this book is going to help you build.
The boat can be captained. §
A NOTE FOR NON-MIGRAINEURS
The boat is not a migraine metaphor. The boat is an invisible-illness metaphor. I built it out of my own disease because that is the water I sail in, but it is the same water any of us sails in when we live with a condition that accumulates.
If you have myalgic encephalomyelitis, post-exertional malaise is the specific accumulation your boat runs on. Every exertion above your energy envelope opens a hole. Every day of rest below it is the bail rate. The crash is the threshold crossing. The Leaky Boat is not a bad description of ME/CFS. It may be the best description of ME/CFS you have ever read.
If you have fibromyalgia, the accumulation is inflammatory, cortisol-linked, and sensitized. The boat still fills. The waterline still moves. A flare is still a threshold event, not an isolated cause.
If you have endometriosis, the cycle itself is the tide. Some days your baseline waterline is higher than others because of where you are in the month, and the same input that would not have swamped you on day nine will swamp you on day twenty-six.
If you have POTS or dysautonomia, the boat's ballast is heart-rate variability and orthostatic tolerance. When they drop, the boat sits lower. The trigger that would have been nothing yesterday floors you today.
You do not need to be a migraineur to sail this boat. You need only to be a chronically ill person whose disease accumulates. If that is you, the Leaky Boat is your model too. Bail accordingly.
The next chapter is how — because triggers do not act alone. They combine. And what they combine against is a capacity that is itself moving. That is the Threshold Model, and it is joint three of the parallel pipe.
 
Chapter Six
The Threshold Model
I want to take you into a Wednesday evening in September of 2024. This one is going to be brief, because I have already asked you to sit through a lot of scenes in this book, and I would like to spend the bulk of this chapter on the framework itself. But the scene matters, because it is where I first understood — with the kind of understanding that changes the way you use a tool — that triggers do not act alone. Triggers combine. And when triggers combine, they do not do so by adding themselves up the way the checklist model wants them to. They do so by multiplying against a capacity that is itself moving underneath them. That is the Threshold Model. It is joint three of the parallel pipe. And once you can see it, the whole map of your disease changes shape.
I was sitting at our kitchen table. Christa was somewhere in the back of the house with the eleven-year-old. I had come out of a two-day attack — a real one, the kind that leaves you thin and unfocused and slightly untrustworthy in your own head for a while after — and I had the log open on the laptop in front of me. The reason I had the log open is that the attack had violated the model. I want to be honest about what I mean by that. By September of 2024 I had been logging the seven dimensions for a little over eighteen months, and I had reached the point in the discipline where my log was actually predicting attacks. Not most of them. But some of them. There were mornings when I could look at the previous forty-eight hours on the page — sleep, resting heart rate, HRV, barometric pressure, screen exposure, fluid intake — and think this is a day I need to protect. And on many of those mornings I could preserve the day. Cancel the after-dinner meeting. Move the family dinner up an hour. Take the walk. Drink the water. Sleep before ten. The attack that had been building would sometimes, not always, dissolve. I had begun to believe I understood my system.
The Wednesday attack should not have happened. That is what the log said. I want you to hear the language of a former Chief Data Officer of the United States Department of the Treasury talking to himself at his kitchen table, because it may be language you know. Sleep the night before the attack was seven hours and twelve minutes, which was above my rolling baseline. Resting heart rate that morning was 60, which was below my baseline. Barometric pressure had been stable for forty-eight hours. I had not been under fluorescent lights the day before. I had not skipped meals. I had not been dehydrated. I had had one glass of wine on Tuesday evening, at dinner, which I have logged as a probable driver in the past but only in a specific hormonal context that did not apply to a forty-nine-year-old man in September. On every dimension I could see, the Wednesday attack was inexplicable. I had gone in expecting nothing and been knocked down for two days.
I sat with that for a while. I want to tell you what I did next, because it is the whole discipline of the framework I want to give you in this chapter. I did not throw the log away. I did not blame the wine. I did not conclude that the framework had failed. I did not fall into the very old temptation that has caught me many times — the temptation to say "well, migraine is just random, my body is a mystery, there is nothing to be done." What I did was open a wider window. Instead of looking at the twenty-four hours before the attack, I looked at seventy-two. Instead of looking at seventy-two, I looked at a hundred and twenty. I widened the aperture until the attack stopped being an unexplained event and started being a pattern.
Here is what I found. On Sunday — three days before the attack — I had slept only five hours and forty minutes. Not because of a bad night. Because we had had friends over late, and we had laughed for the last hour of the visit, and I had gone to bed after midnight for the first time in a month. Monday, I had felt fine. Sleep on Monday night had been solid — nearly eight hours. Tuesday I had been productive. I had eaten well. I had drunk enough water. I had had the one glass of wine at dinner. Tuesday night I had slept just over seven hours. Wednesday I had woken up with a resting heart rate of 60 and an HRV that looked, to my eye, unremarkable. And on Wednesday afternoon, at 3:14 PM, I had crashed.
The Wednesday reading told me nothing. The Sunday-through-Wednesday reading told me everything. There had been a small, isolated hole opened on Sunday — five hours of sleep — and I had not bailed it out. Monday's sleep had refilled some of the deficit, but not all of it. The wine on Tuesday, which would have done nothing on a Tuesday following two nights of solid sleep, had contributed a small additional inflow into a boat that had been quietly sitting one inch lower than baseline for three days. And on Wednesday afternoon, when a specific atmospheric event had happened — a barometric drop of five millibars in the middle of the afternoon that had not shown up on my morning read because it hadn't happened yet — the last inch of capacity had gone. The attack was not inexplicable. It was the exact outcome the model would have predicted, if the model had been drawn wide enough.
What that Wednesday taught me is the piece that took me the longest to learn, and it is the piece I most want to save you the pain of learning the hard way.
Triggers do not act alone. They act in combination. And they combine against a capacity that is itself moving.
If you can hold that in your head, you have joint three of the parallel pipe.
§
Let me unpack it in the plainest way I know. In the checklist model — the model your neurologist has probably been taught to hand you — a trigger is an event that has an intrinsic probability of causing an attack. You have some triggers with high probability (say, red wine). You have some with medium probability (say, dehydration). You have some with low probability (say, screen fatigue). You track them. You avoid the high-probability ones. You get on with your life. In this model, every trigger has a fixed strength, and every attack is caused by whichever trigger happened to be present when the attack occurred.
The model is wrong. It has been wrong for a hundred years. It is wrong because it assumes independence — that each trigger acts by itself, without knowledge of what any other trigger did to your system yesterday or the day before. Independence is the assumption that lets you use a checklist. Independence is the assumption that lets a specialist say "avoid your triggers." Independence is what makes migraine seem, to the checklist-driven patient, like an untreatable random event that keeps happening even when they do everything right.
Independence is not how the human nervous system works. The human nervous system is, in engineering terms, a nonlinear coupled system. That is a mouthful, so let me translate. It means that the effect of doing two things at once is not the sum of doing each one separately. It means that a body under load handles inputs completely differently from a body at rest. It means that the same glass of wine, at the same time of night, on the same day of the week, is a different physiological event when your capacity is at ninety percent than when your capacity is at forty percent. It means, in plain English, that whether something will trigger a migraine for you today depends not on the thing itself but on the state you are already in.
There is a piece of engineering language that captures this exactly, and I am going to introduce it to you not because I want to make you an engineer but because it will change the way you see your own attacks forever. The concept is called a threshold. A threshold, in a physical system, is the level of input at which a fundamentally different behavior begins. Below the threshold, the system behaves one way — usually smoothly, predictably, absorbing what you throw at it. Above the threshold, the system behaves a completely different way — it collapses, it changes state, it moves from one mode of operation to another.
The classic example every engineer learns is a beam under load. A steel beam will bend, slightly, in a very predictable way, up to a specific weight. Below that weight, no matter how many times you load it, the beam returns to its original shape. Above that weight — and this is the crucial part — the beam does not bend a little more. It buckles. It fails. It changes state. There is no fifteen-percent-more-bent version of the beam that got loaded fifteen percent past the threshold. There is either an intact beam or a broken one.
Your migraine works the same way. Below the threshold, your nervous system absorbs load. It bends. It adjusts. Sleep debt, dehydration, stress, weather changes, hormonal shifts, sensory input — all of these accumulate, and your body's regulatory systems compensate. You feel a little tired. A little foggy. A little short. You would not describe yourself as sick. You are inside the linear regime of the system. Above the threshold, your nervous system does not become one percent more tired. It changes state. The trigeminal nucleus depolarizes. The cortical spreading depression begins. The hypersensitivity to light and sound and motion and smell arrives all at once. You are no longer in a body that is a little tired. You are in a body that is having a migraine. There is no gentle transition. There is a threshold, and either you are below it or you have crossed it.
This is the most important thing I am going to tell you in this book, and I want you to hear it exactly. Migraine is a threshold disease. Every attack you have ever had happened at the exact moment the accumulated load on your nervous system exceeded a specific ceiling above which your neurology cannot maintain its ordinary state. Below that ceiling, you feel fine, or a little off, or tired, or foggy. Above that ceiling, you have an attack. The ceiling is not fixed. The ceiling moves with your general health, with your hormonal cycle, with your sleep bank, with your stress bank, with your inflammatory load, with everything the last chapter called your capacity. And the inputs you throw at the system are not scored individually. They are summed against the ceiling.
That is why the checklist fails. The checklist thinks every attack has a cause. The Threshold Model says every attack has a total. And the total is what matters.
§
Let me give you the second piece of engineering language, because it will help you see the practical implications. In a coupled nonlinear system, inputs interact. That is a formal way of saying that the effect of two triggers together is often larger than the sum of the two triggers apart. Sometimes much larger. In the Wednesday attack I described a moment ago, the total input from four small things — three days of mild sleep deficit, one glass of wine, an afternoon barometric drop, and a normal Wednesday of work — added up to an attack, even though no one of them, in isolation, would have caused one. That is interaction. Interaction is what the checklist model cannot see. Interaction is what your neurologist has never been trained to draw. Interaction is what makes the same person a completely different disease on Tuesday and on Friday.
I want to give you the practical shape of it, because I do not want to leave you with a piece of engineering vocabulary and a war story from my kitchen table. I want to leave you with something you can pick up. There are three principles of the Threshold Model that will change how you read your own data forever. Learn them once, and you will never look at a trigger list the same way again.
PRINCIPLE ONE — THE SAME INPUT IS NOT THE SAME EVENT
The single glass of red wine you had on Tuesday of week one and the single glass of red wine you had on Tuesday of week two are not the same physiological event. They are the same drink. They are not the same event. Whether the drink triggers a migraine depends on what state your capacity was in when the drink arrived. On a week when you slept nine hours a night, drank three liters of water a day, exercised on Monday, and had no environmental stress, the wine is nothing. It goes in, it gets metabolized, and by the time you are getting ready for bed you have forgotten it. On a week when you slept six hours a night, worked twelve-hour days, sat under fluorescent lights all day Monday and Tuesday, and had a mildly stressful conversation with a family member on Sunday evening — the wine may be the last drop. Same drink. Same dose. Same time of night. Different disease.
The implication of Principle One is that you cannot maintain a static trigger list. You have to maintain a dynamic capacity picture. A "yes" or "no" verdict on any given input is not something you can attach to the input itself. It is something you have to attach to the combination of the input and the state you were in when the input arrived. This is why, in the log I taught you to build in Chapter Four, you are recording both the attack itself (the seven dimensions) and the day-by-day capacity indicators between attacks. Because without the capacity indicators, you cannot interpret the trigger. The wine, by itself, is uninterpretable. The wine plus five hours of sleep on Sunday is interpretable. Without the second piece, you are guessing.
PRINCIPLE TWO — INTERACTIONS ARE OFTEN MULTIPLICATIVE, NOT ADDITIVE
This is the one sit with the longest, because it is where the intuition breaks. In the checklist model, if red wine has a probability of ten percent of triggering an attack and skipped meals have a probability of twelve percent, then having wine plus skipping a meal gives you something like twenty-two percent — a sum. That is the mental math the model implicitly does, and it is why patients keep asking "which one caused it?" as if a single trigger could be identified in retrospect. In the Threshold Model, the interaction is more like a product. Red wine with a full stomach and a well-slept body may be a five-percent event. Skipped meals with a well-slept body may be a five-percent event. But wine plus skipped meals plus insufficient sleep may be a sixty-percent event, not a fifteen-percent event, because each factor is amplifying the effect of the others in a system that is now sitting close enough to threshold that any additional load can push it over.
I want to give you a concrete example that has nothing to do with wine, because I do not want you to think of wine as the villain in the story. It is not. It is one drop of water in one boat.
I had a stretch in the summer of 2024, before the July leave, when I had a specific interaction pattern I could see in the log. Long days of screen work — six or seven hours in front of a monitor, back-to-back video calls, the specific kind of digital load that comes with running a data organization inside a federal department — were, by themselves, a low-signal trigger. On a week when I slept well and the weather was stable, I could put in three screen-heavy days back to back and be fine. The screens were not causing attacks on their own. But if a screen-heavy day happened to coincide with a night of poor sleep — a night of only five to six hours instead of my baseline seven to eight — the attack rate on the following day was somewhere over sixty percent. Neither one alone was a high-probability trigger. Together, they were a near-certainty. That is multiplicative interaction. That is the coupled system doing its work. And it is the exact pattern the checklist would have told me was not a pattern, because on any given day of the log, the individual triggers were low-signal. The signal was only visible in the interaction.
Let me tell you the practical implication of Principle Two, because it is the most useful sentence in this chapter. On any given day, do not ask yourself "am I about to encounter a trigger?" Ask yourself "what is my capacity right now, and what are the inputs stacked against it today?" If your capacity is high, you can absorb multiple inputs — you can have the wine, do the screen work, take the long drive, and get through the day. If your capacity is low, even inputs that would ordinarily be safe become dangerous, because they combine with each other in ways that a linear read of the day cannot predict. This is the entire discipline of managing chronic migraine at the daily-life level. It is not avoidance of a fixed list. It is capacity management inside a moving budget.
PRINCIPLE THREE — YOUR THRESHOLD IS NOT FIXED, AND YOU HAVE MORE CONTROL OVER IT THAN THE CHECKLIST GAVE YOU
This is the one I want to leave you with, because it is the reason the whole book is worth reading. The threshold — the ceiling above which an attack becomes inevitable — is not a fixed number. It is a moving number. It moves up when you sleep well, hydrate well, eat well, move well, pray well, love and are loved well, treat your preventatives with respect, keep your inflammatory load low, and maintain the practices that keep a chronically ill nervous system as regulated as it can be. It moves down when any of those things is failing. And this is the crucial part — the amount by which your threshold moves is not a rounding error. Across the two years of my log, I could see clearly that my personal threshold moved up and down by what I would estimate as a factor of two or three between my best weeks and my worst weeks. That means that in a good week, I could absorb two to three times the trigger load I could absorb in a bad week and still remain below threshold. Two to three times. That is not a marginal effect. That is the entire game.
The implication of Principle Three is that migraine management is not the same thing as trigger avoidance. Trigger avoidance is a strategy for a body with a fixed threshold — you avoid the inputs, because you cannot change the ceiling. Threshold management is a strategy for a body with a moving threshold — you raise the ceiling, because when the ceiling is high enough, the ordinary inputs of an ordinary life stop being dangerous. It is a completely different orientation. It is the orientation that let me, in the year and a half since I built the tool, have entire months when I would previously have had eight or ten attacks and instead had two or three. Not because I avoided the wine. Not because I gave up the screens. Not because I moved to a windowless bunker in a stable-barometric-pressure region of the country. Because I raised the ceiling. I slept the eight hours. I drank the water. I walked when I could. I preserved the preventatives. I did the small daily work of maintaining the capacity of a nervous system I could no longer take for granted.
§
Christa saw this before I did. I want to acknowledge that here, because she deserves it, and because it is where the boat scene from Chapter Two finds its resolution. On the drive back from Myrtle Beach — the one where she told me she thought I could be doing more — she was not wrong about the general shape of the argument. She was wrong about the direction of it. She thought I was not doing enough of the things a person with a fixable disease does. What she was pointing at, without knowing it, was the ceiling. She had watched hundreds of athletes over her career do the daily unglamorous work of raising their thresholds — a runner recovering from a stress fracture does not "avoid running." She strengthens the surrounding musculature so that the same load becomes absorbable. Christa was pattern-matching against her own professional experience. She was right that there was capacity work to be done. She was wrong that it looked like trying harder in the acute moment. It looked like protecting the surrounding week.
Once I understood the Threshold Model, our conversations about my disease changed. Not because I finally agreed with her. Because she and I had a shared language for what we were actually doing. She stopped asking "have you done what you can do today?" and started asking "what does your capacity look like this week?" I stopped hearing "you are not trying hard enough" and started hearing "help me help you decide where the ceiling is." That is what a framework does. It is not a cure. It is a language. And a shared language between two people trying to survive a chronic illness together is worth more than any medication ever prescribed.
§
I want to teach you how to see the Threshold Model in your own log, because a framework you cannot see is a framework you cannot use. There are three practices, and if you add them to what you have already built in the last two chapters, you will have joint three of the parallel pipe operating in your own kitchen inside of thirty days.
PRACTICE ONE — A DAILY CAPACITY RATING
Every morning, take five seconds. Rate your capacity on a scale of one to ten. Not your pain. Not your mood. Your capacity to absorb load today. A ten is a day when you slept nine hours, feel rested, are hydrated, have no obvious inflammatory load, and could probably take on almost anything the day threw at you. A one is a day when you woke up already in prodrome, already exhausted, already close enough to threshold that the wrong five-minute conversation could push you over. You will find, in the first two weeks of doing this, that your capacity varies more than you thought. It is not "good" and "bad." It is a spectrum, and most days sit somewhere in the middle. Log it. This one number, across sixty days, will teach you more about your own disease than any specialist visit you have ever had.
PRACTICE TWO — THE THREE-DAY LOOK-BACK
Whenever you have an attack, before you write anything else about it, ask yourself what the last seventy-two hours looked like. Not the last hour. Not the last morning. The last three days. What was your sleep doing on each of them? What was your hydration? What was your screen exposure? What was your emotional load? What was the weather doing? Were there interactions you can now see that you could not see in the moment? Write two or three sentences. This is not a full log entry — you will already have your seven-dimension entry for the attack itself. This is a separate short note about the antecedents. You will find, within twenty attacks, that patterns emerge — clusters of two or three factors that seem to travel together in your particular body. Those clusters are your interaction signatures. They are the ones you cannot see with the checklist. They are the ones the Threshold Model was invented to reveal.
PRACTICE THREE — THE CAPACITY-RAISING INVENTORY
Once a month, take twenty minutes. Sit at a kitchen table with the log open. Ask yourself: what has been raising my capacity this month, and what has been lowering it? Write two lists. The raising list is the practices, foods, medications, environments, and relationships that seem to correlate with your higher-capacity days. The lowering list is the ones that correlate with your lower-capacity days. Do not confuse the lowering list with the trigger list — this is not about acute-moment inputs. This is about baseline shifters. Chronic stress belongs on the lowering list even though it does not directly trigger any single attack. A weekly walk outdoors belongs on the raising list even though missing it on any given day will not, by itself, cause an attack. Over months, these two lists become the map of your actual ceiling — the daily and weekly practices that move your threshold up or down. Once you can see the map, you can invest in the practices that raise the ceiling and disinvest from the ones that lower it. That is threshold management. That is the whole practical discipline the model exists to enable.
§
I want to close this chapter with a promise, because I have been asking a lot of you, and I want you to know why. I promised you at the beginning of this book that I would hand you the instrument I built for myself, and that the instrument would give you the one thing the medical system has not given anyone with your disease — visibility. In the last three chapters you have been building the visibility. The Seven Dimensions give you the shape of each attack. The Leaky Boat gives you the shape of the capacity between attacks. The Threshold Model gives you the shape of how the two interact. If you have been reading with a pen in your hand, and if you have been beginning to keep the log even in the smallest way — three lines a day, seven dimensions per attack, a capacity rating every morning — you are already becoming what the American healthcare system does not know how to refuse. You are becoming an evidence-generating patient.
§
A NOTE FOR NON-MIGRAINEURS
The Threshold Model generalizes even more cleanly than the two frameworks before it. Every chronic invisible illness I have encountered has a threshold and a moving capacity — the level of accumulated physiological load above which the system changes state — and the fact that you are living with a disease means, structurally, that you are living close enough to your threshold that ordinary inputs can push you over. That is what chronic illness is, at the systems level. It is not a disease that is always in flare. It is a disease that lives close enough to its threshold that the ordinary inputs of an ordinary life are enough to trigger it.
Post-exertional malaise in ME/CFS is a threshold event. A Long COVID crash is a threshold event. A POTS episode is a threshold event. A fibro flare is a threshold event. An endometriosis pain cycle interacts with a threshold that is itself moving across the month. If your particular disease has a name for its acute state — flare, crash, attack, episode, cycle — that name is another word for threshold crossing.
The three principles of the Threshold Model apply to all of these: the same input is not the same event; interactions are often multiplicative; the threshold moves, and you have more control over its position than the checklist gave you. The daily capacity rating, the three-day look-back, and the capacity-raising inventory work in every one of these diseases. Your log will look different — different triggers, different environmental fields, different hormonal contexts — but the shape of what you are tracking is the same shape. You are tracking the moving ceiling of a nervous system trying to hold state against accumulating load. That is what every one of us is doing, whatever the name on the chart says.
You have not yet built the discipline that turns the seeing you have learned into a permanent, sustainable record. That is what the next chapter is about. The Parallel Pipe — joint four — is the specific documentation architecture that lives underneath the frameworks I have taught you in the last three chapters. It is the actual log, in the actual structure, at the actual cadence a chronically ill life can sustain. After that, in Chapter Eight, comes the Clinician One-Page — joint five — where a full year of your data compresses into the ninety seconds your neurologist actually has to read.
You are further along than you know. The disease has been sitting in the dark for a hundred years because nobody has held a light to it. You are holding the light now. Do not put it down.
 
Chapter Seven
The Parallel Pipe
The first log I built for my own migraine disease had thirty-seven columns in it.
I want to say that plainly at the front of this chapter, because everything else I am going to teach you in the pages ahead is downstream of the specific humbling I received during the summer and fall of 2023 when I discovered that thirty-seven columns is not a data instrument. It is a project. A project is a thing you abandon in week three when the disease is taking your capacity faster than the project is producing insight. I abandoned mine. I want to tell you why.
The columns were, as best I remember them: date, time of onset, pain intensity, pain location, pain quality, aura yes-or-no, prodrome yes-or-no, nausea intensity, vomiting yes-or-no, medication one, dose one, time of dose one, effect of dose one, medication two, dose two, time of dose two, effect of dose two, sleep the night before, sleep quality, morning resting heart rate, morning HRV, blood pressure morning, blood pressure attack, hydration estimate, meals yes-or-no, caffeine intake, alcohol intake, screen exposure hours, fluorescent exposure hours, barometric pressure at onset, barometric change previous twenty-four hours, ambient temperature, humidity, sound level, hormonal notes, functional impact, and notes. Thirty-seven columns. I set it up in a spreadsheet, and I felt, for the first ten days of using it, the specific and beautiful satisfaction that a former Chief Data Officer of the United States Department of the Treasury feels when he sits down at his kitchen table with a well-columned spreadsheet and a problem he intends to solve.
By day thirteen I was skipping columns. By day nineteen I was skipping days. By week five the spreadsheet had entries for maybe half the attacks I had actually had, and most of the entries were incomplete, and the ones that were complete were spread across such varied levels of my own capacity — some captured immediately after the attack, when the data was fresh but my head was still fogged; some captured three days later, when the head was clear but the memory was already lossy — that I could not reliably compare one entry to another. I had thirty-seven columns of what an operations research analyst would call unreliable data. I had built the exact instrument the medical system would have loved me to bring in, and I could not sustain it. Which meant, in the specific arithmetic of chronic illness, that I had not built an instrument. I had built a monument to my own executive ambition. A monument is not a tool. A monument is a thing you look at. A tool is a thing you use.
That failure was the beginning of every framework in this book.
I want to give you what I learned from it, because it is the specific piece of practical craft I do not want you to have to learn the way I learned it — by wasting eight weeks of a chronically ill life on a spreadsheet you cannot maintain.
The parallel pipe is not a spreadsheet.
The parallel pipe is not an app.
The parallel pipe is the specific, sustainable, honest documentation discipline that lives underneath the frameworks I have been teaching you in the last three chapters, and that makes the frameworks in the next two chapters possible. The Seven Dimensions are a description of one attack. The Leaky Boat is a description of what is happening between attacks. The Threshold Model is a description of how the two interact. All three are frameworks for seeing. This chapter is the framework for keeping the record. Without the record, the seeing is transient. The disease will not permit you to remember what you saw. What you can remember, you can log. What you can log, you can look at again. What you can look at again, you can build a case on. That is the entire logic of the pipe.
§
Let me tell you what an honest, sustainable documentation instrument looks like.
It has three cadences. Only three. If you add a fourth you will collapse it, and I will not have done my job as the ghostwriter of your evidence architecture. If you skip one of them you will lose signal you cannot get back, and I will not have done my job either. I want you to internalize these three cadences the way an infantryman internalizes his morning weapons check. They are the discipline. They are the whole discipline.
CADENCE ONE — THE FIVE-MINUTE DAILY
Every morning of your life, take five seconds and record four things. Not five things. Not seven things. Four. I am going to name them for you. A capacity rating on a one-to-ten scale, as I described in the last chapter. A sleep duration estimate, in hours and rough minutes, from your wearable if you have one and from your best guess if you do not. A resting heart rate, if a device on your wrist can give you one, or a subjective one-word cardiovascular note if it cannot — "calm," "buzzy," "flat." A single environmental or life note, in one to three words — "screen day ahead," "rain moving in," "conflict with brother yesterday," "starting cycle day one."
Four fields. Five seconds. Every morning. I want you to notice what I have not included. I have not asked you to log your food. I have not asked you to log your hydration. I have not asked you to log your medication adherence. I have not asked you to log your steps or your exercise or your screen minutes or your caffeine. All of those are useful data, and none of them is worth what it will cost you to log them every day for the rest of your life. The four fields I have named are the minimum viable capacity picture — the smallest number of daily inputs that will let you, six months from now, look at any morning of any past week and know approximately what your body was doing. That is what you need. Nothing more. Anything more, in a chronically ill life, is a monument.
CADENCE TWO — THE PER-ATTACK LOG
Whenever you have an attack, whenever the acute state begins to lift, whenever you can hold a pen or a phone in a room that has not yet gone all the way back to being habitable — you write down the Seven Dimensions. I taught you the Seven Dimensions in Chapter Four. I am not going to teach them again. I want to say instead the specific thing that no one told me and that I had to learn by giving up on my thirty-seven-column spreadsheet: the per-attack log is the single most important entry in the entire pipe, and it should never take longer than five minutes to complete.
Five minutes. Not fifteen. Not thirty. Not "when I have the energy," which in a chronic illness is next week, which is too late. If your Seven Dimensions entry is taking more than five minutes, you are logging more detail than your capacity in that moment can sustain, and the specific consequence is that you will begin skipping attacks entirely. A skipped attack is worse than an abbreviated attack. An abbreviated attack — three of the seven dimensions honestly logged in ninety seconds — preserves signal. A skipped attack loses all signal. In the specific arithmetic of a case that will one day sit on the desk of a claims and ratings evaluator, an abbreviated entry from every attack of the last year beats a perfect entry from half of them, every time.
Log what you can. Log it fast. Log it in your own voice. Move on.
CADENCE THREE — THE TWENTY-MINUTE WEEKLY
Every Sunday evening, or every Monday morning, or whenever your household's rhythm has a small predictable pocket of quiet, you sit down for twenty minutes with the log open in front of you. Not longer. Twenty minutes. I want you to set a timer, because the disease will lie to you about how long you have been at it, and the specific way the disease lies is by hiding the passage of time inside the effort of concentration. Twenty minutes. Timer on. Timer off.
In those twenty minutes you do three things. First, you look back at the previous seven days. What was your average capacity rating? Where were the low days? Did you have an attack, and if so, what did the seventy-two hours before it look like? Second, you look ahead at the next seven days. What are the known stressors on the calendar? What is the weather doing? Where is the risk? Third, you name one specific practice you are going to invest in this week to raise your ceiling, and one specific practice you are going to disinvest from because you have noticed it is lowering it. One of each. Not five. Not ten. One and one. You write them at the bottom of the log with a date next to them, and the next Sunday, when you sit down again, you check whether you did the thing you said you would do.
Twenty minutes. Not more. If you cannot do the twenty minutes this week — a bad week, a family week, an attack week — you skip it. You come back the next Sunday. You do not try to make up the missed week by doing forty minutes the next Sunday. That is the thirty-seven-column mistake. Sustainability beats completeness. Every time.
§
Those are the three cadences. Five minutes daily. Five minutes per attack. Twenty minutes weekly. Roughly forty to ninety minutes a week, total, depending on how many attacks you have. That is the entire time budget the parallel pipe requires from you. If you protect that budget — protect it the way you would protect the small unguarded hour of a Sunday evening when the house is finally quiet — you will, inside six months, have a longitudinal record of your disease more complete and more decision-quality than any specialist you will ever see has ever received from any patient in their practice. I want you to sit with that. It is not a small claim. I make it because I have watched it happen, in my own case, in a career I lost, and I am not going to underclaim what I built during the losing.
Now let me tell you the practical mechanics.
WHERE TO KEEP THE LOG
There are three defensible answers to this question, and there is one bad answer.
The three defensible answers are: a spreadsheet, a purpose-built app, or a paper journal. A spreadsheet — Google Sheets, Excel, Apple Numbers — is what I used, and it is the answer I would recommend to a technically inclined patient who is comfortable with columns and rows and who can build one file that will sustain twelve months of daily entries without becoming unwieldy. A purpose-built app — there are several on the market and I will not endorse a specific one, because the market changes faster than a printed book can — is the answer I would recommend to a patient who wants the friction of logging to be minimal and who does not want to build their own structure. A paper journal is the answer I would recommend to a patient whose relationship with screens is antagonistic — as many chronic migraine patients' is — and whose most sustainable log is one that lives on a nightstand and is written in with a good pen.
The one bad answer is the patient's memory. I want to say that plainly, because too many patients answer this question by saying "I'll remember." You will not remember. Nobody remembers. The disease itself is a machine for erasing memory, especially the memory of what preceded an attack. If you do not write it down, in one of the three formats above, it is gone.
Whichever format you choose, you observe two rules.
Rule one: back it up. I learned this rule the hard way in December of 2024, when the laptop containing the primary copy of my log was, on the day of my removal, turned in to the Department of the Treasury security office and imaged out of existence. I had a backup on a home machine. That backup is the reason this book exists. If you are using a spreadsheet, keep a copy in cloud storage AND on a device that is not the primary logging device. If you are using an app, confirm the app's backup mechanism and use it. If you are using a paper journal, take photographs of each page as you finish it and keep the photographs in cloud storage. The record must survive the destruction of the specific object it lives on. There will be a day when it needs to.
Rule two: timestamp every entry. Not just the date. The time. Time-of-day matters in migraine and in every chronic invisible illness, and a log without timestamps is a log that cannot be cross-referenced against anything else in your life. If you are using a spreadsheet, add a time column. If you are using an app, make sure the app captures time not just date. If you are using a paper journal, write the time next to each entry. Five extra seconds. Enormous downstream value.
§
I want to give you the specific column structure I ended up with, because I have promised throughout this book that I will teach you what I built, and the honest fact is that I built one that works. Not thirty-seven columns. Not five columns. Somewhere in the middle.
Column one: date.
Column two: time.
Column three: entry type. One of three values only: DAILY, ATTACK, WEEKLY.
Column four: capacity rating (one to ten, for DAILY entries only).
Column five: sleep hours (for DAILY entries only).
Column six: resting heart rate or one-word cardiovascular note (for DAILY entries only).
Column seven: environmental / life note (one to three words, for DAILY entries only).
Column eight through fourteen: the Seven Dimensions (for ATTACK entries only). One column per dimension.
Column fifteen: weekly reflection (one paragraph, for WEEKLY entries only).
Column sixteen: investment (for WEEKLY entries — the practice you are going to invest in this week).
Column seventeen: disinvestment (for WEEKLY entries — the practice you are stepping back from).
Seventeen columns. Not one column ever has a value on every row. That is the point. A DAILY row uses columns one through seven. An ATTACK row uses columns one, two, three, and eight through fourteen. A WEEKLY row uses columns one, two, three, and fifteen through seventeen. The columns are structural placeholders that let you filter and sort later. Nothing else.
If seventeen columns feels like a lot, remember: you are only ever filling in a small fraction of them at any given entry. The daily takes five minutes because you touch seven columns. The attack takes five minutes because you touch seven columns. The weekly takes twenty minutes because you touch three columns and think. The total cognitive load of any single logging session is small. The total data yield across a year is enormous. That is the ratio you want.
§
I want to close this chapter with the honest sentence I owe you about failure.
You are going to fail at this.
You are going to have weeks when the daily does not get done. You are going to have months when the weekly does not get done. You are going to have attacks that go unlogged because you were in too much pain to hold a device. You are going to have periods, sometimes long ones, where the whole discipline falls off a cliff and you stop opening the log entirely. That will happen because you are chronically ill, and chronic illness does not permit uninterrupted discipline in any domain of a life. I want to tell you what to do when that happens, because I did it wrong for the first eighteen months and I do not want you to lose eighteen months to the same mistake.
Do not try to make up the missed days.
Do not go back and reconstruct.
Do not scold yourself.
Open the log tomorrow morning. Do the five-minute daily. If you had an attack in the missed stretch, log whatever you remember of it in the ATTACK format, honestly, with a note that says "logged from memory five days later — signal degraded." That note is itself signal. A neurologist reading a log with that note in it can see the shape of the missed week and knows to weight it appropriately. A claims and ratings evaluator can too.
Then keep going. The log does not need to be beautiful. It only needs to be true. What it needs across a year — the shape, the shadow, the movement of your capacity, the pattern of your attacks — will emerge from an imperfect record with the same fidelity as from a perfect one, provided you do not stop. Do not stop. Stopping is the only failure mode that matters. All the other failure modes are life.
§
A NOTE FOR NON-MIGRAINEURS
The parallel pipe works, unchanged, for every chronic invisible illness I have encountered. The three cadences are the same. The seventeen-column structure is the same. The rules — back it up, timestamp every entry, do not try to make up missed days — are the same.
What changes is what goes into the Seven Dimensions columns. If you have ME/CFS or Long COVID, those columns hold your fatigue profile, your post-exertional malaise, and your autonomic readings. If you have fibromyalgia, they hold the map of your tender points and the specific inflammatory signature of a bad day. If you have endometriosis, one of them holds a running record of cycle day and the specific pain pattern of the month. If you have POTS, several of them are your orthostatic readings.
The pipe does not care what the disease is. The pipe cares that you build it, that you feed it, that you back it up, that you honor the twenty-minute rule. Do that, and inside six months you will have the same instrument I have, translated into the language of your particular body.
§
The next chapter is the first artifact your pipe will produce. I have taught you how to see, in three joints. I have taught you how to log, in this one. The next chapter is where the log begins to speak — where a full year of your data compresses into a single page a neurologist can read in ninety seconds and cannot ignore. That is the Clinician One-Page. It is joint five of the parallel pipe. It is the artifact that changed my relationship with every doctor I saw in the last eighteen months of my career. It will change yours too.
 
Interlude
Sunday Evening
It is a Sunday in November. It has begun to get dark early. Christa is upstairs on a phone call with her sister. M. is in the family room in a specific corner of the couch he has appointed as the corner from which he does his best work with the AAC device. A. is at the kitchen island with a book that is not for school and that she has been reading in silence for forty minutes, which for a nine-year-old is a specific accomplishment I want to note without interrupting. The house is doing the small settled hum it does on a Sunday evening after supper, and I have opened the log.
It has been running now for almost three years.
I sit at the kitchen table with the laptop open in front of me and a mug of tea — I still do not drink coffee unless a migraine is asking me to — and I set a timer on my phone for twenty minutes. I have said, in the last few chapters of this book, that the twenty-minute timer is not a decoration. It is a discipline. It is a Sunday. The timer is on.
I look back first. Last week, on the page, was a five-point-eight average capacity. Not a great week. Not a bad week. The average is a small blessing that would not have been available to me three years ago because three years ago I did not have averages. I had impressions. Impressions in a chronic illness are unreliable in a specific direction I have described earlier in this book, which is that they tend to overweight whatever the last forty-eight hours were. My last forty-eight hours were fine. Without the log, I would say to Christa, if she asked, that last week was fine. With the log, I can say to Christa, if she asks, that the last two Sundays have averaged five-point-eight, that the two Sundays before that averaged six-point-four, and that my capacity trajectory has been softening for about a month in a way that I want to pay attention to. That is a different sentence than fine. It is also, if she asks, a sentence I can produce in about eleven seconds.
There was one attack. Wednesday afternoon. A four-hour job, aborted at hour two by an on-time triptan. The Seven Dimensions on it are already in the log because I wrote them Wednesday evening, in about seven minutes, from the specific chair I use when I am recovering. I open the entry. I skim it. I add one thing I did not have the presence of mind to add on Wednesday, which is a small note in the environmental column: "Long screen morning, unplanned." That is a piece of signal I could not have seen without the log. Wednesdays are my long screen mornings. This one had extended into the early afternoon because a client call had run over. The interaction signature involving screen work is the second bullet on my current Clinician One-Page. I add the note. I move on.
I look ahead. Next week is a normal week on the calendar with one exception. Wednesday I have a two-hour meeting on-site that is going to be a screen-and-fluorescent day back-to-back with the client call that ran over this week. I know now, from the log, that back-to-back Wednesdays of that specific character have a threshold-crossing risk I want to protect against. I move the client call to Thursday. Christa can drive me on Wednesday because I do not want to drive home tired. I text her: "Can you drive me Wed?" I do not explain. She will understand. She reads about three chapters of the log on her own account and she knows what a screen Wednesday is. The instrument has given us shared vocabulary I described in the last chapter. This is that vocabulary in an ordinary text message on a Sunday evening.
She writes back: "Yes. Also we're out of the good yogurt."
I laugh. I do not know why the yogurt is the good yogurt. I know only that she is upstairs on a phone call with her sister and has apparently opened the refrigerator during the call because that is the specific quiet efficiency of my wife on a Sunday. I add yogurt to the grocery list. This is not part of the log discipline. This is the ordinary domestic exchange of a household with children in it. I mention it because I do not want to give you the impression that the twenty-minute Sunday timer is a monastic thing. It is a kitchen thing. It happens in the same room where the yogurt shortage is discovered.
I have four minutes left on the timer.
I write the two sentences of the week. The investment I am going to make this week is a walk on Saturday morning that I have been letting slip. The disinvestment is scrolling news on my phone during the specific late afternoon window when my capacity is at its lowest and doom-scrolling costs me more than it should. I write both of those with a date next to them and I close the log.
The timer has forty-one seconds left. I use them to look at A., who has not looked up from her book in the entire twenty minutes. I use them to notice M., who has done seven things on the AAC device and is currently negotiating with one of his sisters, in text on the app, about a snack I did not know was on the table. I use the last ten seconds of the timer to say a specific thing in my head that has become a Sunday-evening habit of mine and that I want to give you as a small pastoral gift because it is the closest thing to a prayer that I have found in this discipline.
Thank you, God, for the record. Thank you for the week that is behind me. Thank you for what I can see of the week ahead of me. Keep me honest inside the instrument. Keep me honest outside of it. Amen.
The timer rings.
I close the laptop. I get up. I go find out what the negotiation about the snack is. A. keeps reading. M. hands me his device to show me something small he has just discovered on it and has been waiting for me to be done so he could share. Christa comes down the stairs at some point in the next fifteen minutes and says something about her sister.
The next chapter is about the first output of the pipe. This one was just a Sunday.
 
Chapter Eight
The Clinician One-Page
Your neurologist has twelve minutes with you.
I want you to sit with that number, because everything I am going to teach you in this chapter is downstream of it, and if you refuse to accept it — if some part of you still believes that a specialist practicing in an American healthcare system in 2026 is going to spend forty-five minutes with you at your next appointment — you will build the wrong artifact. The number is twelve. Sometimes it is fifteen. Sometimes it is seventeen, on a Tuesday morning when the schedule has not yet slipped and the patient before you canceled. Sometimes it is nine, on a Thursday afternoon when everyone in the practice has been running behind since ten in the morning and the physician you are seeing is triaging her own cognitive load between what your chart says and what the next patient's chart is going to say when she opens it in six minutes. The average is twelve. I am going to build the rest of this chapter on twelve.
If you have twelve minutes with a specialist and you spend nine of them narrating the last three months of your life in the language of a chronically ill person who has been erased from every conversation about her body for the last several years — the specific too-much, too-slow, apologetic-and-then-defensive narration that we all fall into, myself included, when we are given the rare rooming with someone who might listen — you will run out of clock before you have said the thing your neurologist actually needs to hear in order to make a decision that will change the next quarter of your life. She will nod. She will renew the prescription she gave you last time. She will schedule a follow-up in three months. You will walk out of the appointment having spent nine minutes talking and having received three minutes of medical decision-making, and the specific tragedy of the visit is not that she was a bad physician. She was a physician doing her best inside a twelve-minute container. The tragedy is that you gave her the container, and she filled it.
The Clinician One-Page is the correction.
The Clinician One-Page is a single sheet of paper — literally one page, printed on one side, handed to your neurologist at the beginning of the visit or emailed to her office forty-eight hours in advance — that compresses twelve months of the finest longitudinal data ever assembled about your particular body into ninety seconds of reading time. Ninety seconds. I mean that literally. When I hand my one-page to a physician now, she reads it in ninety seconds. Sometimes she reads it in sixty. When she looks up from the page, she has more accurate information about my last twelve months of migraine disease than any spoken narration I could have given her in the nine minutes she was going to grant me. She has that information because I built the page for her, not for me. That is the whole design principle. The page is not a document about your suffering. The page is a document about her decision. Everything on the page is there to answer a question she is going to have to make a call on. Everything not on the page is there to protect the ninety seconds.
I want to teach you how to build one.
§
The page has a header, five sections, and a closing question. That is the entire architecture. I have used variations of this format across the last eighteen months of my care. I have watched neurologists, primary-care physicians, pain specialists, and one occupational medicine officer working for the Department of Veterans Affairs read the page and change their approach to me within ninety seconds. It is the single most consequential artifact your parallel pipe will produce. It is joint five of the pipe. It is where the log begins to speak.
THE HEADER — three lines, no more
Line one: your legal name and your date of birth. Line two: the twelve-month window the page covers, e.g. FEBRUARY 2025 THROUGH JANUARY 2026. Line three: the version number and the date you generated this specific one-page, e.g. VERSION 7, GENERATED 24 JANUARY 2026. The version number and the date are important. They tell the physician that this is not a one-time document — that you have been building and rebuilding it, that you will be back with an updated version at the next visit, and that she is looking at a snapshot from a specific point in a running record. That framing changes how she reads it. It moves the page from "one patient's dossier" to "a signal from a system that is going to keep producing signal." That is a different kind of clinical artifact. She will read it with different attention.
SECTION ONE — THE ATTACK SUMMARY
Four numbers. Not five. Not eight. Four.
Number one: total attacks in the covered period. Number two: mean attack duration in hours. Number three: mean peak intensity on your own honest calibrated scale, one to ten. Number four: total hours of acute-phase disability across the period. That fourth number is the one your disability adjudicator will care about the most, and it is the one your neurologist has never been given by any patient, and it is the number that will, more than any other single item on your one-page, cause her to sit up in her chair. I am going to say it again. The total number of hours of acute-phase disability across a twelve-month period is a number no chronic migraine patient walks into a neurologist's office with. When you walk in with it, calculated from a real log, you are already in a different category of patient. That category matters.
SECTION TWO — THE PATTERN VISUAL
A single small chart. I want you to hear me on this: one chart, on a one-page, is worth eight sentences of narrative. Two charts is worth zero sentences, because two charts is where the physician's eye glazes over and she starts skimming. One chart, well-chosen, well-labeled, occupying no more than about a fifth of the page.
The chart I use is a bar chart of attacks per week across the twelve-month window. The horizontal axis is week number. The vertical axis is number of attacks that week. The bars are simple grey. There is no color. There is no legend. There is one horizontal line drawn across the chart at the level of "a normal migraine patient's weekly attack rate," which for my population is somewhere between one and two. Anything above the line is a bad week. Anything below is a good week. My neurologist can, in the five seconds she looks at the chart, see the entire trajectory of my year — the two-month stretch in the summer when a preventive medication was working, the crash in September when it stopped, the slow climb back down in December after a course adjustment. She sees, in five seconds, what forty minutes of my talking could not have given her.
If a bar chart is not the right format for your particular disease, pick a different one. A heat calendar — the small twelve-by-thirty-one grid where each cell is shaded darker for a worse day — is excellent for conditions where attacks are less discrete than migraine. Long COVID fatigue often shows better as a heat calendar than as a bar chart. Fibromyalgia can go either way. Pick the one chart that lets a physician see the year in five seconds. There is only one such chart per disease per patient. Find yours. Use only that one.
SECTION THREE — THE INTERACTION SIGNATURES
Three bullet points. I want to say that in a chapter that has otherwise been avoiding bulleted lists: the interaction signatures section is one of the few places on the whole one-page where bullets do actually work, and the reason is that bullets are the specific formatting a physician's eye is trained to consume in under fifteen seconds. A physician reads three bullets in under fifteen seconds. A physician reads a paragraph of the same content in over sixty. On a one-page whose entire promise is ninety seconds of reading time, you cannot afford paragraphs where bullets do the same work.
The three bullets should name the three most consistent interaction signatures you have identified from your log. These are not triggers. These are the combinations of factors that, in your particular body, precede an attack more often than any single factor alone. The Threshold Model chapter taught you how to see these. This section reports them.
An example from my own one-page, from a version I used with my neurologist in November of 2025:
• Sleep under six hours + barometric drop of four or more millibars within twenty-four hours: attack rate 68%.
• Screen work over five hours + poor sleep the preceding night: attack rate 61%.
• Emotional-load week (defined) + missed magnesium dose within seventy-two hours: attack rate 47%.
Three bullets. Three interaction signatures. Each with an approximate rate — I use percentages rounded to the nearest whole percent, calculated from at least twenty prior occurrences of the interaction in my log. Nothing in a bullet exceeds one line of text. My neurologist, reading those three lines in fifteen seconds, knows more about my particular disease than she could have learned from three visits of open-ended conversation.
SECTION FOUR — THE MEDICATION RESPONSE MATRIX
A small table. Three columns.
Column one: medication name and dose. Column two: what it is intended to do — abortive, preventive, adjunctive. Column three: what it has actually done in your case, quantified. The third column is where the whole matrix earns its space on the page. Instead of "rizatriptan — helps sometimes," you have "10 mg rizatriptan ODT — 74% partial abort when taken within 20 min of onset, 12% partial abort when taken after 90 min, sample size 47 attacks."
The medication response matrix is the single most useful thing your neurologist will see on the entire page for the purposes of adjusting your current regimen. Physicians make medication decisions based on the response data they have. Their default response data is what you tell them in the room, which is unreliable because the disease compromises the memory of medication response, and what previous chart entries say, which is what previous physicians wrote down in previous unreliable rooms. The medication response matrix breaks that cycle. It gives your physician a clean, quantified table of what your body is actually doing in response to what you are currently taking. She reads it in twenty seconds. She then spends the remaining eight minutes of your appointment making decisions with the benefit of the twenty seconds she just spent, instead of the four minutes she would have spent trying to reconstruct the response history through conversation.
SECTION FIVE — THE CAPACITY TRAJECTORY
Two sentences. Not three.
Sentence one: a statement of whether your capacity — your ability to function at the level you would ordinarily function at — has been improving, stable, or declining across the covered period. Sentence two: what you believe is currently the largest single lever you and your physician could pull to improve it.
That is it. Two sentences. I want you to notice what the second sentence does. It flips the frame of the visit. Instead of the physician deciding what the appointment is about and you responding, you have arrived at the appointment with a stated hypothesis about what would matter most, and the physician's job is now to evaluate your hypothesis rather than construct her own from scratch. You have not seized authority. You have offered a specific starting point for the shared decision-making that is supposed to be happening in the room. Physicians almost always welcome this. It gives them a place to begin.
An example from a version I used at a January 2026 appointment:
Capacity has been slowly improving since the medication change in November. I believe the largest current lever is titrating up the CGRP dose to see whether it further reduces the interaction signatures identified in Section Three.
Two sentences. A whole appointment's worth of collaborative direction.
THE CLOSING QUESTION
The last line of the page — the very last line, positioned so the physician's eye lands on it after she has read everything else — is a single specific question you need her to answer at this visit.
Not three questions. One.
The choice of the one question is the single most important curatorial decision you make on the whole page. You will have five or six questions you would like to ask. You have to pick one. The one you pick should be the one that will affect the largest possible surface area of your care going forward. It should not be a question about a symptom. It should be a question about a decision. Compare:
Bad closing question: "Are these visual changes anything to worry about?"
Good closing question: "Given the interaction signatures in Section Three, do you want to change the abortive strategy, the preventive strategy, or leave both in place and give the current regimen another quarter?"
The bad question asks the physician to render an opinion about a symptom. The good question asks her to render a decision from a bounded menu, given the data you have already put in front of her. Physicians answer good questions much better than they answer bad ones. Not because they are unwilling to answer bad ones. Because a good question is easier to answer well. You are, quietly, making her job easier. She will thank you for it, and the answer you get will be worth ten of the answers you would have gotten from an open-ended question.
§
Those are the six components. Header. Attack summary. Pattern visual. Interaction signatures. Medication response matrix. Capacity trajectory. Closing question. All on one page. All designed to be read in ninety seconds.
Let me tell you the rules of the page, because they are not decorative and they will not survive a first draft if you do not know them going in.
RULE ONE — ONE PAGE MEANS ONE PAGE
Not two. Not one and a half with the last third of the second page nearly empty. One physical page, printed on one side. If your one-page is running to a page and a half, the discipline of cutting it back to one page is the discipline of building the artifact correctly. Something on the page is a decoration. Cut it. Every square inch of the page has to earn itself against the alternative of being blank. Blank space, on a well-built one-page, is not wasted space. Blank space is the specific breathing room that allows a physician's eye to move through the page in ninety seconds instead of two minutes.
RULE TWO — NO JARGON YOU DID NOT LEARN FROM A DOCTOR
If you cannot point to a specific previous visit where a specific physician used a specific term to describe your condition, do not put that term on the page. The one-page is not the place to demonstrate that you have read the migraine literature. It is the place to demonstrate that you are a rigorous observer of your own body in the language your care team already uses. If your neurologist has never used the word "sensitization" in your presence, do not use it on your one-page. If she has, use it accurately and only once.
RULE THREE — DELIVER IT BEFORE THE APPOINTMENT WHEN POSSIBLE
Some physicians' offices have a patient portal that accepts uploaded documents forty-eight hours in advance. Use it if yours does. The physician who has had ninety seconds to read your one-page BEFORE she walks into the room walks into a different appointment than the physician who reads it during the room. The first appointment starts at the point the second one would have arrived at in minute ten. If the patient portal does not accept documents, hand her the printed page at the moment she walks in, before you sit down, and give her the ninety seconds explicitly: "I know we have limited time. I've prepared a one-page summary of the last twelve months of my case. Take a moment to read it — I'll wait." I have never had a physician refuse that request. Most have thanked me for it.
RULE FOUR — UPDATE IT FOR EVERY VISIT
The version number in the header is not decoration. It is a discipline. Every appointment gets a fresh one-page, generated within seventy-two hours of the visit, reflecting the current state of your log. Do not reuse an old one. The point of the version number is that your physician can, if she chooses, ask for the version you gave her at the last visit and compare — and if she can compare, she is seeing the trajectory of your disease with a precision that no chart review has ever given a specialist about a patient. That comparison is the highest and best use of the one-page across time. It is where the artifact stops being a document and becomes a record.
§
I want to give you the specific thing that happened the first time I used a one-page in front of a physician, because it is the moment I understood what I had built and I do not want you to miss the analogous moment in your own care.
I had been seeing a neurologist for about a year at that point — a good one, in a practice that took my case seriously, and one who had been gently trying, across several appointments, to adjust an abortive regimen that was not fully working. I had, until that visit, been giving her the ordinary chronic-migraine patient narration: the last month has been rough, the medication is helping a little, the sleep is inconsistent, I've had maybe six or seven attacks. She had, until that visit, been making the ordinary responses: we could try increasing the dose, we could add a preventive, we could wait another quarter. Both of us had been operating inside the twelve-minute container the appointment permits. Both of us had been doing our best inside it.
At that visit I handed her a one-page as she walked into the room. I said the sentence I gave you two rules back: I know we have limited time, I've prepared a one-page summary, take a moment to read it, I'll wait. She sat down. She read the page. It took her about seventy seconds. She looked up. She said, in a tone I had not heard her use before: "Roger, this is remarkable."
That is not a self-congratulatory sentence. I include it because of what she said next. She said: "I'm going to change what I was about to do." She had, walking into the room, been prepared to increase my abortive dose. She saw, on the one-page, that the interaction signature involving screen work and poor sleep was accounting for more of my recent breakthroughs than the abortive dose was. She revised her plan in the room, based on data she could not have accessed by asking me questions. We ended the appointment with a targeted intervention on the screen-and-sleep interaction — a specific behavioral change she had never before had the data to prescribe — and a small change to the preventive rather than the abortive. Two weeks later my attack rate dropped by roughly a third.
I want to tell you the moral of that story, because it is not what most patients would guess. The moral is not that I saved my own appointment by handing her a document. The moral is that I gave her the ability to make a better decision inside the twelve minutes we had. She would have made the best decision she could have made without the page. With the page, she made a decision she could not have made without it. That is the entire promise of the Clinician One-Page. Not that it makes you a better patient. It makes your physician a better physician within the container the American healthcare system has given her to work in. She was always ready to help you. The system has been asking her to help you with a signal so degraded she could not have. The page fixes the signal. It does nothing else. It does not need to.
§
A NOTE FOR NON-MIGRAINEURS
The Clinician One-Page works, unchanged in architecture and changed only in content, for every chronic invisible illness. The header is the same. The five sections are the same. The closing question is the same. The rules are the same.
What changes:
If you have ME/CFS or Long COVID, Section One's Attack Summary becomes a Crash Summary or an Exertion Threshold Summary. Section Two's Pattern Visual is almost always a heat calendar rather than a bar chart, because your bad days are less discrete than migraine attacks and a heat calendar shows the shape of them better. Section Three's interaction signatures will lean heavily on exertion combined with cognitive load combined with sleep debt.
If you have fibromyalgia, Section One's summary becomes a Flare Summary. Section Three's signatures often include specific inflammatory triggers combined with sleep quality. Section Four's medication matrix expands to include the non-pharmacological interventions — heat, movement, specific stretch protocols — that are more central to fibromyalgia care than to migraine care.
If you have endometriosis or a cycle-linked condition, Section Two's visual almost always benefits from a cycle overlay — a small annotation showing the phase of the cycle across the twelve months. Section Three's signatures will center on cycle day interactions.
Whatever the disease, the page is one page, read in ninety seconds, delivered before the appointment when possible, refreshed for every visit. Your neurologist / rheumatologist / gynecologist / cardiologist / primary care physician is the same physician regardless of what she has been trained in. She has twelve minutes. Give her the ninety seconds. She will use the remaining ten and a half.
§
The next chapter is where the same discipline moves from the exam room to a much colder room: the desk of a claims and ratings evaluator who has never met you, will never meet you, and will make a decision about your life on the basis of what fits inside a claims folder. The Disability Evidence Chain — joint six of the parallel pipe — is how the artifact you built for your neurologist translates into the artifact chain a disability adjudicator cannot dismiss. Same instrument. Different reader. Different stakes. It is the final joint of the pipe, and it is the one that changed my ability to have my case seen at the highest and hardest level of the system. I want to show you how to build it.
 
Chapter Nine
The Disability Evidence Chain
The claims and ratings evaluator has never met you.
She will never meet you. She will make a decision about the next portion of your life on the basis of what fits inside a claims folder — a physical or, more often now, digital folder that contains the artifacts you have submitted, the artifacts your physicians have submitted on your behalf, and whatever her adjudicative system has automatically pulled from other federal or insurance databases. She has, depending on her workload, somewhere between forty and ninety minutes to read the folder and render a decision. Some of that time will go to procedural checklist items. Some of it will go to reading medical records that were not written for her. Some of it will go to your own words, if you have submitted a personal statement. And some of it — the small, uncounted, decisively important minutes that will make or unmake your case — will go to whether the folder in front of her tells a coherent story or a chaotic one.
The Disability Evidence Chain is how you build the coherent one.
I want to be careful about how I frame this chapter, because I do not want to give you the impression that a disability adjudicator is a bureaucratic villain. She is not. She is a human being reading a hundred and forty case files a month, working inside a system that has been designed to be adversarial in exactly the way the American healthcare system's evidentiary layer has been designed to be lossy. She is not going to be helped by an angry file. She is not going to be helped by a file that is trying to move her. She is going to be helped, in the specific and narrow sense that any professional reader is helped by good source material, by a file that has been assembled so that she can find what she needs to find, verify what she needs to verify, and reach the decision the law and her workload permit her to reach. The chain you are going to build in this chapter is designed to give her that file. Nothing else.
I want to give you the sentence that captures the whole discipline:
The clinician cares what your body is doing. The adjudicator cares what your file is saying. The gap between those two objects is where most chronic-illness disability claims are lost.
The Clinician One-Page from the last chapter closed the first gap. The Disability Evidence Chain closes the second.
§
Let me tell you what a chain is, in the specific meaning I am using here.
A chain is not a single document. A chain is a sequence of linked artifacts, each of which points to the next, so that any place a reader enters the chain she can find her way to any other place in it within seconds. A well-built chain has six links. Each link does one specific thing. Each link references the links on either side of it. Together they form the coherent story an adjudicator needs to read a case fairly.
I want to walk you through the six links one at a time, in the order they should appear in the folder. Every link has a specific job. Every link is short. Every link can be built, in a chronically ill life, without a lawyer, though a good disability attorney can improve any one of them if you can afford one and I am not going to pretend otherwise.
LINK ONE — THE COVER PAGE
The first document in the folder is a single-page cover. Its job is to orient the reader in under ninety seconds. It is a close cousin of the Clinician One-Page, and if you have already built one of those, you have already done most of the work of this one. The differences are:
The header includes not just your name and date of birth but the case number, the claim number, the date of your application, and the name of your primary treating physician.
The Attack Summary section is expanded to include the twelve-month figures AND a rolling twenty-four-month comparison — because a disability determination is a determination about a trajectory, not a snapshot. A single bad year, in adjudicator eyes, could be an anomaly. Two bad years, well-documented, is a case.
The Pattern Visual is the same single chart from your Clinician One-Page, though I would recommend extending the horizontal axis to cover the full window of your claim rather than only twelve months.
The Interaction Signatures section stays exactly the same. Three bullets. Same percentages. Same discipline.
The Medication Response Matrix expands to include every abortive and preventive you have tried across the full period of the claim, whether or not you are still taking them. Failed medications matter to adjudicators the way successful ones matter to physicians. A record of ten failed abortives is not a record of a difficult patient. It is a record of an evidence-generating patient whose disease has been rigorously investigated and has resisted the interventions offered by her care team. Both readings of that record exist. Your file has to make sure the adjudicator gets the second one.
The Capacity Trajectory section closes with a functional statement rather than a treatment hypothesis. Not "the largest lever we could pull is X" but "my capacity is currently insufficient to sustain the essential functions of my occupation, and the trajectory across the covered period supports that conclusion." Same instrument. Different reader. Different close.
LINK TWO — THE PHYSICIAN CORROBORATION
The second document is a letter — or, better, two or three letters — from your treating physicians that specifically reference the artifacts you have built. I mean that literally. A physician letter that says "the patient has severe chronic migraine, unable to work" is worth less than a physician letter that says "the patient has provided me with a longitudinal data set consisting of X months of daily and per-attack logging, from which the following patterns are directly observable" and then goes on to describe the patterns from your one-page.
I want to say the specific thing you need to do to get that letter, because it is not intuitive:
You do not ask your physician for a general letter. You ask her for a letter that references your data. When you make the request, you provide her with a copy of your Clinician One-Page and a copy of the cover page from Link One, and you say something like: "I am filing a disability claim. Would you be willing to write a letter that specifically references the data I have been providing you across our appointments? I have a one-page summary of the current state of my case that you may want to reference in the letter."
Every physician I have made that request of has said yes. Not because they are being generous with their time. Because it is easier to write a letter that references specific data than it is to write a letter that generalizes about a patient. Your data is doing the work. She is signing off on it. Both of you are doing your appropriate share of the labor of the letter. That is a good exchange.
If you have more than one physician involved in your care — a primary, a neurologist, a specialist for a comorbid condition — get a corroborating letter from each of them, and make sure each letter references the same underlying data set. Adjudicators notice cross-corroboration. Three physicians independently referencing the same twelve months of your longitudinal log is not three redundant letters. It is a triangulated case.
LINK THREE — THE FUNCTIONAL IMPACT STATEMENT
The third document is your own written statement of what the disease has done to your capacity to function. I want to distinguish this from the medical narrative that most patients try to write in this position, because the two are not the same.
A medical narrative describes what the disease is. "I have chronic migraine. My attacks are severe. They involve nausea, light sensitivity, cognitive impairment..." That is not what an adjudicator needs from you. She has three physician letters and a cover page describing what the disease is. What she does not have — what no one has given her but you — is a specific, quantified, first-person account of what the disease has prevented you from doing.
The functional impact statement is a list. It should be five to eight items long, no longer. Each item is a specific function you were able to perform before your disease progressed to its current state, and that you can no longer reliably perform. Each item is written in operationally verifiable language. I mean that in the specific sense a claims examiner uses it: a fact that could, in principle, be verified by a third party.
Compare the following two ways of describing the same reality:
Not usable in an adjudication file: "I can no longer concentrate the way I used to. I get tired. I miss things at work."
Usable in an adjudication file: "I am unable to sustain focused attention on written material for more than approximately twenty minutes without visual disturbance or the onset of cognitive symptoms. I am unable to attend meetings of more than one hour in duration without requiring a break to a dark environment. In the twelve months preceding this claim, I missed a total of 1,147 documented work hours to acute migraine events, an amount equivalent to thirty full working weeks."
The second is a functional impact statement. The first is a lament. Both are true. An adjudicator can only act on the second. If you write the first, she will nod sympathetically and read down. If you write the second, she will make a decision based on it.
Every item in your functional impact statement should be paired, wherever possible, with a specific number from your log. Not the whole number — the specific number that supports that specific item. This is the point at which the parallel pipe you have been building for the last five chapters starts paying dividends that no chronically ill patient without a pipe can access. Your log has the numbers. You have already built the discipline. You are now cashing it in.
LINK FOUR — THE TREATMENT HISTORY
The fourth document is a comprehensive treatment history. Everything you have tried. Every medication. Every non-pharmacological intervention. Every specialist consulted. Every diagnostic procedure. Dates. Duration. Outcome.
I want to be specific about what an adjudicator is looking for when she reads a treatment history, because it is not what most patients think it is.
She is not looking for evidence that you have tried hard. She is looking for evidence that the standard-of-care interventions available to you have been exhausted, and that the residual disability that motivates your claim exists in a patient who has done the reasonable work of trying to be treated. She is looking, in adjudicative language, for the presence or absence of "failure of conservative treatment." If your treatment history documents the failure of a full course of conservative treatment — first-line abortives, first-line preventives, adjunctive therapies, specialist consultations — you have provided her with the exact category of evidence her rating system is designed to weight most heavily.
Format the treatment history as a table. I know I have said in the last two chapters that tables are dangerous. In an adjudication file, a table is the correct format for this specific document. Adjudicators are trained to read tables. A table gives her a scannable inventory she can cross-reference against her rating criteria without having to reconstruct the inventory from prose. Every row is one intervention. Every row includes: intervention name, prescribing physician (if applicable), start date, end date, outcome one-word summary (effective, partially effective, ineffective, discontinued for side effects, ongoing).
LINK FIVE — THE THIRD-PARTY STATEMENTS
The fifth document — or set of documents — is what claims examiners call collateral evidence, and what I want you to think of as the witnesses. You have been paying the Convincing Cost for years, and one of the things the Convincing Cost has done to you is convince you that only your own report of your own body is admissible. It is not. In a disability file, the observations of the people who have been around you — your spouse, your adult children, your close colleagues, your supervisor, your neighbors, in that rough order of weight — are admissible evidence. They will often not carry the day. But they are, in the specific case of an invisible illness, some of the most valuable evidence in the folder.
The reason is exactly the reason this book started with: the world cannot see your disease. A witness statement is the closest thing the folder will contain to somebody else corroborating what the disease has been doing to you. A statement from your spouse describing the specific frequency, duration, and severity of your acute attacks — written in her own voice, from her own observation, dated and signed — is worth more than three additional pages of your own narration. It is not a substitute for the medical evidence. It is the connective tissue that binds the medical evidence to the functional impact.
Solicit two to four such statements. Provide each witness with a short guidance document — one page — that describes what you are asking for, what you are not asking for, and what the format should be. Do not write the statements yourself. Do not tell your witnesses what to say. Provide the guidance and let them write. The specific value of the collateral statements comes from their independence. If they read as though you drafted them, an experienced adjudicator will notice and the whole set will be discounted.
LINK SIX — THE APPENDICES
The sixth document is not a document. It is a set of appendices. Everything the first five links reference, in the order they are referenced.
Do not include the full log. A twelve-month day-by-day migraine log with two thousand rows of data is not something a claims examiner is going to read, and it is not something you are going to send her. What she needs, and what the appendices should contain, is the specific subset of your data that supports the specific claims made in the first five links. A one-page summary of your monthly attack counts. A one-page medication response summary. A one-page cycle overlay if relevant. A copy of the Clinician One-Page you are currently using with your neurologist. A copy of the previous version. Nothing else.
An adjudicator who wants more can request more. A well-built folder tells her, implicitly, that more exists and can be produced. That is the position you want. You do not want a folder that overwhelms. You want a folder that credibly promises depth.
§
Those are the six links. Cover page. Physician corroboration. Functional impact statement. Treatment history. Third-party statements. Appendices. In that order. In that folder.
Let me tell you the rules of the chain, because they are not obvious and they will cost you months of appeal work if you get them wrong.
RULE ONE — EVERY LINK REFERENCES THE LINKS ON EITHER SIDE OF IT
The cover page references the physician corroboration by name. The physician corroboration references the specific data in the cover page. The functional impact statement references the treatment history by intervention name and date. The treatment history references the physician corroboration by physician name. The third-party statements are the only link that does not follow this rule, and even they benefit from a small header sentence that names the case number and the date they were prepared.
Why does this matter? Because an adjudicator entering the folder at any point should be able to trace her way to any other point in under ten seconds. Cross-referencing is what turns a stack of documents into a chain. Without it you have a stack. A stack is what most chronic-illness disability claims look like. A chain is what a rare few of them look like. The rare few are the ones that win.
RULE TWO — THE CHAIN IS BUILT IN LAYERS, NOT IN ONE SITTING
Do not try to build the whole chain in a weekend. Build it in the order I gave you. The cover page first, in a week. The physician corroboration next, over three to six weeks depending on your physicians' calendars. The functional impact statement over another week. The treatment history over one long focused session, then reviewed after two days. The third-party statements over two to four weeks. The appendices last, once the referenced pieces are stable.
If you try to build the chain in a single push you will get one of two outcomes. Either the disease will not permit you to finish it, and you will submit a partial folder against a deadline you cannot afford to miss. Or you will finish it and it will read as though you built it in a single push, which claims examiners can also sense. A folder that has been assembled slowly, with the specific patina of documents that have gone through revisions and reviews across several weeks, looks different from one that has been thrown together. Give yours the patina. It will read as evidence.
RULE THREE — LAWYER YOUR APPEAL, NOT NECESSARILY YOUR INITIAL FILING
There is a category of chronic-illness patient — often one who has been ground down enough by the Convincing Cost that she believes she cannot advocate for herself — who hires a disability attorney to file her initial claim. I do not want to tell you not to do this. A good disability attorney can be worth every penny, especially in complex cases involving comorbid conditions, prior denials, or federal employment.
What I do want to tell you is this: the chain you have built in the last several chapters is often sufficient, in a first-round adjudication, without an attorney. Attorneys become more valuable at the appeal stage, when the specific arguments turn on legal interpretation of the rating criteria rather than on the underlying evidence. If cash is tight, submit the initial claim yourself with a well-built chain. If it is denied, hire the attorney to build the appeal. You will have preserved thousands of dollars of attorney time for the phase of the process where it matters most, and you will have submitted an initial folder that is often stronger than the folder an attorney would have submitted for you.
That is not universally true. It is not legal advice. It is a practical observation from the specific position of a chronically ill patient watching other chronically ill patients spend money they did not have on attorneys who added little to what a good chain could have delivered. Read that paragraph twice if you are considering hiring.
§
A NOTE FOR NON-MIGRAINEURS
The Disability Evidence Chain works, in structure, for every chronic invisible illness that meets an American disability adjudicator's desk. The six links are the six links. What changes is what fills them.
For ME/CFS and Long COVID, the treatment history link is often the most important of the six, because the disease's underlying illegibility to the medical system means that the record of failed conventional interventions is itself part of what establishes the seriousness of the case. For fibromyalgia, the third-party statements often carry disproportionate weight, because the invisibility of the symptoms means witnesses are one of the few sources of collateral corroboration. For endometriosis, the pattern visual on the cover page benefits enormously from a cycle overlay that shows the temporal linkage between hormonal events and disabling pain — a linkage that adjudicators may not otherwise recognize.
For all of them, the discipline is the same: build the chain, link it, layer it, deliver it cleanly. Do not try to move the adjudicator. Try only to give her the file she needs to reach the decision the law and her workload allow her to reach. That is the entire task. Do it well, and the file will do the rest.
§
The parallel pipe is now built.
I have taught you the Seven Dimensions. I have taught you the Leaky Boat. I have taught you the Threshold Model. I have taught you the Parallel Pipe itself and its three cadences. I have taught you the Clinician One-Page. I have taught you the Disability Evidence Chain. Six joints. All held together. All sustainable, at the level of a chronically ill life, if you do the work in the small, honest, twenty-minute-timer sessions I have described across the last several chapters.
The next chapter is not another framework. There are no more frameworks. The next chapter is about what happens after you have built the instrument — about the specific and unexpected shape of a chronic illness lived from inside the discipline of the pipe. I want to talk to you about what changes when you become an evidence-generating patient. I want to talk to you about what does not change. I want to talk to you about the specific psychology of running the instrument for a second year, and a third, and every year after that. Because building the pipe is the first year of your work. Living inside it is the rest of your life. Both are the subject of this book. The second one is what the next chapter is about.
 
Chapter Ten
Living With the Instrument
I want to begin this chapter by telling you what will not happen once you have built the parallel pipe.
Your disease will not go away. I said in the Preface that there is no cure, and I have said it in different forms in every chapter since, and I want to say it one more time at the point in the book where the pedagogical arc concludes and the lived arc begins. A patient who has finished reading Chapter Nine and who has spent the previous six months honestly building the six joints of the pipe still has migraine disease. She still has attacks. She still has bad weeks. She still has the specific fatigue of postdromal fog and the specific grief of a canceled evening with her children. The instrument does not change any of that. Nothing was going to change any of that. Anyone who has told you that a good enough tool, a good enough diet, a good enough neurologist, a good enough attitude was going to change any of that has been selling you something, and I told you at the beginning of this book that I refuse to be one more voice on that pile. The disease continues. What changes is your relationship to it. That is worth almost as much as a cure would be, and it is the only thing on offer at the price the world can actually charge.
I want to tell you what changes.
§
The first thing that changes — the one that surprised me the most, and the one I hear most often from other patients who have built the pipe — is that you stop trying to remember your disease.
I do not mean that you forget it. You do not. The disease will not permit forgetting. What you stop doing is the specific, exhausting, low-grade mental labor of trying to hold your own case in your own head — the constant background computation of "when was the last bad one, how many have I had this month, is this getting worse, is that new medication working, did I take my preventive at the right time, when did I last see the neurologist, what did she say" — that every chronically ill patient without an instrument runs, without noticing, as the operating system underneath her ordinary life. That computation is one of the largest hidden costs of chronic invisible illness. Nobody talks about it, because nobody knows to. You do not know you have been running it until you stop.
You stop when the instrument stops requiring it. The log has been keeping the record for you. The pipe has been running behind your ordinary life. When you need to know "when was the last bad one," you open the log and it tells you in three seconds. When you need to know "is this getting worse," you look at the pattern visual and you can see the trajectory. When you need to know "is the new medication working," the medication response matrix is either building the case for it or against it. You do not have to remember. You have to log. Those are different demands on a compromised nervous system, and the second one is dramatically less expensive than the first.
I did not know, until my instrument had been running for about eighteen months, that a specific low frequency of my cognitive load had been dedicated to this holding function since the disease had started. When I noticed it was gone, I noticed it the way you notice a background hum in a house going silent when the compressor finally cuts out. You did not know the hum was there. You feel the silence.
That silence is one of the specific things a well-built pipe gives you. It is not the biggest thing. It is the first thing. I mention it here because most patients do not know to expect it, and if you do not know to expect it, you will not recognize it when it arrives, and it is worth recognizing. It is one of the small quiet miracles of a chronically ill life managed with an instrument. Enjoy it when you feel it. You have earned it.
§
The second thing that changes is how you talk about your disease with the people who love you.
I want to tell you a small scene from my own life to explain what I mean. It is a scene that happens now, roughly once a month, with Christa, and that used to happen not at all before the pipe. She will notice — because she is an athletic trainer by profession and pays specific attention to bodies in ways most spouses do not — that I have been holding something for a day or two. A fatigue in the eyes. A slower pace at the sink. A shorter answer than the moment called for. She will ask about it. In the years before the instrument, my honest answer would have been I don't know. Not because I was withholding. Because I genuinely did not know. The disease is a fog for the person inside it, and the specific question "how are you doing this week" is one of the questions the fog is worst at answering.
Now my honest answer is that I open the log and I look.
I say something like: "My capacity ratings have been about a five for four days in a row. Sleep the last three nights has been about six and a quarter. There's a barometric drop moving in this evening. My heart rate variability was depressed this morning. I haven't had an attack, but I'm sitting closer to threshold than usual, and I've been protecting my afternoon accordingly."
That sentence is not a symptom report. It is a status report. Christa can hear it, and she can respond to it, in a way that the old I don't know did not permit her to respond to anything. She can say "do you want me to take the eleven-year-old to therapy this evening so you can rest before dinner?" She can say "the walk we were going to do tomorrow morning is going to be a walk we do on Saturday instead." She can say, and she does say, "I'm glad you have that." The instrument has given us a shared vocabulary about my body that the disease had, for years, made impossible. That shared vocabulary is one of the reasons our marriage works the way it works. I want to be careful about how much I claim for it, because Christa and I have a lot of reasons our marriage works and only one of them is the log. But the log is one of them. I would not want to be married while chronically ill without one.
I think most spouses of chronically ill patients would say something similar if asked. The disease is hard on marriages. A shared vocabulary makes it less hard. The pipe produces the vocabulary as a byproduct of doing what it was originally built to do. This is the specific class of unexpected dividend that comes from building an evidence architecture and then living inside it. You get things you did not build the architecture to get. This is one of them.
§
The third thing that changes is your relationship with your care team.
I want to give you the specific shift, because it is unlike anything I encountered in a decade of ordinary neurology visits before the pipe. Physicians treat you differently once you have started handing them well-built artifacts. It is not that they respect you more, exactly. It is that they can see you more. When you walk in with a Clinician One-Page, you are, in the language of medical culture, a different kind of patient than the patient you were the previous visit. You are an evidence-generating patient. That category is rare enough that most physicians have never before encountered it in their practice, and it changes what they can do for you inside the twelve minutes they have.
I want to describe two specific ways this has manifested in my own care, because they are the ways it will most likely manifest in yours.
The first is that you get more time.
Not always. Not on every visit. But often. When a physician sees, in the ninety seconds she has read your one-page, that the patient in front of her is bringing organized signal to a diagnostic problem that has historically been solvable only through prolonged and unreliable conversation, she will sometimes — quietly, without announcing it, without necessarily even realizing she is doing it — extend the appointment. She will spend twenty minutes with you instead of twelve. She will follow a thread her curiosity wants to follow instead of the thread her schedule requires her to follow. She will do this because you have given her something worth spending her time on. I do not want to overclaim this. It does not happen every visit. When it happens, it is a gift. When it happens repeatedly, it is a relationship. You will have built one with your care team that most chronically ill patients never get to build.
The second is that you get taken seriously as a partner in your own care.
Every physician who has treated a chronic disease patient has, at some point, felt the specific frustration of trying to make a therapeutic decision without the data she needs to make it well. She has been trained to make those decisions anyway. She has been trained to make her best guess, prescribe the plausible next intervention, and iterate. When you hand her a one-page and a chain, you are offering her the option of a completely different mode of collaboration. You are saying, quietly: I have been doing the diagnostic work that this system does not have time to do for me. I have brought you the results. Would you like to make a decision with me from inside a real dataset, rather than for me from outside one?
I do not know a physician who has said no to that offer, when it is made in good faith with a well-built one-page in the room. It changes the appointment. It changes the relationship. Over eight or ten visits, it changes what your care becomes. It becomes, quietly, the kind of care most patients dream about and almost no chronically ill patient ever gets: a partnership between an evidence-generating patient and a good physician doing her best inside a broken system. You get to that partnership through the instrument. There is no other route to it that I have ever seen work. If someone tells you there is one, I would ask her to show you her records.
§
The fourth thing that changes is that you start to see other patients differently.
This is the change I did not expect and that has, in the eighteen months since I began sharing the framework publicly, become the most important of all of them. You start to notice, when you talk to another chronically ill person, whether or not she has an instrument. You can tell within two minutes of a conversation. A patient with a pipe talks about her disease with the specific texture of somebody who has been generating her own evidence. She says things like "my attack rate this quarter has been higher than my rolling twelve-month average" instead of "I've been bad lately." She says "the interaction between sleep and screen work is the driver" instead of "I don't know what's causing it." She says "my neurologist and I are working on the preventive strategy" instead of "the doctor won't help me." She says these things not because she is more articulate than the patient without a pipe. She says them because she has organized her signal in a way that permits her to say them.
The patient without a pipe often knows her body just as well as the patient with one. She has been paying the Convincing Cost for years and has been holding an interior sense of her disease that is often astonishingly accurate. What she lacks is the vocabulary and the artifacts to translate her interior sense into a form the world can read. The pipe is that translation.
When you have built one, you become — sometimes without meaning to, sometimes very much meaning to — a person other chronically ill people ask questions of. A neighbor whose daughter has Long COVID. A colleague whose sister has fibromyalgia. A woman at your church whose migraines are just beginning to look like the trajectory yours had ten years ago. They ask, in various forms, the same question: how did you build what you built? I want to tell you what to say when they ask, because it is one of the specific responsibilities that comes with having built the instrument, and I do not want you to be unprepared for it.
Say: I built it slowly. I built it in twenty-minute sessions. I built it during the collapse of a career, and I built it because the alternative was to have my disease continue to be invisible. It is a set of six joints. I can show you the first one in fifteen minutes. I can show you all six of them over the course of a few conversations. I do not have any secret knowledge and I am not going to charge you for it. If you want to build one for yourself, sit down with me on a Saturday morning and we will start on the log.
That answer is the answer you give. It is not evangelism. It is not marketing. It is a specific act of solidarity with somebody who is standing where you were standing before you began to build. When you help her begin to build, you extend the reach of the instrument beyond your own life, and you do it in the specific way the instrument was always meant to be extended: hand to hand, at a kitchen table, in the small unglamorous minutes of an ordinary Saturday. That is how the discipline is actually going to reach the forty million migraine patients and the hundred and fifty million chronically ill patients in this country. Not through my book alone. Not through my company alone. Through you, and through the person you next help build, and through the person she next helps.
§
I want to close this chapter by telling you what does not change.
You will still have bad days. You will still have weeks when the log does not get updated. You will still have appointments that do not go the way you had hoped. You will still have claims that are denied on first review. You will still have friends and family members who, no matter how much data you give them, will not fully understand what you are living with. You will still have moments when the disease wins the specific engagement and you spend a night on the bathroom floor with the tile against your forehead. I have. Christa has watched me have that night more than once since the instrument has been running. The instrument does not prevent the night. The instrument only prevents the specific compounding of the night — the specific way that a bad night, undocumented and unsituated in the trajectory of a case, used to feel like a random blow that might be the first of many. With the instrument, the night is still the night. But it is a night whose context you can see. It is a data point. It gets logged in the morning, or the next morning, and it takes its place inside a record that already contains hundreds of nights like it, and you can look at the record and know that the shape of your disease continues, and that this night is not the beginning of anything you have not already seen the beginning of before.
That is the specific consolation the instrument provides. It is a small consolation. It is enough.
§
A NOTE FOR NON-MIGRAINEURS
Everything in this chapter — the silence where the hum used to be, the shared vocabulary with the people who love you, the shift in your care team, the recognition of other patients, the small consolation on the bad night — is the specific texture of a chronically ill life managed with an instrument. The texture is the same across invisible illnesses. I have talked to enough patients with ME/CFS, Long COVID, fibromyalgia, endometriosis, POTS, and interstitial cystitis to know that the specific shape of what changes when the pipe is running is remarkably consistent across conditions. The disease you have is not the disease I have. The life you are building around it, if you build the pipe, is going to look a great deal like the life I have built around mine. Different disease. Same architecture. Same consolations. Same responsibilities to the next patient behind you.
§
The next chapter is about the specific things the tool does not do — the limits of the instrument, the places where I have watched it fail me, the honest ceiling of what an evidence architecture can accomplish for a chronically ill life. I would not be being straight with you if I did not spend a chapter on that. The instrument is powerful. It is also finite. You should know both.
 
Chapter Eleven
What the Instrument Cannot Do
The instrument does not cure the disease. I have said that in the Preface and in Chapter Ten and in three or four places in between, and I want to close the last unfinished piece of the honest arithmetic of this book by saying it once more with more specificity than I have said it before.
The instrument does not cure the disease. The instrument does not stop the attacks. The instrument does not raise the ceiling higher than the physiology of your particular nervous system will permit it to be raised. The instrument does not make good weeks always follow bad weeks. The instrument does not, and this is the piece I want to spend the majority of this chapter on, protect you from the specific ways that the world outside your body will still, even with a perfect log and a perfect Clinician One-Page and a perfect Disability Evidence Chain, refuse to see you at the moments when being seen would most change your life.
I want to tell you a story from my own case, because it is the story I have been holding for four chapters now with a promise that I would tell it later. Later is now.
§
In the fall of 2024, after the Notice of Proposed Removal had arrived and before the removal itself had been finalized, I was interviewing for a job.
I want to be careful about how I describe the job, both because I do not want to embarrass the organization involved and because the specific details are not the point of the story. What I can tell you is this. The role was a Deputy Chief Data Officer position inside a federal agency other than Treasury. It was a significant step up from the role I was in the process of losing. It carried an additional eighty thousand dollars a year in salary. The organization was looking for someone to bring business innovation to a data function that had been operating in a fairly traditional posture for several years. My résumé — the Marine Corps years, the Naval Postgraduate School operations research training, the twelve years at Treasury culminating in the CDO role — was, in the specific way federal executive hiring works, an unusually strong fit for what they said they wanted. They liked me. I liked them. We had three rounds of interviews across about six weeks. At the end of the third round they made me a verbal offer and told me a formal offer was going to be extended within about two weeks.
I want to tell you what I did next, because it is the specific mistake this chapter is about.
I told them I had been sick.
I did not tell them what kind of sick. I did not tell them that I was on medical leave. I did not tell them that a Notice of Proposed Removal was sitting on my kitchen table with a thirty-day response window I had already elected not to use. I told them, in the language of a chronically ill person who has spent the last year and a half being careful about how much she discloses, that I had been dealing with a health issue, that I was working to return to full capacity, and that I did not anticipate it being an obstacle to the role.
I did not lie to them. I want to say that clearly. Every sentence I said to them was, in the technical sense, true. I had been dealing with a health issue. I was working to return to full capacity. I did not anticipate — this is the piece that in retrospect was the mistake I most wish I could take back — that it would be an obstacle to the role. I had, by that point, been the CDO of the United States Department of the Treasury for four years. I had done the job at a level that had earned me the loyalty of a strong deputy and the respect of a Cabinet-appointed rating official. I did not fully accept, in the fall of 2024, that the disease had by then progressed past the point where I could reliably perform work of that level in an ordinary work-week. I was, in the specific denial that I described earlier in this book, still telling myself that the tool was going to bring me back. I was going to be an evidence-generating patient who returned to executive federal service inside of a year. That was the story I told myself. That is a piece of the story I told them.
Some period of weeks passed. There was a background check involved. I was, in the meantime, notified of my removal from Treasury. I did not tell the new organization about the removal because I did not know how to. I did not know whether the removal would show up in the background check. I did not know whether the fact that I had disclosed being sick without disclosing that I had been separated for medical inability would be read, by the specific people who were about to read it, as an act of concealment or as a reasonable act of privacy. I did not know what to do. So I did nothing, which is what a chronically ill person often does at the moment she is most in need of doing something.
Communication from the organization slowed. Then it stopped. I called. I emailed. I did not receive a response. I understood, in the specific way a former Chief Data Officer understands institutional silences, that something had shifted on their end. I did not know what it was.
Some further period of weeks passed. A new presidential administration was preparing to take office. Federal hiring freezes were being telegraphed publicly in advance of the transition. Eventually the organization got back in touch with me. They cited the incoming hiring freeze as the reason they could not, at this time, extend the offer. They were regretful. They wished me well. The letter was procedurally correct in the same way the Treasury letter had been procedurally correct. There was no villain in the room. There was, however, no job.
I want to tell you what I believe was actually happening, without accusing anyone of anything, because I think the honest version of the story is more useful to you than a version in which I claim to have been discriminated against. I believe the hiring freeze was the true policy reason they could no longer bring me on. I also believe — and this is the honest part I owe you — that they had, at some point in the preceding weeks, come to know that I had been terminated for medical inability to perform, and that this piece of information had shifted their internal disposition toward the offer. I do not believe they thought I was well enough to do the job. I do not know that. I have no evidence of it that would sustain a claim. I believe it because I know how those decisions get made inside federal executive hiring, and because I know what it looks like when an organization's tone toward a candidate cools without a stated reason. The hiring freeze became the available cover. It was a real reason. It was, in my belief, not the whole reason.
I lost eighty thousand dollars a year for the foreseeable future.
I lost a specific role I had wanted, in a specific organization I had been excited to work with.
I lost, more importantly, the specific narrative I had been telling myself throughout the fall — that the disease was going to end my job at Treasury but was not going to end my career. Because in the specific way that these things add up, the loss of the new offer was the moment my career, as a career, actually ended. Treasury had been the collapse. The new offer had been the recovery. The rescission was the recovery collapsing. I want to give you that language plainly, because I did not have it in the moment and it took me months to find it. When the collapse and the recovery both collapse, what you have is not a setback. What you have is a life that is going to have to be rebuilt on different foundations from the ones you have been rebuilding on for the last several months.
The log had not prevented any of that.
The Clinician One-Page had not prevented any of that.
The Disability Evidence Chain had not prevented any of that.
The instrument had done what the instrument was designed to do — it had preserved my signal, it had given my care team something to work with, it had positioned me to file a defensible disability claim. What it had not done, and could not have done, was change the specific ways that the world treats a fifty-year-old federal executive who has been separated for medical inability to perform. The world treats him with a specific, procedurally correct, quietly closing coolness. The instrument does not fix that. Nothing fixes that. The instrument only lets you see it clearly enough that you do not blame yourself for it.
§
I want to tell you what I did with the loss of the offer, because it is the specific piece of practical spiritual craft I have to teach you and that I do not think anyone else will teach you as bluntly as I am about to.
I did not respond to it well in the first month.
I want to say that plainly. I was angry. I was angry at the organization. I was angry at the specific person who had made the offer and then who had, in my imagination if not in fact, quietly worked to withdraw it. I was angry at the incoming administration. I was angry at the disease. I was, in the specific and quiet way that people of faith sometimes are, briefly angry at God. I want to be honest about that because I do not want to give you a sanitized version of a lost job that I have processed into a lesson. It was not a lesson in the first month. It was a wound.
Christa was patient. She let me have the wound. She did not try to move me off it prematurely, and she did not try to make it about her, and she did not offer me the specific brand of reassurance that would have made the wound smaller without helping it heal. She let me sit with it. She did the domestic labor that a household with five children requires while I sat with it, and she did not make me feel guilty for sitting with it. I do not know how to thank her for that on the page adequately, so I am simply going to say that she is one of the reasons the instrument is not the only reason my chronically ill life is bearable. She is another reason. A larger one.
After about a month I began, quietly, to do the specific work that a chronically ill patient of faith does with a loss of this magnitude, which is to hold it up and ask what it was for.
I do not mean I looked for a silver lining. I do not mean I convinced myself that everything happens for a reason. I mean I asked, in the plainest and most stubborn way I know how to ask, whether there was a use I could put the loss to that I could not have put the offer to. It is not the same question as "why did this happen to me." It is a better question. It is a question a person with an instrument in her hands can ask, because the instrument teaches you, over enough years of running it, that events are inputs and that what you do with an input is often the only thing about the input you can control.
The use I put the loss to was this book.
I want you to hear that not as a triumphant sentence but as a factual one. If the offer had come through in the winter of 2025, I would have gone back into federal service inside of a Deputy CDO role, and the disease would have chewed on me at that level for six months or a year until it had removed me from that role too, and the book you are reading would not exist, and the tool I built for myself would not be being distributed to any patient anywhere. The loss of the offer was, in the specific arithmetic of the year 2025, the event that forced me to stop trying to return to a career the disease had already ended and start building the vessel that could carry the work the disease had made possible. I did not know that in the first month. I was too angry to know it. I have known it, more and more clearly, in every month since.
I am not saying the disease is a gift. I am not saying the loss of the offer was a gift. I do not believe in gifts of that kind, and I would find it insulting if somebody offered me one. What I am saying is that when the instrument has taught you to see your own body as a source of signal rather than as a source of shame, it teaches you, by extension, to see the events of your life as sources of signal too. Losses are signal. The signal from the loss of the offer was that the world was closing a door I had spent two years trying to keep open, and that another door was going to have to be walked through, and that the walking of it was going to have to be my responsibility now. That is not a lesson. That is a piece of information. It is what the instrument gives you.
§
I want to close this chapter by telling you the other things the instrument does not do, because the offer story is the biggest one but it is not the only one.
The instrument does not make your friends understand.
Some of them will. A few will read what you are doing and want to know more and will, in the small and specific ways that friendship works, sit with you at a kitchen table on a Saturday morning and ask you to show them the log. Most will not. Most of your friends love you and will be, throughout your chronic illness, exactly as attentive to the specifics of it as they would be if they had a friend with any other chronic condition, which is to say attentive in a general way and not attentive in a specific way. This is not a failure of your friends. It is the specific truth of being an ordinary human being with an ordinary bandwidth for other people's problems. The instrument does not increase your friends' bandwidth. It only increases the fidelity of the small share of their bandwidth you already had. Do not resent them. They are doing their best.
The instrument does not make your workplace safer.
If you are still working, or if you find yourself working again in a different context, you will discover that the specific institutional response to a well-organized chronically ill employee is not always the response that a well-organized employee deserves. Some employers will treat your data as an asset. Some will treat it as a threat. A few will treat it as both simultaneously and will handle you with a specific administrative caution that costs you promotions, roles, and opportunities in ways you will find difficult to name. The instrument does not prevent this. It only lets you document it, and it lets you decide, at each juncture, whether to fight or to let the moment pass.
The instrument does not remove the loneliness.
Chronic invisible illness is lonely in ways that no amount of documentation can fully address. You will still have three-in-the-morning stretches when nobody in the world knows what your body is doing to you and when the fact of nobody knowing is a wound of its own. The log does not sit with you at three in the morning. Christa sometimes does. God sometimes does. The log is on the laptop, in the other room, closed. The instrument helps in the daylight. It does not follow you into the night.
I want to tell you those things not to darken this chapter but to keep you from being surprised by them. A book about the instrument that pretended the instrument was more than it is would be a worse book than one that told you honestly what its limits are. I have tried to write the honest version. I think it is a better book for it. I think you will make better use of the instrument for it. You will know, going in, what to expect and what not to. That is worth as much as the tool itself. Maybe more.
The instrument is finite. So are you. So am I. We work with it anyway. That is the whole of it. The next chapter is about what happens when you decide to give the instrument away.
 
Chapter Twelve
What You Owe
I want to begin this chapter with a sentence that is going to sound like a departure but is not.
Once you have built the instrument, you owe it.
I do not mean you owe it to me. I do not mean you owe it to Precipice Health, the small business-of-one I have quietly stood up around the frameworks in this book, and which is still — as I write this in the winter of 2026 — a company that has not yet released its first product. I do not mean you owe it to any organization at all. I mean that once you have built an instrument that lets you see your own body more honestly than the American healthcare system has ever let anyone see hers, you are, whether you asked to be or not, a person with a responsibility. The specific responsibility is to help the next patient behind you begin to build one for herself. That is what this chapter is about.
I want to be careful about how I frame the responsibility, because I do not want it to feel like a debt. It is not a debt. Nobody handed you the instrument. You built it in the collapse of a life the disease was taking from you, in twenty-minute sessions at a kitchen table, with a discipline that cost you thousands of hours of energy you could barely spare. You do not owe your instrument the way a person who receives charity owes her benefactor. You owe it the way somebody who has learned a rare and useful thing owes it — not to any particular person, but to the shape of the world. A woman who has learned to sail owes the next sailor a few sentences at the dock. A man who has learned to garden owes the next gardener a look at his tomato plants and a bag of the seeds that worked. Those are not debts of obligation. They are debts of practice. You have practiced something rare. There is a way of paying that forward that does not require you to become an evangelist. I want to describe it, because it is the way I have come to think about it and because I hope it will make you less afraid of the responsibility once you understand its actual shape.
§
The shape of the responsibility is small, local, and specific.
I want you to picture a Saturday morning in your kitchen. Not a marketing event. Not a webinar. Not a nonprofit board meeting. A Saturday morning, at your kitchen table, with a coffee or a tea or a small water in a glass, and one other chronically ill person who has come to sit with you for an hour.
That is the meeting. That is the whole meeting.
You do not need slides. You do not need a copy of this book, though if she has read it that will save you both some time. You need the log open on your laptop or your tablet, and you need forty-five minutes to walk her through the first two joints of the pipe. The Seven Dimensions. The Leaky Boat. That is what fits into forty-five minutes. The rest — the Threshold Model, the log discipline, the Clinician One-Page, the Disability Evidence Chain — she can come back for on subsequent Saturdays if she wants to, and she will, because the first two joints will produce enough insight in her own case within thirty days that she will be ready for the next two, and the two after that.
The forty-five minutes is what you owe.
The subsequent Saturdays are what you offer if she comes back.
That is the whole responsibility. You are not on the hook for her disease. You are not on the hook for her disability claim. You are not on the hook for whether she builds the pipe or lets it lapse or comes back six months later having lost the log to a hard drive failure. You are on the hook for the forty-five minutes and the honest teaching of two joints. Once she has those, she has enough to build the rest herself if she chooses to, and she has enough to know she is not alone if she does not.
I mention this because chronically ill patients, in my experience, often decline to help other chronically ill patients out of a fear that helping will require more than they have. It will not. The specific shape of the help that is actually useful is a shape that fits inside the capacity of a person with an instrument. It is a chapter of one of my books, delivered across a kitchen table, in forty-five minutes. You can do that. You have the capacity for that. You have already done things harder than that in the last two years.
§
I want to tell you what I have learned about giving the instrument away, because I have been at it for about a year and a half now — since the winter of 2024, when I first began to sit down with other chronic migraine patients at various kitchen tables — and there are three things I did not expect that I want to warn you about.
The first is that the person on the other side of the table is often not ready.
She has picked up the phone. She has arranged the Saturday. She has driven to your house or your favorite coffee shop or wherever you are meeting. She wants what you have. She may not be ready to build it. What that looks like, in the room, is a specific kind of resistance to the discipline of the first joint. She will ask if she can just download an app. She will ask if there is a shortcut. She will explain, warmly, that she is not the kind of person who keeps a log. She will tell you about her friend who tracked her migraines for six months and it did not help. She will, in the specific and honest way that a chronically ill person exhausted by her disease sometimes talks, be looking for you to hand her something that requires less of her than the instrument actually requires.
Do not hand her a shortcut. There is not one. Hand her the honest first joint. Show her the Seven Dimensions. Tell her it takes five minutes per attack and that the discipline is the discipline and that the reason her friend's log did not work is that her friend had thirty-seven columns and no framework and gave up. Then let her decide. Some patients will decide to build. Some will not. The ones who decide not to build are not failures. They are people who, at the moment they sat down at your kitchen table, did not have the capacity for the specific kind of work the instrument requires. Respect that. Do not push. She may come back in six months, when the disease has taken enough more from her that the cost of the discipline finally becomes cheaper than the cost of continuing without one. When she does, you will be ready for her. So will she.
The second is that the person on the other side of the table sometimes changes what the instrument means.
I want to tell you about a woman I sat with about eight months ago whose primary condition was not migraine but a specific form of connective tissue disorder that I did not know very much about before we sat down. She wanted to know if the frameworks generalized. I told her what I have told you in every sister-condition callout in this book — yes, with translation — and I walked her through the first two joints in the language of migraine because that is the language I know. She translated as I went. She was, in fact, a better translator than I would have been if I had tried to teach her framework in her own condition's language. Within about thirty minutes she was showing me how one of her disease's specific hallmarks — joint hypermobility measured across time — would fit into a modified version of the Seven Dimensions. I had not thought of it that way before. She had.
I want to give you the lesson from that Saturday, because it is one of the least expected dividends of giving the instrument away: the people you teach it to will teach it back to you. They will show you uses you had not imagined. They will bend the frameworks in ways your own disease would never have required you to bend them. If you approach the giving with the humility of a person who has built one instrument and is now watching another patient build a related but different one, you will learn from every conversation. You will become, in the specific pastoral sense that this book has occasionally reached for, part of a community of evidence-generating patients. That community does not currently exist at scale. It will exist, in ten or fifteen years, because of Saturdays like the one I just described. You will be one of the reasons.
The third is that the people you least expect will be the ones who use it.
I sat with a nurse once who did not have chronic migraine but who cared for her elderly mother who did. She wanted to build the instrument for her mother. Her mother was seventy-eight years old, cognitively intact but physically frail, and had been getting migraines for forty-five years without ever having been given a language for them. The nurse built the instrument for her mother over about eight weeks. Her mother, at seventy-eight, learned to rate her capacity on a one-to-ten scale each morning. She learned to write the Seven Dimensions in a paper journal that her daughter set up on her nightstand. Within four months her mother's primary care physician had changed her medication regimen based on the data. Within eight months her mother was reporting fewer attacks than she had had in a decade.
The person who used the instrument was not the person I had been teaching. It was her mother.
The specific shape of that dividend — the daughter as the intermediary, the mother as the beneficiary — is not something I would have predicted from inside my own case. I would not have expected a seventy-eight-year-old woman with no prior experience of longitudinal data would find the discipline sustainable. She did. The nurse held the pen for the first eight weeks and then her mother took the pen and kept holding it. I mention this to you because you may be, at some point after you have built the instrument, the intermediary for somebody who cannot build it for herself. A parent. A child. A spouse. A partner. A friend. You will hold the pen for them for a while, and then you will pass it to them, and they will surprise you.
§
I want to say one thing about Precipice Health, the company I have started to build around the frameworks in this book, and then I want to move on.
The company exists because I could not, from a single kitchen table, teach the instrument to every patient in the country who needs it. The Saturday-morning model I described in this chapter is the model I believe in. It does not scale. Some things do not need to scale. Some things need to be replicated, and replication is not the same as scaling. Precipice Health is my attempt to build a small, honest vessel that can help patients begin the discipline in a lower-friction way than I began it, and that can then, at the right moment, hand them off to the small and local and Saturday-shaped support the discipline actually requires. There is a flagship product in development called Migraine Data Detective that will, when it launches later this year, be the first piece of that vessel. There will be others. The Well — a small family calendar application my wife and I built for our own household — is a different product in a different category, and it is being brought to market first, quietly, as a way of learning how to run a small business well before the more consequential health tools follow it.
I mention any of this only because you may wonder, having read the book, whether I have a commercial stake in what I have been teaching you. I do. The stake is small and the company is early and I have chosen to tell you about it in one paragraph in one chapter rather than in a marketing frame around the whole book because that is the honest weight it deserves. The book is the book. The tools I have taught you are yours. If Precipice Health builds something that helps you, use it. If it does not, do not. Nothing I have taught you in the preceding twenty chapters requires the company to exist. The company is downstream of the book. It is not the reason for it.
I want to be very direct about that, because I have read enough business books disguised as memoirs to know how tired the format can make a reader, and I refuse to make you tired. The book is a book. The tool is a tool. The company is a small vessel. All three of those things can exist without collapsing into each other. I am doing my best to keep them separate.
§
What you owe, then, is this.
If you build the instrument, and if the instrument produces for you the kinds of small quiet consolations I described in Chapter Ten, sit down with one other chronically ill person and give her forty-five minutes. Do it once a quarter for the rest of your life. Four Saturdays a year. Sixteen people over four years. In ten years, the small local network that will have grown around the sixty or seventy patients you have personally seated at your kitchen table will be doing more for the movement of evidence architecture into invisible illness than any book or company or professional society will do.
That is the movement.
It is a kitchen-table movement.
It does not require a foundation. It does not require a hashtag. It does not require anyone's permission. It only requires that you, having built the instrument, be willing to sit down with one other patient a quarter and teach her the first two joints.
The next chapter is where I try, briefly and carefully, to say what this book has been about at a scale larger than one disease. I have been holding back on that scale for the whole book. I want to say it once, cleanly, before the book ends. I am going to try.
 
Chapter Thirteen
The Wider Field
I want to try, in this chapter, to say the thing I have been holding back for the whole book.
I have written this book about migraine because migraine is the disease I know from the inside. I have made, in each of the framework chapters, a small note showing how the joint I have just taught you translates to some of the other invisible illnesses that share migraine's specific structural cruelty. Those notes have been careful. They have been short. I have been, throughout, unwilling to claim more than my own experience gives me authority to claim, and I want to be clear before this chapter goes any further that nothing in the next few pages is going to change that. I am not a specialist in myalgic encephalomyelitis. I am not a specialist in Long COVID. I am not a specialist in fibromyalgia or in endometriosis or in postural orthostatic tachycardia or in any of the other conditions on the sister-illness list I gave you in the Preface. I am a chronic migraine patient who, in the collapse of a career, built an evidence architecture for himself that he now believes, on the basis of the patients he has sat down with, generalizes.
I want to say what the generalization actually is, because it is more careful than "this book applies to invisible illness generally" and it is more expansive than "this book is about migraine."
The generalization is that a specific class of chronic invisible illness — the class that shares the structural cruelty of not being able to be seen — has been treated, for a hundred years, as a set of unrelated diseases with unrelated evidentiary problems. Migraine over here. Fibromyalgia over there. Endometriosis somewhere else. Long COVID as its own new category. ME/CFS as its own contested category. POTS in a corner. Each of them has developed its own clinical specialty, its own advocacy community, its own suite of medications, its own set of specialists. Each of them has, from the patient's side, been experienced as a lonely and singular fight against a specific medical system.
The specific medical system is not a specific medical system. It is one medical system. The fight is not a specific fight. It is one fight. The problem is not a specific problem. It is one problem, and the problem is the evidentiary layer. The gap between what a chronically ill person's body is doing and what the medical, insurance, and disability systems are able to read about it is the same gap regardless of the name on the chart. It is the gap between subjective experience — the actual, ordinary, all-day-every-day experience of being alive inside a body that is doing something the world does not know how to measure — and objective evidence. For a hundred years, medicine has been extraordinarily good at closing that gap for diseases that can be imaged, biopsied, cultured, or blood-tested. It has been extraordinarily bad at closing it for diseases that cannot. The diseases that cannot are the invisible illnesses. All of them. There is not a special exception for one of them and a general problem for the others. There is one general problem. I have been calling it the Evidence Funnel, and I have described its shape in Chapter Three, and I have taught you a specific counter-instrument to it over the course of the following six chapters. The instrument works because the problem is one problem.
I am not the first person to notice this. There are researchers and clinicians and patient advocates who have been describing something like this generalization for years, using various frames — the "medically unexplained symptoms" frame, the "central sensitization" frame, the "contested illness" frame, the "biopsychosocial" frame that has done as much harm as good. I am not claiming to have discovered the generalization. I am claiming, in the specific limit of what a patient-built evidence architecture can claim, that the practical response to it — the instrument I have taught you — is one instrument. Not seven. One. What differs across the diseases is the specific content of the columns. What does not differ is the architecture of the log, the discipline of the cadences, the six links of the chain, the twelve-minute problem in the specialist's office, and the forty-to-ninety-minute problem in the adjudicator's folder. The architecture is the same because the underlying evidentiary problem is the same.
§
I want to name the field, because a thing that is named is a thing that can be worked on, and this thing has not until now had a name that stuck.
I have been calling it, for lack of a better term, evidence architecture for invisible illness.
I do not love the term. It is a mouthful. It has the specific dryness of a phrase that a former Chief Data Officer might come up with over coffee, which is because it is a phrase a former Chief Data Officer came up with over coffee. I would welcome a better one, and if this book ever finds a wider audience I hope somebody in that audience will propose a better one and I will happily adopt it. Until then: evidence architecture for invisible illness. It captures three things that I want to keep in front of the reader.
The first is that this is architectural work. It is not clinical work. It is not the work of the doctor. It is not the work of the researcher. It is the work of the person who lives inside the body, building an artifact that translates the body's signal into a form the world can read. Architecture is the right word because the discipline is structural. There is a load-bearing shape to it. There are joints. There are seams. A bad architecture collapses under the weight of a chronic illness. A good one holds.
The second is that this is evidentiary work. It is not about advocacy. It is not about narrative. It is not about awareness. Those are all things and they matter and I do not want to disparage them. They are not what I have been teaching you. What I have been teaching you is how to generate, curate, and deliver evidence — the specific and unglamorous kind of evidence that a physician can use in a twelve-minute appointment, that an adjudicator can use in a forty-minute review, that a supervisor can use in a workplace accommodation conversation, and that a spouse can use in a conversation about whether Wednesday is going to be a driving day or a not-driving day. That is the work. Everything else is downstream of it.
The third is that this is invisible-illness work. It applies to a specific class of conditions. It does not apply, or applies less directly, to visible illnesses — cancers, fractures, infections, cardiovascular events — for which the medical system's ordinary evidentiary apparatus already produces adequate signal. Somebody with a well-imaged tumor does not need to build a parallel pipe. Somebody with fibromyalgia does. This is a book for the second person. The field, if it is going to exist as a field, exists for the second person. That is the specific population that has been quietly waiting for a hundred years for somebody to build them tools that do not require them to first become their own translators.
§
I want to be careful, in what remains of this chapter, about how much I claim for the field.
The field does not exist yet. I have described it as a thing that could exist, and I have named it, and I have taught you a specific instrument that I believe is one of its founding practical artifacts. It is not a discipline. There is no professional society. There is no journal. There is no graduate program that will train a young data professional to spend her career on evidence architecture for invisible illness. If any of those things come into being in the next ten years, they will come into being because thousands of patients, sitting at thousands of kitchen tables, teach the instrument to other patients in the small local Saturday-morning way I described in the last chapter. Movements grow that way. Fields grow that way. A field that grows any other way — by fiat, by grant, by celebrity — is not a field. It is a program. Programs end when their funding ends. Fields, once they have taken root in a population of practitioners, do not end.
I want the field to take root. I want it to take root because, in the specific arithmetic that a former data executive still performs in the back of his head, there are somewhere between one hundred and fifty and two hundred million Americans living with the conditions the field would serve. Most of them are women. Most of them have been disbelieved. Most of them have been trying, for years or decades, to close the evidentiary gap with tools that do not exist. If the field takes root, some of them will build their own instruments, will teach the next patients, will win their disability claims, will get taken seriously by their care teams, will have shared vocabularies with the people who love them, will reach the specific quiet consolations I described in Chapter Ten. That is a lot of people. It is not everyone. I do not want to overclaim. It is enough people that the field is worth being founded.
I am not, in this book, going to try to found the field. I am one man with one disease and one book and one small pre-launch company. What I can do, and what I have tried to do, is provide a beginning artifact. If the beginning artifact is good enough, other people will build the other artifacts the field needs. If it is not good enough, somebody else will write a better version of it, and I will be one of the earlier attempts. Either outcome is fine. The work will get done. I have made peace with not being the person who does all of it, or even most of it. I am the person who wrote this book. The next person is going to write the next book.
§
I want to close this chapter with a specific thought about the wider field, because it has kept me awake in the small hours more than once in the past year and I do not want to leave it out.
The wider field, if it comes into being, is going to have to solve a problem I have not solved in this book.
The problem is what to do about the patients who cannot build the instrument.
I have written this book for a specific reader — a chronically ill person with enough cognitive bandwidth, enough time, enough education, and enough basic access to technology to build a longitudinal log and to sustain the discipline over years. I have known, as I have written it, that this reader is a fraction of the total population of invisible-illness patients. Some patients are too cognitively impaired by their disease to build the instrument themselves. Some are too poor. Some are too old. Some are too young. Some are in caregiving situations that consume every hour of the day. Some are homeless. Some are incarcerated. Some have severe comorbid mental illness that competes for the same cognitive resources the instrument requires. The instrument I have taught you is not a solution for any of these patients. It is a solution for the reader who can build it and who can help the next patient behind her.
The wider field will have to figure out how to serve the patients who cannot build. I do not know what that solution looks like. It probably involves a set of intermediaries — clinicians, social workers, community health workers, adult children caring for elderly parents, patient advocates working inside disability offices — who can hold the pen for a patient who cannot hold it for herself. It probably involves institutional adoption of the framework by primary care practices and by federally qualified health centers and by disability adjudication offices themselves. It probably involves policy work I am not qualified to do.
I am pointing at the gap rather than filling it because that is the responsibility of the first book in a field. You do not fill every gap in the first book. You name the field, you teach one useful instrument, you point at the gaps that remain, and you invite the next people in. I am doing that. The next people will do the next thing. I hope one of you reading this is one of those next people. I hope you are already thinking about the patients who cannot build. I would like to sit down with you at a kitchen table someday and hear what you have been thinking.
The next chapter is the last one before the Epilogue. It is about the title of this book. I have been holding it back for thirteen chapters, and I want to give it to you cleanly.
 
Chapter Fourteen
The Precipice
I have called this book Precipice of Power.
I want to tell you what I mean by that phrase, because I have used it in the title, and I have referred to it in the Preface, and I have not, until now, told you plainly what I mean by it. A book that names its own title in the final chapter is a book that has been earning the title across everything that came before it. I have tried to earn this one. Whether I have earned it is not for me to say. But I owe you the specific meaning I have been holding, and I want to give it to you now.
§
A precipice is not a summit. I want to start there because most books that use the word precipice in the title mean something like the last hard climb before the peak. I do not mean that.
A precipice is a specific kind of edge. It is the place where a piece of solid ground meets a fall. It is not a mountaintop. It is not a triumph. It is not the moment when a struggle ends and the view opens. It is the moment when a person who has been walking, unremarkably, along the ground finds herself at the specific spot where the ground stops and the drop begins. She can look over. She can not look over. She can step back. She can stay. She can, if she chooses to, step off. What the precipice offers her, at that moment, is a choice she could not have made a mile back where the ground was still the ground. A precipice does not confer power. It offers a place from which power becomes available in a way it was not available before.
That is the meaning I have been holding. A chronically ill life is not a story about arriving at a summit. There is no summit. The disease does not stop. The instrument does not cure it. The chain does not always win the claim. The company does not always change the world. The Saturday-morning teachings do not always take root in the person on the other side of the table. A chronically ill life is a story about walking, and walking, and walking, and eventually finding oneself at a specific place where a choice becomes available that was not available before.
The choice is whether to become an evidence-generating patient.
The place at which the choice becomes available is the precipice.
The power that becomes available if you make the choice is the power of the title of this book.
§
I want to describe what the power actually is, because I do not want you to think it is anything mystical or motivational. It is not. It is specific.
The power of being seen — first by yourself, and then, because you have built the record, by the world around you — is what I have been calling the precipice of power throughout this book. It is not a power over the disease. The disease keeps doing what the disease does. It is a power inside the disease. It is the specific ability, developed over eighteen or twenty-four months of the discipline I have taught you, to be a person who knows what her body is doing, who can tell somebody else what her body is doing, and who can hand a specific artifact to a specific professional and have the artifact do the work of translation that language cannot do.
It is a very small power, in one sense. It does not change the weather. It does not change your neurology. It does not change the specific set of medications that will or will not work in your particular case. It changes only what happens between the body and the world. That is not everything. It is enough. When enough of a chronically ill patient's life happens in the space between the body and the world, changing what happens in that space is worth a great deal.
I want to tell you the specific moment I understood this in my own case, because it is the moment I want to leave you with before the Epilogue.
§
I was in a neurologist's office in the spring of 2025.
It was the third appointment I had had since I had started using the Clinician One-Page. I had, by that visit, become something the American healthcare system does not know very well what to do with — a patient who arrived with organized evidence, who did not require the physician to reconstruct her case through conversation, and who was operating from a specific hypothesis about what the next intervention should be. The physician read the page. She looked up. She asked me a question I have not been asked by a physician before or since.
She said, "Roger, do you know how few patients ever get to this point?"
I said I did not.
She said, "I want to tell you. In twenty-two years of practice, I have had maybe three."
I sat with that for a moment. I did not know what to say to it. It was not a moment I had prepared for. The Clinician One-Page had been designed to let her do her job well inside the twelve minutes she had. It had not been designed to produce a sentence like the one she had just said.
She said one more thing. She said, "The reason I am telling you this is that I want you to understand what you have built. I have patients who have been sicker than you are, for longer than you have been sick, and who have not gotten here. I have patients who are much better off than you and who are also not here. This is not about how bad your disease is. This is about what you did with it. You should know that. Nobody else in your life will tell you. I am telling you."
That is what she said. I am telling you what she said because I want you to know, before this book ends, that the specific moment I understood what precipice of power meant was a moment when a physician who had never before in twenty-two years of practice seen a patient reach the specific place I had reached, told me I had reached it. I did not know I had. I knew I had built a tool. I knew I was using it. I did not know I was standing at the edge of anything. She told me I was. She was not telling me the disease was over. She was telling me the disease had stopped being the only thing about my case that mattered. What now also mattered was what I had done in response to it. What I had done in response to it was to build an instrument that let her — and, on subsequent visits, let other physicians, and the disability adjudicator, and Christa, and my sons, and eventually the readers of this book — see me.
Being seen is the power. Being seen is what a chronically ill patient has been unable to arrange for a hundred years. Being seen is what the instrument arranges, one artifact at a time, one Saturday at a time, one visit at a time. The precipice is the place from which being seen becomes possible. The choice that made me a person standing at that precipice was the choice to build the instrument. You have the same choice. You are, if you have read this book from Chapter One to this chapter, already standing closer to the precipice than you were when you began. Whether you take the last few steps to it is up to you. Whether you look over is up to you. What you do once you have looked over is entirely up to you.
I have not looked over the precipice of power and stepped off it. I have looked over and stepped back, because there is still a lot of walking to do — Christa, and M., and A., and my two older boys, and the book, and the small company, and the next patient I am going to sit with next Saturday morning, and the day after that, and the day after that. A chronically ill life is a lot of walking. The precipice is not the end of it. The precipice is what happens along the way, once, or twice, or on certain rare days when the light is a certain way and you look up from the log and understand, without needing anyone to tell you, that you have become the specific kind of person you needed the world to see you as.
That is the whole meaning I have to give you.
The Epilogue is small. It is about Christa. It is about the ordinary morning that follows the precipice. It is not a dramatic close. A book like this one should not close dramatically. It should close the way a chronically ill day closes: with the light going down, and the tools of the day put away, and the small honest business of tomorrow's breakfast being organized on a corner of the counter. I want you to close this book that way. I have tried to close it that way. I hope I have succeeded.
 

The Morning
I am writing this in the winter of 2026. It is a Tuesday morning in the first week of February. The house is quiet in the specific way it is quiet after the two youngest have left for school and before the workday's first meeting has begun. Christa is standing at the counter with a mug of coffee — she likes coffee, in the way I do not — and she is scrolling through something on her phone. She looks up occasionally and says small things to me. I say small things back.
I do not have a migraine today. I did not have one yesterday. There is a low-grade sensitivity in my left temple that the log will register at about a three when I sit down to do the daily entry, and if the pattern holds it will fade by mid-morning without becoming anything larger. My capacity rating this morning will be a seven. Seven is a good number for me. It used to be a number I only occasionally saw. It is now the average of my last thirty days. I do not attribute that fully to the instrument. I attribute it to the specific combination of the instrument, a preventive that has been working since December, a wife who insists on the walk on Saturday mornings, three children in the house who require a specific rhythm of ordinary attention that turns out to be one of the better things a chronically ill body can be asked to perform, two older sons who have grown into their own lives in ways that make being their father a source of unlooked-for peace, and a God who I believe has kept me alive for reasons that have become clearer to me across the last eighteen months than they were during the fifty years that preceded them.
I want to say the specific thing about Christa that I have been holding since the first chapter.
She is my second wife. I did not write this book, and I would not have written this book, if she were not. I loved my first wife. I raised my two older boys with her. When that marriage ended in 2023, I did not know what came next. I did not know for most of a year. I met Christa in January of 2024, at a point in my life when I did not have the specific reserves that a normal courtship requires, and she loved me anyway. She loved me through the July that I stopped being able to work. She loved me through the November of the Notice of Proposed Removal. She loved me through the December of the removal itself. She loved me through the winter that followed, when I did not know if I would ever work again in the specific way I had been trained to work. She married me in August of 2025 in a small ceremony in a specific place that we chose because it was where we had been ourselves most fully during the year before.
I do not want to sentimentalize her. I have not sentimentalized her in this book, and I do not want to break that discipline in the last page of it. Christa is not a saint. She is not a martyr. She is a working professional, an athletic trainer, an adaptive physical education teacher, a mother of three, a stepmother of two, a Christian woman with a small, specific, unpretentious faith that has held her through her own trials in ways that would fill her own book if she ever chose to write it. She married me knowing what she was marrying. She has, in the eighteen months since, done the small unglamorous daily work of loving a chronically ill man without ever once making me feel that the work was a burden to her. I do not know how to be adequately grateful for that. I have decided, in the specific way a man of faith decides these things, that the appropriate response to a gift you cannot repay is to try to live in a way that honors it. I am trying. Most days I am doing all right. Some days I am doing better than that. A few days, I am not. She lets me have the not-days without making them larger than they need to be. That, more than anything else in the record of my life, is the thing I want you to know about my marriage before this book ends.
§
The two youngest are the ones who will not know, until they read this book someday, how much of it was built for them. M. is eleven. A. is nine. The older girl is fourteen. The older two boys — mine, from before — are twenty-three and twenty. They are their own men now. The three younger ones are still children, in the specific meaning that word carries in a household with kids in it, which is to say that they take up much of the daily surface area of a life without necessarily knowing that they do. I want to say a small thing to them here that I will say to them out loud someday when they are older, but that I want on the record now, because this book is going to outlast me in ways that spoken words cannot.
You did not have to be nice to me while I was sick. You were, mostly, anyway. You did not have to accept me into your life the way you did in the year and a half before Christa and I were married. You did. You did not have to be patient with the specific weeks when I could not be the adult in the room that you needed me to be. You were.
The instrument in this book was built for me. Some of it — I will not tell you which parts until you are older and can read them and decide which parts you want to claim — was built for you. I built it because your mother deserved a husband who could show up for her, and because you deserved a stepfather who could show up for you, and because the disease had been threatening for a long time to take from all of you a version of me that would have shown up if he could have. The instrument has helped me show up. It has not always let me show up. It has let me show up more than I would have shown up without it. I am grateful for what it has given me and I am at peace with what it has not. I hope you read this someday and know that.
To my two older sons: the same, in the specific version that adult sons need to hear from their fathers. I have loved you at every point of the last twenty-five years including the years when the disease was making me hard to love back. I have not always known how to say that during the years the disease was worst. The record is here now. You will read this book someday. You will know, when you read it, exactly what your father was doing during the years he was not always available to you. He was learning to see himself. He was learning to be seen. He was building the tool that let him be a better father to you across whatever remaining years the two of you and I get to share, and it worked, and I am sorry it took the disease to teach me what a well person should have known without one. Forgive me the years it cost. I have forgiven myself. I hope you can.
§
I want to close with a small pastoral thought, because a book about chronic illness that ends without one is a book that has not told its readers everything they need to know about how the writer stayed alive.
I stayed alive because I built the instrument. I stayed alive because Christa is Christa. I stayed alive because my older sons are my older sons and the younger three are the younger three. I stayed alive because a specific handful of friends and colleagues, whose names are not mine to put on this page, kept in touch with me during the worst months in ways that meant more than they know. I stayed alive because I had good physicians who tried their best inside a system that was not built for them to succeed inside of. I stayed alive because a disability process, procedurally correct and administered by people doing their jobs, gave me enough of a landing to catch my footing on.
I also stayed alive because I have a small, specific, unshowy faith that I have not made a large part of this book because I did not want to alienate readers who do not share it, but that I want to name here in a single paragraph before the book ends.
I have prayed a great deal about all of this. About the disease. About the career. About Christa. About the children. About the book itself. About the small company that has grown up around the frameworks and about the patients who I do not yet know but who I hope this book will find their way to. I do not know how much of my life is the result of what I built and how much of it is the result of what has been sustained in me by a God who I have not always been the best about giving credit to. I know only that I am here. I know that a lot of chronically ill people, at fifty-one years old, after two years of the specific losses I have described in this book, would not be. I want to say the thanks that is owed for that. I have owed it for a while. I want it on the record.
Thank you, God, for the record.
Thank you for the disease that taught me to keep it.
Thank you for Christa, and for M. and A. and the fourteen-year-old, and for my two older boys, and for the people who I have loved and who have loved me across the fifty years that preceded this winter.
Thank you for the reader who has walked through this book with me and who is now standing at her own precipice, or approaching it, or beginning to see it from a distance. I hope she finds it. I hope she looks over. I hope she becomes, in whatever way the specific disease she is carrying will permit, an evidence-generating patient in the small local Saturday-morning sense that this book has tried to teach. I hope her care team sees her. I hope her adjudicator sees her. I hope her spouse sees her. I hope she sees herself.
That is the whole prayer. That is the whole book.
Amen.
— R.M.
Winter, 2026