The Assumptions Doctors Make

I’d just started seeing patients, as an intern at Ben Taub Hospital, in Houston, when the senior resident on my team received notice from the emergency room that a woman needed to be hospitalized overnight, to receive treatment. He wasn’t the type to argue with other doctors over the phone. Still, it was clear by his questions—“So no fever? And she’s totally hemodynamically stable?”—that he wasn’t impressed by what the doctor in the E.R. was telling him. After hanging up, the resident grabbed a colored marker and wrote a name on the whiteboard in our team room, followed by the letters “U.T.I.” He capped the marker with a frustrated look: patients aren’t usually hospitalized for urinary-tract infections.

I hurried down to meet my new patient, armed with “Pocket Medicine”—my Bible, a book summarizing all of internal medicine in microscript. Downstairs, the hospital’s E.R. was in full, frenetic swing—the nurses drawing blood, the techs gathering vital signs, the doctors migrating from stretcher to stretcher in small teams. I followed the numbers posted above each stretcher until I found the one that I was looking for. A young and fit Hispanic man in his thirties sat at its foot, his jeans and T-shirt looking stretched and worn. A woman lay in the bed—my patient. I introduced myself, explaining who I was, before sliding past him to collect her history.

I asked the woman a few questions and got little response. In the way she moved her eyes, I could see why the E.R. doctor had referred her to us. She was in her forties, broad-shouldered and tall, with bushy black hair and wide pupils that drifted from me to a nearby noise or a passing nurse, then back. I wondered if I had the wrong language, so I tried Spanish.

Ella sólo habla inglés,” the man said. She only speaks English.

Now it was his turn to field my questions. Instead of short, inattentive nods, I got descriptions, in Spanish, about how frequently the woman was urinating. It was happening so often that she had to wear diapers.

“Since when?” I asked, looking at her. Once more, her eyes drifted away and then quickly returned before drifting again.

I turned toward the man. “Is everything O.K. with her?” I said.

His look—not toward me, but directed at her—said it all. “This is how she is now,” he said. When they’d married, a few years earlier, she’d been different. They’d lived the routine life of couples: working, eating together, going out occasionally. But, since the mental deterioration started, it had been relentless. The woman was now someone he barely recognized. She looked much older than him, but in fact they were similarly aged.

I asked him more questions, and soon enough I’d built a hypothesis in my mind: this woman had early Alzheimer’s—or Lewy body dementia, or Wilson’s disease, in which the copper we eat isn’t properly processed, leading to buildup in the brain. I flipped to the neuro chapter in “Pocket Medicine.” I decided that I’d organize my patient history, which I’d soon present to my senior resident, around these neurological conditions, rather than the urinary-tract infection.

On the way back upstairs, I happened to run into Robert Graham, one of the most well-known internal-medicine professors at the hospital. Some people called Graham the Cowboy Doctor: on his weeks off, he lived on a ranch, hunting animals and farming tomatoes. He’d once shot and killed a man who had broken into his home, and often shared stories about his boxing days, as a white kid who was often taken in by a Mexican family in Houston’s rough Third Ward.

Graham presided over our daily Morning Report, at which we reviewed interesting medical cases. If he found a case interesting, he jotted notes on his Styrofoam coffee cup. More often—if a vital sign or a finding from a physical exam sounded incorrect to him—he grilled the presenter. If someone cited a journal article, he pried into how the study had been conducted—“They enrolled only five patients?”—and whether we could derive any real meaning from it. Doctors, he believed, must admit when they’re wrong—an insight we’d discussed after he read “The Confessions of St. Augustine.” He argued openly and vigorously with other professors, but always with the same calm, measured demeanor. I imagined that he shot his rifle this way, too.

I told Graham the story I’d just gathered from my new patient and her husband. “Something seems wrong,” I said. I kept on repeating the part that stood out to me most: “She’s still young.” I mentioned the diapers, how her eyes couldn’t keep still. I said that this was the first time anybody had mentioned her cognitive problems—which grabbed Graham’s interest.

“Let’s go,” he said, motioning quickly toward the computer. He logged into the system with two index fingers. I gave him the patient’s name and medical-record number and he began scrolling through the data.

“What’s she in for?” Graham asked.

“U.T.I.,” I said.

He examined the woman’s latest head CT scan. Then he released the mouse and sank back in his chair. “Well, I definitely don’t think she has early-onset Alzheimer’s, or any of that other stuff you mentioned,” he said.

I began to defend my reasoning, though not too aggressively. “But she’s so young,” I said. “She can hardly talk. She can’t control her urine.”

Graham’s blue eyes lasered in on me. “This woman has untreated diabetes,” he said. “It’s caused her to have thousands of small strokes that have killed all these parts of her brain. That’s what’s causing her dementia.”

“I didn’t know it could do that,” I said—a response that revealed my ignorance. I’d never even looked up the symptoms of untreated diabetes; I had assumed that a patient would never grow this sick from a common and manageable disease. “Pocket Medicine,” in its section on “Change in Mental Status,” described only problems such as vitamin deficiencies or opiate intoxication—afflictions that its writers presumably believed were a danger to typical patients. There was no listing for “untreated diabetes.” The book didn’t mention poverty, or a lack of health insurance—conditions that were common among patients at Ben Taub—as a cause of any illnesses. I was starting to realize that its perfect world, in which medicine straightforwardly confronted disease, was a fiction.

As I spent more time in the hospital, a belief began to take shape in my mind: whether someone had insurance could affect their health as much as, or possibly more than, their genetics. I wasn’t the only one wrestling with this idea. In 2017, Annals of Internal Medicine published a review of eleven studies that examined whether people are more likely to die if they lack health insurance. Two of these studies found that Medicaid expansion reduced adult mortality by as much as six per cent. Another paper described a randomized, controlled study of a group of people in Oregon who were given Medicaid through a lottery. Researchers looked at how the winners fared compared with people who remained on the Medicaid wait list. People who received coverage were far more likely to have their diabetes diagnosed and managed.

Diabetes is an expensive illness. Even for people with health insurance, the cost of needles, syringes, extra medical visits, and lost wages can add up to an average of forty-eight hundred dollars a year—around a tenth of an average American salary. The uninsured pay much more for care, and often have less income. One out of every ten Americans is diabetic; in 2017, twelve per cent of all patients who visited the E.R. were diabetics; and nearly eight per cent of the work done in nursing homes involved treating the effects of the disease. My forty-something patient now needed the kind of twenty-four-hour care one gets in a nursing home—her husband acted as her nursing attendant, feeding her, bathing her, making sure she didn’t fall, or burn herself on the stove. Of course, neither she nor her partner could afford it.

Not long after the woman entered the hospital, she left: my supervising resident discharged her, after starting antibiotics for a U.T.I. that we didn’t think existed. (The symptom of urinating on herself was probably attributable to her dementia.) The most valuable thing we did, apart from normalizing her blood sugar and giving her a prescription for insulin, was bringing in a social worker who could help her partner look for community resources to help with her care. I never learned if he was able to access that help, or if she continued her decline. I never saw the woman or her husband again.

The following year, I received a page about a patient whose blood sugar was too high for him to leave the hospital. The E.R. staff had pricked his index finger and fed the test strip into a glucometer. They’d thought the machine might be broken: a normal blood-sugar level is around a hundred, but the machine read “> 499.” Another test confirmed that his blood was nearly saturated with sugar. “He can’t leave like this,” the E.R. doctor told me over the phone. And so I descended the stairs dutifully to the E.R. with a plan already brewing in my mind: I would lower my new patient’s blood sugar enough to get him home.

I was now a senior resident, and Graham was my supervisor. Nothing about the patient’s history or physical exam changed my understanding of the case. After organizing my presentation, I called Graham so that he could sign off on my plan of attack. We met at the patient’s bedside. “Thirty-eight-year-old male, history of uncontrolled diabetes, presents with dizziness and malaise,” I said. I made sure to take into account some of the nuances other residents might have missed—for instance, I noted at what times in the day the man took his already sizable doses of insulin—and concluded with my plan: an even larger dose of insulin. I started to explain the dose I wanted to give.

“Oh,” Graham said, stopping me. “I’m not sure I’d do that.”

He turned to face the patient in the stretcher, who weighed more than four hundred pounds. Now Graham led the conversation. He asked the man some of the questions I’d already asked, like what he ate. But he also asked him when he ate, with whom, and if he ever found himself feeling extremely hungry after injecting insulin.

“I gotta eat,” the man said of how he felt after an injection.

Graham told the man that he was caught in a vicious circle. Higher doses of insulin were pushing sugar into his cells; this had the effect of making him hungry, which caused him to eat, which caused his blood sugar to go up again, which resulted in the need for more insulin, which resulted in more eating, and on and on. The man nodded.

Graham asked him about his activity level, too—not only if he exercised, but also who he walked with, where he would go, if he had a bike at his house or a gym nearby.

“I need to find one,” the man said.

I had planned to keep my patient in the hospital for a few hours, or a day at most—long enough to document a normal glucometer reading. But Graham told us both that we would be keeping him in the hospital for at least three days. “Your diabetes is curable,” he said. “Every hour, I want you to walk around the hospital three times.” He turned to me. “Cut his insulin in half!”

An average diabetic patient on insulin might use ten, twenty, or fifty units of insulin daily. At home, the man took nearly a hundred units each day. I knew that Graham liked making maverick moves; I didn’t argue with him partly because I doubted his plan would work.

“Just remind him to walk,” I told the nurses.

The new dose controlled his blood glucose well, and so our new goal was to wean him fully off of insulin. For two days, the man walked around the hospital every hour, prodded by me, the physical therapists, the nurses, and Graham. He ate only a low-carbohydrate diet. And, for two days, I documented blood-sugar levels that never went above a hundred and ten. My assumptions had been wrong. Insulin wasn’t the solution; what my patient needed was a whole new approach to his diabetes. Before he left the hospital, I told him to keep walking at home. “I can’t wait,” he said. I never saw him in the emergency room again, and I never heard whether he kept up with this new approach to his diabetes. What I knew for sure was that a seemingly insurmountable problem had been at least temporarily solved. Graham had understood the science of diabetes well enough to tailor a plan for this man in particular.

Leave a Reply

Your email address will not be published. Required fields are marked *