Two days into my second hospital admission in a week, the GI doctors were still pointing me in the direction of a second colonoscopy. At least, that’s what the internal medicine team — hospitalists, they’re often called — was telling me.

“Let me talk to the GI attending,” I told the gang of six as they made their morning rounds. The group included an intern, a resident, a hospitalist Fellow, the attending hospitalist, and two medical students.

Penetrating the hospitalist wall to communicate directly with the specialists who are actually making the key diagnostic and treatment decisions would become a daunting challenge. This was especially true on the “general medicine” floor where I spent my first two hospital stays.

“We’ll try to reach him,” said the team resident. “But don’t worry. We’re in consultation with his Fellow regularly about what should happen.”

“That’s not particularly reassuring,” I said. “I don’t know what’s being lost in translation between you and the GI Fellow and her attending physician.”

In the ER, the GI Fellow’s tunnel vision caused her conclude without any corroborating evidence that my problems required another colonoscopy. I’d encountered such personalities before: she was accustomed success, but rarely faced the kind of tough questioning that I’d posed. (My 30 years as a Kirkland & Ellis litigator had made me a pretty effective interrogator.) In fact, when she couldn’t answer in a consistent way that made sense, she was threatened and became defensive. She wasn’t going to change course. Oh, for the comfortable — and sorely misplaced — certainty of youth.

At that point, I took a page from a former client’s playbook. He was the best negotiator I’d ever met. When he didn’t like the way things were going, he walked away from the table.

“I don’t like this plan,” I said. “The last time I had a colonoscopy here, the preparation process to clear me out caused my hemoglobin to crash and required a blood transfusion. I’m not going to do that again. I’ll sign whatever releases you want me to sign. But I’m not doing it.”

I showed the young resident a printout of my blood levels during my previous admission.

“That drop in hemoglobin was probably a coincidence,” she ventured.

Her reaction was a vivid example of confirmation bias. Facts didn’t matter to her conclusion. “Coincidence” explained away anything at odds with what she wanted to see.

“Are you kidding?” I was incredulous and becoming angry. “When you have to rely on the concept of coincidence to explain away correlations that seem pretty obvious to anyone else, you need rethink what you’re doing.”

I wasn’t finished.

“Let me put it to you this way,” I said pointedly as I framed my closing argument. “Are you willing to bet two units of your own blood that my crash during the last colonoscopy prep was a coincidence? Because I’m not.”

“So here’s what I want you to do,” I continued. “Take a look at the stool sample I left in the bathroom this morning. It’s called a melena — a classic symptom of what usually is an upper GI bleed, not a lower one that a colonoscopy will find. I understand how the hospitalist system is structured. You have your turf to protect. Like most doctors, you think in protocols based on typical cases. I need you to think outside the boxes you’ve created for yourselves. You need to talk to the attending GI — not his Fellow or whoever your contact person is for the GI team on this particular shift. Tell him that I want to talk to him personally so he can come up with a new plan because I’m not having another colonoscopy. We ran the preparation process experiment on me once. I’m not repeating it. Period. Not negotiable.”

A few hours later, I’d heard nothing in response to my demands. I ventured into the hall and, fortuitously, encountered the attending GI physician and his Fellow — the same one who had decided on another colonoscopy when she saw me in the ER.

“I assume you’re on the way to see me soon?” I asked the GI attending.

“No,” he seemed puzzled.

“No one has talked to you about my problems with the current colonoscopy plan?” I wondered.

“No,” he said. “Why?

“Can you walk with me for a minute? I want to show you something.”

We returned to my room where I showed him the stool sample.

“Melena,” he said.

“Right,” I said.

“So tell me how another colonoscopy will identify the source of that bleed?” I asked.

“I can’t,” he replied. “It’s an indicator of a probable upper GI bleed. It’s could also be the result of a lower GI bleed, but that’s less likely.”

I then showed him the time series printout of my blood draws — including the critical crash period during the colonoscopy prep that the hospitalist resident had dismissed as coincidence.

At that point, his Fellow walked in. I invited her to take a look at the stool sample I’d left in the bathroom. She refused. I invited her to look at the printout of blood results during my previous colonoscopy prep. She didn’t want to see that, either.

“I need to rethink this whole thing,” the attending GI physician continued. “This looks more like an upper GI bleeding problem.”

“Precisely,” I said. “Thank you.”

At last, I’d found a doctor willing to reconsider his own assumptions about what was happening to me. It took physical evidence — a bowel movement that I’d saved for hours and my printout of blood results associated with my prior admission — to persuade him. Without that hard evidence to show him, I think that my words would have continued to fall on deaf ears.

About 30 minutes later, the GI attending returned to my room — without his Fellow. He said that he’d canceled the colonoscopy, ordered another endoscopy because the first one might have missed something important (he said there were statistics on that), and wanted to see the results of a capsule endoscopy — a pill that transmitted images to a battery pack that I would wear for eight hours.

“We have to get this going,” he said. “I’ll schedule the endoscopy for the morning and have them place the capsule so that at 5:30 pm, someone can pick up the image pack. We should get those results by Tuesday.”

(Despite his desire for a prompt retrieval of the capsule endoscopy imaging pack, it didn’t get picked up as quickly as he wanted. We’ll come back to that not-so-funny comedy of errors in an upcoming post.)

“Thanks for being willing to reconsider your own conclusions,” I said. “Only a confident professional is willing to do that.”

“Not at all,” he said. “Not at all. It’s good that you acted as you did. And it’s good that you are so informed about your situation. It seems that you have become an expert in this area yourself.”

Whenever we get a new case, that’s what litigators do, I thought.

***

Hospitalists purport to function as a liaison between the patient and the specialists who are really calling the shots. In fact, if you ask the hospitalists whether some procedure will happen — in my case, a second colonoscopy — they’ll tell you that they defer to the specialists.

“Who, exactly, ordered another colonoscopy?” I asked.

“The GI team,” came the response.

“Give me a name,” I persisted.

“It’s a team that includes the GI Fellow who saw you in the ER,” the resident hospitalist replied. “The attending GI rotates. But they all have access to all of the notes that all of us leave in the computer.”

How reassuring — computer notes that someone may or may not read have become the primary conduits for continuity of care.

***

Years ago, primary care physicians (“PCPs”) actually visited patients in the hospital. It was part of the professional service that they rendered. Two doctors coined the term “hospitalist” in a 1996 issue of the New England Journal of Medicine. In some respects, it’s a good idea that, all too often, has gone terribly bad.

According to one of the co-authors of the 1996 article, the hospitalist movement grew from a desire to move away from a procedure-oriented billing and revenue system to one that emphasized coordination of patient care. He described the key challenge to expanding the hospitalist program: “We recognized from the start that hospital support payments were crucial, and would hinge on hospitals’ perception of a positive return-on-investment.”

Lost in the search for this positive investment return is continuity and advocacy that a PCP is uniquely suited to provide on behalf of his or her patient. By 1996, maybe the PCP had already become a marginal player, but once upon a time, having hospital privileges meant something to a patient and his primary doctor. Among other things, it meant that the patient could expect his PCP to quarterback hospital care.

Today, the hospitalist embodies ways in which the core mission of modern medicine has become muddled, if not lost. All too often, they are an obstacle to direct doctor-patient interaction that is essential to effective care. Experts should be basing their decisions on accurate information that they receive directly from the patient, rather than from the medical system’s version of the “telephone game” whereby a patent’s initial reports to a young medical student or intern eventually make their way to the real doctors who matter.

With his or her PCP absent from the hospital picture, the patient alone must do what is necessary to get direct access periodically to the specialists who are making the most important decisions about diagnosis and treatment. That means piercing the hospitalist wall.

I didn’t go into the hospital expecting to fight with doctors. The system left me with no other option.