Studies comparing angioplasty and thrombolytic drugs have shown a clear advantage for angioplasty if it is performed by an experienced cardiologist in a high-volume catheterization lab within three hours of the onset of symptoms. Death rates after thirty days are lower by almost 50 percent. Also, angioplasty results in an open coronary artery 90 percent of the time, compared with 54 percent for thrombolytics. Moreover, angioplasty drastically reduces bleeding complications, especially in the brain.
But whether these advantages could be maintained in the real world, where delay in getting patients to catheterization labs is inevitable, had been an open question. Then a Danish study published about a year before I started at LIJ notched a major victory for angioplasty. In the study, about fifteen hundred patients who were admitted to community hospitals with acute heart attacks were randomly given immediate thrombolytic treatment or transferred by ambulance to angioplasty centers up to a hundred miles away. Even with the delays, angioplasty resulted in a 40 percent reduction in death, recurrent heart attack, or stroke after thirty days. The data were compelling enough that the study was stopped early by a data-monitoring committee.
In a related study published in The Journal of the American Medical Association, about five hundred patients at eleven community hospitals in Maryland and Massachusetts were randomly assigned to receive thrombolytic therapy or angioplasty. Angioplasty has traditionally been performed only at hospitals with cardiac surgeons on duty, in case there are complications, but these relatively small hospitals had none. In following up, the study found that even without surgical backup, angioplasty reduced the occurrence of heart attacks and strokes in the next six months by almost 40 percent and shortened hospital stays for the original visit by an average of a day and a half.
The net result of these studies was that LIJ was doing more outreach to try to get ambulances to bring patients with acute heart attacks directly to the hospital for angioplasty. At the same time, we were trying to reduce D2B time to less than an hour. But I quickly learned that coordinating ambulance fleets on Long Island is a gargantuan task, in part because a large number are privately owned. However, as my colleagues at hospital meetings pointed out, paramedics had already learned to take trauma and burn victims to specialized hospitals. Heart attack victims deserved no less, and many more lives were at stake.
* * *
My first year as an attending was packed with all sorts of such meetings. Once a month my cardiology colleagues would convene over chicken marsala and baked ziti and talk about the faculty practice plan. Rajiv would often bring up his “Queens strategy”: finding a way to tap into the borough’s large ethnic population—Russian, Greek, South Asian—which suffers disproportionately from diabetes and heart disease. “Out here on Long Island, we practice more preventative care,” he’d say. “Patients are on the right meds. But in Queens it’s a different story.” He’d gripe about the hospital’s refusal to accept Healthfirst and other low-paying insurance plans commonly carried by Queens patients. He’d warn that cardiologists at Jamaica Hospital, a competing facility, were already poaching patients in their emergency room who were slated for transfer to LIJ. He’d decry the clinic that LIJ had proposed to build in Queens because it was within a block of a busy internist, who might feel threatened and start referring his patients elsewhere. “As much as we hate to admit it, patients are a commodity,” he once unashamedly declared when we were brainstorming about how to increase procedural volume (which would help determine his and his interventional colleagues’ salaries).
For my part, I couldn’t help wondering how the hospital was going to handle more patients. Where were we going to put them? The ER was already overflowing, and patients on the wards were sometimes stowed in the corridors. Did we really need to keep putting stents into everyone? But I kept my mouth shut. If I ever had any doubts that medicine is distinct from business, these gatherings absolved me of that notion.
The meetings multiplied, and soon I was attending hospital conferences on quality improvement and resource utilization. Published studies show wide variation in death rates between best and worst hospitals. For example, thirty-day mortality ranges from 11 to 25 percent for patients with myocardial infarction and from 7 to 20 percent for patients with congestive heart failure. The mantra at these meetings was “evidence-based medicine,” which meant looking for reliable metrics, not clinical judgment or experience, to quantify performance and identify best practices. There was a kind of staleness at these meetings, a saccharine, artificial quality that permeated everything from the agenda to the muffins. People would say stuff like “We need a holistic approach that addresses the issues of variability and interdependencies” and “Let’s review our program across the entire spectrum of processes,” or they’d use phrases like “continuum of care” or “integrate and transform.” It was corporate gobbledygook, and it never made much sense.
I discovered a tension between my dual roles as a faculty physician and a hospital employee. As a faculty physician in the Division of Cardiology, I was trying to generate more revenue for my particular section. As a hospital employee, I was trying to keep down overall hospital costs. Nowhere was this conflict more evident than in my charge to reduce the length of stay of patients hospitalized with heart failure. Like all acute care facilities, LIJ received a set payment for each admission based on the patient’s diagnosis. (Such a prospective payment system for hospitals had been in operation since the early 1980s.) The longer a patient stayed in the hospital, using up a greater amount of resources, the more money the hospital stood to lose. Of course, the longer a patient stayed, the greater the likelihood of hospital-acquired infections or harm from tests and procedures, which meant that timely discharge, in most cases, was good for the hospital and patients alike.
But individual doctors, paid separately by insurers for patient visits, had little motivation to discharge patients quickly. As long as their patients were in the hospital, they could bill and be paid for each visit they made. The incentives were misaligned. The hospital was paid a fixed payment. Physicians were paid à la carte. (The incentives might have gotten aligned if hospitals had been allowed to share savings with doctors, but the law at the time prohibited any practice that might influence doctors to provide less care.) Reviewing heart failure cases, both mine and other physicians’, on rounds, I frequently encountered patients getting diagnostic workups or trivial medication adjustments that could have been performed on an outpatient basis. What was keeping these patients in the hospital?
The administration did not think it was because private doctors with admitting privileges were eager to bill more. “I am not jaded enough to think that doctors are going to keep patients in an unsafe environment for an extra seventy or eighty bucks a day,” Bill Remsen, a senior hospital executive, told me when we discussed it at a utilization meeting. “I really don’t believe that is happening.” Instead, he attributed the problem to a fragmented delivery system. “Most hospitals don’t have the number of physician’s assistants and nurse practitioners we do,” he said. “It’s a unique model: more handoffs, more confusion. Who’s taking responsibility for the patient?”
However, some private doctors had a different take. One afternoon, while nursing a cup of coffee at the doctors’ station on ward 7-South, I found myself discussing my administrative responsibilities with Samuel Oni, an affable Nigerian internist in private practice whose patients had some of the longest lengths of stay. When I brought up the issue of hospital stays for heart failure, he confirmed what I suspected. “I understand why hospitals want to cut down length of stay,” he told me matter-of-factly. “As the length of stay goes up, they keep less money. But if I discharge a patient early, I don’t get paid at all. It’s okay if you have enough patients in the hospital, but if you don’t, you sometimes have to drag out the stay. I don’t like to do it, but sometimes you have to.”
Overutilization was the gorilla in the room. Everyone could see it, but few seemed to acknowledge it was
there. A fifty-year-old patient of Oni’s was admitted to the hospital with shortness of breath. During his monthlong stay, which probably cost upward of $200,000, he was seen by a hematologist; an endocrinologist; a kidney specialist; a podiatrist; two cardiologists (me and another doctor, named Chaudhry, who joked, “I am just going to write, ‘Agree with Jauhar’”); a cardiac electrophysiologist; an infectious-diseases specialist; a pulmonologist; an ear, nose, and throat specialist; a urologist; a gastroenterologist; a neurologist; a nutritionist; a general surgeon; a thoracic surgeon; and a pain specialist. The man underwent twelve procedures, including cardiac catheterization, a pacemaker implant, and a bone marrow biopsy (to work up mild chronic anemia). Every day he was in the hospital, his insurance company probably got billed nearly a thousand dollars for doctor visits alone. Despite this wearying schedule, he maintained an upbeat manner, walking the corridors daily with assistance to chat with nurses and physician’s assistants. When he was discharged (with only minimal improvement in his shortness of breath), follow-up visits were scheduled for him with seven specialists.
This case, in which expert consultations sprouted with little rhyme, reason, or coordination, reinforced a lesson I learned many times in my first year as an attending: In our health care system, if you have a slew of physicians and a willing patient, almost any sort of terrible excess can occur.
There are many downsides to having too many doctors on a case. Specialists’ recommendations are often at cross-purposes. The kidney doctor advises “careful hydration”; the cardiologist, discontinuation of intravenous fluid. Because specialists aren’t paid to confer with one another or coordinate care—at least as of this writing; Obamacare is looking to put into place payment systems that will do just this—they often leave primary attendings without a clear direction as to what to do. More important, patients don’t always require specialists. Patients often have “overlap syndromes” (we used to call it aging), which cannot be compartmentalized into individual problems and are probably best managed by a good general physician. When specialists are called in, they are apt to view each problem through the lens of their specific organ expertise. (Perhaps the hardest thing in medicine is to do nothing, especially when you’re called for help.) Patients generally end up worse-off. I have seen it over and over again.
Oni was hardly the worst offender. He once called me about a patient who had a right lung “consolidation”—probably pneumonia, though a tumor could not be excluded—that a lung specialist had decided to biopsy. Oni wanted me to provide “cardiac clearance” for the procedure.
“Sure, I’ll see him,” I said, sitting in my office, checking e-mails. “How old is he?”
“Ninety-two.”
I stopped what I was doing. “Ninety-two? And they want to do a biopsy?”
Oni started laughing. “What can I tell you? In my country we would leave him alone, but this is America, my friend.”
Though accurate data are lacking, the overuse of health care services in this country probably costs hundreds of billions of dollars each year, out of the more than $2.5 trillion that Americans spend on health. Are we getting our money’s worth? Not according to the usual measures of public health. The United States ranks forty-fifth in life expectancy, behind Bosnia and Jordan; near last in infant mortality, compared with other developed countries; and in last place in health care quality, access, and efficiency among major industrialized countries, according to the Commonwealth Fund, a health care research group.
And in the United States, regions that spend the most on health care appear to have higher mortality rates than regions that spend the least, perhaps because of increased hospitalization rates that result in more life-threatening errors and infections. It has been estimated that if the entire country spent the same as the lowest-spending regions, the Medicare program alone could save about $40 billion a year. In some cities Medicare pays more than twice per person what it pays in others. For example, Medicare spends $8,414 per person per year in Miami but only $3,341 in Minneapolis. Even within states there are huge variances. People who live in St. Cloud, Minnesota, are half as likely to undergo cardiac bypass surgery as those in Detroit Lakes, and more than twice as likely to undergo back surgery as those in Rochester. If you have gallstones and live in Wadena, you are three times more likely to have gallbladder surgery than someone who lives in Minneapolis. Among the sixty largest communities in Minnesota, there is a fourfold variation in the frequency of coronary angioplasty and a more than threefold variation in carotid surgery. The greatest variation is in a type of prostate operation, for which rates vary an astonishing sevenfold, ranging from 1.6 per 1,000 Medicare beneficiaries in Rochester to 11.6 per 1,000 in Bemidji.
We don’t know exactly why these variations exist, but we do know that in regions where there are more doctors, there is more per capita utilization of doctors’ services and testing, including consultations, hospitalizations, and stays in intensive care. (I am reminded of a particle collider: the more energy present, the more mass that’s created.) The Dartmouth Atlas of Health Care, a publication of the Dartmouth Institute for Health Policy and Clinical Practice, has shown that at my hospital—hardly an outlier—Medicare beneficiaries will see on average seventeen physicians and receive more than fifty physician visits during the last six months of their lives. John Wennberg, a researcher at Dartmouth, has dubbed this supply-sensitive care—the volume of care provided is influenced by the local supply of doctors and health services—and as is by now well-known, health care expenditures don’t translate into better outcomes. In fact, health outcomes in the highest-spending regions of the country may be worse. “The hospital is a great place to be when you are sick,” Remsen, the senior hospital executive, told me. “But I don’t want my mother in here five minutes longer than she needs to be.”
Overutilization in health care is driven by many forces: “defensive” medicine by doctors trying to avoid lawsuits (unnecessary tests add an estimated $150 billion each year to the health care budget); a reluctance on the part of doctors and patients to accept diagnostic uncertainty (thus leading to more tests); lack of consensus about which treatments are effective; and the pervading belief that newer, more expensive drugs and technology are better. The most important factor, however, may be the perverse financial incentives of our current fee-for-service system. Doctors are usually reimbursed for whatever they bill. As reimbursement rates have declined to control runaway health costs, most doctors have adapted—even if subconsciously—by increasing the quantity of services. If you cut the amount of air you take in per breath, the only way to maintain ventilation is to breathe faster.
Overtesting and overconsultation have become facts of the medical profession. The culture today is to grab patients and generate volume. Internists aren’t gatekeepers (as managed care advocates once envisioned) as much as ushers, escorting favored specialists onto a case with “friendly” consults, or calls for help. The probability that a visit to a physician results in a referral to another physician has nearly doubled in the past ten years, from 5 percent to more than 9 percent. Referral rates to specialists in the United States are estimated to be at least twice as high as in Great Britain. The rates reflect several aspects of American medicine: increasing specialization, the lack of time for any doctor to give to complex cases, and fear of lawsuits over not consulting an expert. At the same time, referrals are also a way for cash-strapped doctors to generate business.
Bob and Joe and Dave have an unwritten agreement to call one another when patient issues arise outside their scope of expertise. If Bob, the nephrologist, sees a patient, he finds a cardiac and a gastrointestinal issue and consults the other two specialists, and vice versa. It’s not kickbacks per se, which are illegal, but there is a mutual scratching of backs. Physician-to-physician referrals are doctors loudly declaring their independence from insurers and the federal government. Insurance companies can restrict medications, tests, and payments. But they still cannot tell us whom or when we can a
sk for help.
When I was in training, simple referrals from internists, like patients with only mild hypertension, bothered me as a waste of time. Now that I am in practice, I have learned to welcome them. I haven’t changed my mind that these referrals are probably unnecessary, and there is plenty of evidence that wasteful expert consultation is adding to health costs and creating redundant care. But as a full-fledged doctor I appreciate the business. It is hard not to view a referral as an overture from another physician, and it is equally hard not to return the favor. However, referrals can put you into a moral bind. Should you refer patients back to certain doctors—not necessarily the best doctors; sometimes even assholes who aren’t particularly good with their patients—just to sustain your business?
It is a paradox of specialized medicine. Specialists are better paid than primary care physicians, but they are also less autonomous because unlike primary care physicians, they depend on other doctors for referrals. So there is tremendous pressure on specialists to keep referral sources happy, especially in doctor-saturated areas like Long Island.
In the spring of my first year at LIJ, a cardiologist named Richard Adelman sent a seventy-four-year-old woman with a leaky heart valve to the hospital for valve surgery. Adelman no longer did rounds at LIJ, so his patient, Mildred Harris, was assigned to me. Ms. Harris had few teeth and hollow cheeks and kind of gummed her words when she spoke. Besides having emphysema and diabetes, she was frail and virtually bedbound, making postoperative rehabilitation difficult. After talking to her, I decided that surgery would be too risky, so I canceled the operation, increased the dosages of her medications, and, after several days, told her I was going to send her home.
Doctored Page 11