The Diminishing Returns of Modern Medicine: (Rapid Angioplasty Not so Good)
For several decades, doctors have taken it as an article of faith that more rapid treatment of an acute heart attack improves patient survival. The treatment of choice today is angioplasty, where tiny balloons and wire-mesh cylinders called stents are used to open up blockages in the arteries that feed the heart. A “door to balloon,” or D2B, time – the period from a patient’s arrival in the hospital to inflation of the coronary balloon – of less than 90 minutes has become a quality metric for American hospitals. “Time is muscle” is the operative mantra, and the shorter the delay, the better.
However, an article published in September in The New England Journal of Medicine is forcing a rethinking of this faith. In the study, which examined nearly 100,000 patients treated at 515 American hospitals over four years, shorter D2B times did not improve in-hospital survival. The median D2B time dropped to 67 minutes, from 83, in the period studied, but short-term death rates did not change.
There are several plausible explanations for this result. Perhaps heart-attack patients at low risk for death are already getting expeditious treatment, and those who are sicker – and therefore at higher risk for death – are experiencing the most delays. Perhaps the follow-up time in the study was too short, and if we waited a bit longer a survival benefit would be seen. Or perhaps there is another reason: modern cardiology has reached the limits of what it can do to reduce mortality after a heart attack.
This would be a mixed blessing, of course. Perhaps more than any other medical specialty, cardiology has been at the forefront of technological innovation and quality improvement in the past 50 years. This golden period has witnessed a hailstorm of life-prolonging advances, including implantable pacemakers, coronary bypass surgery and heart transplantation. Cardiovascular mortality has dropped significantly over this period, but the rate of that decline has slowed.
This should not come as a surprise. The law of diminishing returns applies to every human enterprise, and medicine is no different. In-hospital mortality after an acute heart attack has dropped 10-fold, from 30 percent to 3 percent, since the invention of the modern cardiac care unit in the 1960s. Can shaving a few more minutes off D2B time possibly yield any additional benefit?
There are other examples of such diminishing returns. In my field, heart failure, medications such as beta blockers and ACE inhibitors have profoundly improved survival since their advent in the mid-1980s. Yet recent studies of newer agents – endothelin blockers, vasopressin antagonists – have shown little benefit. Today, patients’ cardiac risk factors, such as hypertension or high cholesterol, are better controlled. It is getting harder to improve on existing successes.
No doubt we should celebrate the rise of evidence-based therapies. For example, more than 90 percent of patients who present directly to angioplasty-capable hospitals today have D2B times of less than 90 minutes, with a median time of approximately 60 minutes – a major improvement from only a few years ago. However, this means the bar is continually being set higher for every new treatment.
Clinical research in its current form, focusing on “me too” drugs or add-on therapies or optimizing existing procedures, is increasingly producing only marginal advances. For example, no medicine developed over the past 20 years to treat acute heart failure has improved survival over the standard of care. Should we continue to invest in exorbitant treatments that provide only modest gains when one dollar out of every six in this country is already spent on health care?
We have to get smarter about how we try to improve care. The push to reduce D2B time is revealing. Hospital D2B times are publicly reported and linked by insurers to reimbursement. With the threat of financial penalties for any delays, roughly a third of angioplasty-team activations are false alarms. Treatment is frequently rushed, potentially compromising safety. Individual patient factors, such as mental status or kidney dysfunction, are often not taken into consideration.
A cardiologist in my department recently told me about a case when he was on call. “By the time I got into my car, we were already at the 60-minute mark,” he recalled. “I live a half-hour from the hospital, so to get there on time I had to drive 90 miles an hour at 4 o’clock in the morning. We ended up inflating the balloon at 87 minutes. But I don’t think it helped the patient one bit.”
Instead of focusing on making small cuts to D2B time, we should be focusing on reducing pre-hospital delays – patients with chest pains waiting several hours before calling 911, for example – which dwarf the delays in our current D2B processes. We need better educational campaigns to reduce “symptom to door” times before cardiac teams even get involved.
Quality improvement in medicine is too often a blunt instrument. We try to take what works in certain situations and apply it to all situations. Our methods yield results for populations, not individual patients.
A shift to more personalized medicine will be needed to continue to make the kind of progress to which patients and doctors have become accustomed. Until then, maybe we should be satisfied with securing the gains that have already been made.
Sandeep Jauhar, a cardiologist in New York, is the author of “Intern: A Doctor’s Initiation” and the forthcoming memoir “Doctored: The Disillusionment of an American Physician.”