Tweet This
The good news according to recent reports is that Medicare won’t become insolvent until 2030, four years later than previously estimated. The bad news of course is that Medicare costs are still a huge, and growing, burden on federal budgets. A recent study concludes that part of the blame for high Medicare costs can be pinned on patients. Places like McAllen, TX, famous for its high healthcare costs, are populated by patients who simply want more medical care than patients in other parts of the country.
Atul Gawande brought attention to McAllen in an influential New Yorker essay, exploring geographic variation in healthcare expenditures. For several decades now, healthcare researchers have documented often dramatic differences in the type and amount of medical care Americans receive as a result of where they live. A woman’s chance of delivering a baby by C-Section may have as much to do with the state she lives in as it does with the state of her pregnancy. Researchers at Dartmouth, in fact, created an Atlas summarizing these geographic differences.
But what explains these differences? Researchers have an incomplete understanding. They are pretty confident that the supply of hospital beds partly explains these geographic differences. All else equal (“all else” including measures of patient age and health), cities with a plethora of hospital beds per capita will have more patients in the hospital at any given time than ones with fewer beds. Build a hospital, the thinking goes, and physicians will find a way to fill it up with paying customers. Physicians fill the beds not simply because hospitals want more revenue, but also because when hospitals are near capacity, physicians start handling more patients’ illnesses in the outpatient setting, illnesses they might be more comfortable treating in the hospital.
New research shows that patients now, too, deserve credit (or blame) for a portion of these geographic differences. In an elegant study, three healthcare economists linked data on healthcare spending with a national survey of people’s attitudes towards healthcare, and discovered that the two were linked to each other in ways that do not appear random.
The attitudinal survey asked people questions like:
- Would you see a specialist even if your doctor thought it was unnecessary?
- Would you decline drugs that shorten your life even if they made you feel better?
When I first saw these questions, I was skeptical. My answer to each question would probably have been “it depends.” For instance, how much would the drug shorten my life and how much better would it make me feel?
And yet even with imperfect questions like these, the researchers discovered astonishing results. Medicare expenses were higher in places of the country where more people said “yes” to the first question and “no” to the second. By their estimate, Medicare spends $129 more per patient in places where patients hold these kinds of preferences. In other words, patient preferences explain 4.6% of the difference between high and low spending regions :
Costs at the end of life were even more strongly associated with people’s answers to these questions, contributing more than $500 of the difference between high and low cost regions.
Keep in mind that even with end-of-life care, patient preferences as assessed by these survey questions explained less than 10% of the difference between high and low cost regions. In other words, patients are partly to blame for causing the high spending we see in places like McAllen, TX, but are probably only a small part of the problem. Nevertheless, healthcare spending is a big problem in the U.S. – with Medicare contributing to federal deficits – so even being a small part of the problem is, in itself, a big problem.
As someone who has worked in behavioral economics for several decades now (Yikes, I must be old!), I am obliged to offer one important word of caution about this research. Those of us who work in behavioral economics have shown that many preferences are quite flimsy. Subtle factors can dramatically alter what people focus on when they make decisions. Thus, while this new study provides compelling evidence that patient preferences contribute to high healthcare spending, those preferences are not necessarily fixed. If we truly want to curb healthcare spending, we should think about ways to shape people’s attitudes towards medical care.
Patients do not deserve the lion’s share of blame for healthcare spending. But they are not passive sheep either, simply following doctors’ orders without any say in their medical care. Patient demand for medical care is an important part of our healthcare spending problem. Taming this demand may be an important part of any efforts to curb healthcare expenditures.
