Is a Poorly Functioning Health Care System to Blame for Low Life Expectancy in the U.S.?

Featured in print Bulletin on Aging & Health

Life expectancy in the U.S. lags below that in other industrialized countries. This is particularly true for life expectancy at age 50, which is 3.3 years lower than in Japan and 1.5 years lower than in Australia, Canada, France, Italy, Iceland, Spain, and Switzerland. The U.S. also spends more on health care than other countries - 16% of GDP in 2007, or as President Obama recently noted, "almost fifty percent more per person than the next most costly nation."

The coincidence of these two facts has led some policy makers and health analysts to wonder if a highly inefficient U.S. health care system is to blame for poor health outcomes. This is the question examined by researchers Samuel Preston and Jessica Ho in their recent working paper, "Low Life Expectancy in the United States: Is the Health Care System at Fault?" (NBER Working Paper 15213).

One reason to be cautious in drawing a causal inference from the coincidence of high spending and poor health outcomes is that health outcomes do not depend solely on what transpires within the health care system. Personal health behaviors such as diet, exercise, smoking, and compliance with medical protocols play a critical role.

The authors focus on two diseases - cancer and cardiovascular disease - that jointly account for over 60 percent of U.S. deaths after the age of 45. Focusing on specific diseases rather than aggregate mortality from all causes may make it easier to distinguish the role of the health care system in health outcomes from that of health behaviors and other factors. It is also useful to focus on treatment and outcomes for those with the disease, rather than disease incidence rates, since health behaviors are likely to play a bigger role in incidence.

The authors begin by looking at cardiovascular disease. The share of those with heart disease that receive medication is higher in the U.S. than in Europe.(61 percent vs. 55 percent). For those with high cholesterol the US advantage is 88 percent vs. 62 percent. The share of individuals whose high blood pressure is successfully controlled with medication is 66 percent in the U.S., versus 25 to 49 percent in other countries.

The authors also examine treatment for a heart attack or stroke and survival rates following these acute events. The use of aggressive surgical treatments following a heart attack or strokes - such as angioplasty, coronary bypass, or surgical removal of plaque inside the carotid artery - is more common in the U.S. than in other countries. Evidence suggests that the use of these aggressive treatments significantly boosts the patient's survival prospects.

The authors begin their discussion of cancer by examining cancer screening. Compared to European countries, the U.S. has higher screening rates for many cancers, including prostate, breast, cervical, and colorectal. Absent other differences in health behaviors or health care systems, higher screening rates would be expected to lead to a higher prevalence of cancer diagnoses. Indeed, 12.2 percent of Americans over age 50 report ever having been diagnosed with cancer, versus 5.4 percent of citizens in ten European countries. More screening can lead to cancers being caught at an earlier stage, when they are easier to treat. Consistent with this, the average stage at diagnosis is lower in the U.S. and 5-year survival rates are higher.

The authors offer a more in-depth analysis of prostate and breast cancer, two cancers that account for a large share of cancer deaths and (in the case of prostate cancer) are relatively unrelated to behavioral factors. The main screening test for prostate cancer is the Prostate Specific Antigen (PSA) test. The test is controversial, as it can produce false positives, and prostate cancer treatment can have unpleasant side effects. As PSA screening became more common in the U.S., the reported incidence of prostate cancer doubled, but the share of tumors that had spread at time of diagnosis fell from 25 percent to 4 percent. This is important, as early stage prostate cancer is highly treatable.

As with cardiovascular disease, the U.S. is more aggressive in treating prostate cancer than are other countries, using radical prostatectomy or radiation relatively more often and "watchful waiting" less often. These more aggressive approaches have been shown in clinical trials to lower disease progression and mortality. The U.S. has experienced significantly faster declines in prostate cancer mortality since the 1990s than have other countries, and studies have attributed most of the decline to expanded PSA testing and improvements in treatment.

Turning to their analysis of breast cancer, the authors document that the U.S. has historically administered mammograms, the most important diagnostic tool for breast cancer, more widely than have other countries, though the gap has narrowed in recent years. Mammography for women ages 50 to 69 has been shown to lower breast cancer mortality, and consistent with its higher use in the US, breast cancer is diagnosed at an earlier stage in the U.S. than in Europe.

In terms of treatment, the use of breast conserving surgery, multi-agent chemotherapy, and the drug tamoxifen have been shown to reduce mortality. While international data on the use of these treatments is limited, it is clear that their use in the U.S. has grown over time. Breast cancer survival rates are better in the U.S. than Europe, a difference thought to result from earlier diagnosis. The U.S. has also experienced a significantly faster decline in breast cancer mortality over time than have other countries.

The authors conclude that mortality reductions from prostate and breast cancer have been exceptionally rapid in the U.S., relative to peer countries, a finding they attribute to wider screening and more aggressive treatment. They note "it appears that the U.S. medical care system has worked effectively to reduce mortality from these important causes of death."

They caution "it is possible that the U.S. health care system performs poorly in preventing disease in the first place," although this is difficult to study because of lack of good data on incidence and because health behaviors may play a bigger role in incidence than treatment and mortality post-diagnosis. They also caution that in spite of their results, there could be great inefficiencies in the U.S. health care system, resulting from misallocation of physician and patient incentives, defensive medicine, and the like. But to the question "does a poor performance by the U.S. health care system account for the low international ranking of longevity in the U.S.?" the authors answer no.

The authors gratefully acknowledge financial support from Social Security Administration through a grant to the National Bureau of Economic Research as part of the SSA Retirement Research Consortium.