Medical Device Daily Washington Editor

WASHINGTON – Healthcare costs keep rising in many economically advanced countries, and governments tend to point to advances in medical technology as the major contributor to these increases.

But now, the medical device and pharmaceutical industries may find some solace in a recent analysis that the growth of medical insurance – both public and private – over the past few decades has driven much of that increase.

A panel discussion at the American Enterprise Institute (AEI; Washington) on Monday concerning the respective roles of insurance and med-tech was prompted by the publication of a paper by Amy Finklestein, PhD, assistant professor of economics at the Massachusetts Institute of Technology (MIT; Cambridge). In that paper, she makes the case that increases in health insurance coverage have played a more significant role in the rise in healthcare costs than previously thought.

As the gold standard, Finklestein cited a 15-year study of the impact of free healthcare coverage on consumption of health services, titled the RAND Health Insurance Experiment (RHIE), funded by the U.S. Department of Health, Education and Welfare (the renamed Department of Health and Human Services ) and conducted by RAND (Santa Monica, California).

According to a summary of the original study penned by Emmett Keeler, PhD, of RAND: “Compared with free care, cost sharing reduced spending consistently, with the stingiest plans spending about two-thirds of free care.” Keeler notes that “the cost of treatment episodes was no smaller . . . but people did not seek care as frequently.”

Finklestein pointed out that as a percentage of gross domestic product, U.S. healthcare spending has shot up from 5% to 16% between 1960 and 2004, and between 1960 and 2000, the percentage of total healthcare outlays covered by insurance went from under 50% to 80%.

Much of the medical innovation since the 1960s has been in the area of cardiovascular care, and major reductions in mortality and morbidity have come in the over-65 age group, life expectancy in that group jumping 25% bet-ween 1960 and 2000.

However, Finklestein said that the RHIE looked at the impact only on those who were enrolled in the study and failed to examine the possibility that “market-wide changes in health insurance can fundamentally alter the nature and character of medical practice in ways that small-scale changes will not.”

She said that her extrapolations of RHIE data on the impact of Medicare on health spending would “explain half of the increase” in that spending. But she also advised “considerable caution” about the conclusions because of concerns regarding “external validity.”

Finklestein also noted that while “the introduction of Medicare appears to have had no impact on elderly mortality in its first 10 years,” it did reduce out-of-pocket spending by beneficiaries sufficiently to cover “between one-half and three-quarters of the costs of Medicare.”

Her study also looked at regional impacts. “In New England, about half [of Medicare eligibles] had insurance, but in the Southeast only about 12%” were covered by some other insurance, hence a rather different impact of Medicare in these two regions.”

She also noted that for the time-frame of her analysis – 1965 to 1970 – those who were old enough to qualify for Medicare made up only about 10% of the U.S. population, which was less than 200 million at the time.

Finklestein's conclusion: “Medicare is associated with a 37% increase in real hospital expenditures (for all ages) between 1965 and 1970,” a figure she said dwarfed by a factor of six “the impact of an individual's health insurance on health spending.” She stated also that a number of other Western nations “have experienced sustained growth in the healthcare sector over the past half-century” and that it might be instructive to examine “whether other health insurance systems have had a similar impact on health spending.”

Melinda Beewkes Buntin, co-director of the Center for Health Care Organization, Economics and Finance at RAND, displayed a chart that showed that total expenditures for healthcare in the U.S. were on the rise prior to 1965, when Medicare was legally established, and a similar trend was seen in the payroll expenditures on insurance. She made the classic economic case that demand for healthcare was inelastic – relatively insensitive to price increases because “it's a necessary good for which there are no close substitutes.”

Her analysis indicated that earlier extrapolations of RHIE data showed that only 10% of the increase in healthcare spending could be chalked up to the aggregate increase in insurance, “maybe as much as an eighth.”

She noted that Medicare may have triggered another classic economic phenomenon, that of spillover costs and benefits. Medicare may have prompted “greater adoption of cardiac technologies,” she said, for non-Medicare patients, and she pointed out that policymaking should ask the tough question: “Are we getting our money's worth out of this technology?”

Buntin also made reference to a “medical/industrial complex” that has produced a sort of feedback mechanism between increased insurance coverage and ever-improving technology.

Philip Ellis, a senior analyst at the Congressional Budget Office (CBO), reminded the audience that “doing these extrapolations is a dicey thing,” but offered numbers from two analyses that suggested the spread of insurance boosted health costs by 10% and 13% between 1940 and 2000. However, he pointed out that the share assigned to technology was something of a catchall for factors that were difficult to capture statistically. Those two estimates for the impact of technology on healthcare expenditures were 65% and 49% (Newhouse 1992 and Cutler 1995).

Ellis also offered charts, one depicting sharpening increases in the cost of Medicare over the projections offered in 1965. “Things didn't really get out of control until CBO came along, so maybe it's our fault,” he quipped.

According to his chart, Medicare expenditures were originally projected to double between 1965 and 1995, but that they have nearly quadrupled. Ellis's numbers also indicated that the share of healthcare costs paid out of pocket had begun to fall before 1965 and that between 1960 and 2005, hospital costs paid by patients fell from 20% to less than 5%.

A similar trend appears for prescription drugs, 95% of which was paid by patients in 1960 and slightly more than 30% in 2005.