The following originally appeared on The Upshot (copyright 2015, The New York Times Company).
The expansion of health care insurance can drive the development of new technologies that can drive up health care spending. But it doesn’t have to.
By historical standards, health spending growth has been low since 2002. As the federal and state governments expand health insurance coverage, will this relatively low growth continue? The answer depends in large part on how investments in health care technology are encouraged.
Health economists have shown that both insurance expansion and technology drive health spending upward. A few studies have teased out a deeper relationship between the two: By increasing the market for health care products (like prescription drugs and medical devices), coverage expansion encourages investment in health sector research and development. And, depending on what technologies are subsequently brought to the market, this can lead to more costly, though often better, care.
By increasing coverage, the Affordable Care Act will expand the market for prescription drugs and medical devices, so we might expect growth in new products. For example, Daron Acemoglu and Joshua Linn, in a study published in the Quarterly Journal of Economics, found that a 1 percent increase in a drug category’s potential market size was associated with a 6 percent increase in the number of new drugs entering the market in that category.
We might also expect to spend more for new health technology encouraged by coverage expansion. Indeed, the expectation of greater drug and device manufacturers’ revenue was an argument for Obamacare’s reduction in prices paid for prescription drugs by government programs and its excise tax on medical devices. (Though in reality, consumers, not the medical device industry, will pay most of that excise tax.)
But it need not be the case that technology encouraged by coverage expansion increases per person costs. In an insightful paper published in 1991 in the Journal of Economic Literature, Burton Weisbrod distinguished between two types of health care technologies: those that enhance quality at potentially high cost and those that reduce cost without substantially sacrificing quality. Historically, quality-enhancing but cost-increasing innovations were relatively common, particularly before the managed care era of the 1990s. Studies from the 1980s found that when hospitals competed, they tried to outdo each other on high-cost amenities and technologies rather than reduce prices.
This “medical arms race” partly explains the historical rapid growth in health care spending. However, it’s also the case that much of health care’s cost-increasing technology increased the quality of health care. In the New England Journal of Medicine, David Cutler, Allison Rosen and Sandeep Vijan showed that from 1960 through 2000, life expectancy grew almost seven years and at a cost of nearly $20,000 per year of life gained. The vast majority of gains were because of improvements in care for high-risk, premature infants and better treatments for cardiovascular disease, both of which can be costly.
History also offers examples of cost-reducing health care technology. Some did not degrade quality. For instance, when Medicare reduced payments for kidney dialysis in the mid-1970s, new equipment emerged that halved treatment time, saving labor costs. Others sacrificed quality. When changes to Medicare payment policy in the early 1980s made cochlear implants less profitable, 3M ceased making them and ended research on a more advanced, and presumably more costly, model.
So how, historically, have policy makers and insurers driven health care costs downward? Mr. Weisbrod argued that shifting the cost risk onto providers is one traditional means. For instance, H.M.O.s that did so were more likely to encourage investment in cost-reducing health care technologies. Additionally, paying a fixed, diagnosis-based price for treatment — as Medicare began to do in the early 1980s, with private insurers following — increased the profitability of drugs that substitute for more costly treatments, like beta blockers instead of coronary bypass surgery and cimetidine instead of ulcer surgery.
What does all of this suggest about the future of health care spending growth? We know that coverage expansion increases the profitability of certain kinds of health care technology. The market for costly but effective new drugs for hepatitis C — Sovaldi and Harvoni — is certainly larger under coverage expansion than it would otherwise be. Correspondingly, revenue of Gilead Sciences, the drugs’ manufacturer, has soared, along with its stock price.
But we also know that the new cost-conscious climate could put the brakes on other kinds of costly technology. Medicare’s Accountable Care Organizations — which incentivize cost saving by providers — higher cost sharing for the privately insured, and other new arrangements between private insurers and providers all might slow the adoption of certain high-cost medical devices.
One example is the proton beam accelerator, used to treat certain kinds of cancer. Construction of facilities for them has grown in recent years. Proton beams can substitute for more traditional radiation treatment of prostate cancer, but at twice the cost and with no evidence of better outcomes. For this reason, insurers like Blue Shield of California and Aetna have stopped covering it, and others may follow.
Putting these and other factors together, Amitabh Chandra, Jonathan Holmes and Jonathan Skinner, in a paper for the Brookings Institution, made a quantitative prediction. Health spending will grow at a pace above that of the overall economy, but slower — 1.2 percent above gross domestic product — than it has historically — 2.4 percent — though the authors also express caution about this prediction.
The uncertainty in predictions like this is vigorously debated in health policy circles. Will deductibles and other cost sharing continue to grow larger, hold steady or decrease, for example? Higher cost sharing would weaken the market for certain new health care technologies. Another issue is the payment level of Medicaid programs, which has typically been well below that of private plans and Medicare, leading to reduced access to care for Medicaid enrollees. (A temporary increase in Medicaid payments for primary care expired at the end of December.) Lower Medicaid payments would reduce the rate of adoption of technologies applicable to the population it serves.
We may not be able to precisely predict the extent to which health care technology will drive health spending in the future. What’s clear, however, is that changes in technologies brought about by the shifting market for them will play a central role in health care spending growth in the future.
