The debate over whether higher health care spending in America has translated to value may be analogous to today’s cost of a car. The average price of a new car in 2016 was $34,000—about the price of a Honda Accord.
Adjusted for inflation, $34,000 in 2016 was the equivalent of $4,600 in 1966. Back then, $4,600 would buy you two new Chevy SS Chevelle hardtops, according to John Pearley Huffman’s analysis on TheDrive.com. Considering that the Accord is full of toys (like power windows and back-up cams) that didn’t exist in 1966—not to mention that it’s considerably safer, more reliable, and gets better gas mileage than the Chevelle—you could argue that the Accord is twice the car that the Chevelle was and, thus, worth twice the price.
Oh, and when you floor it, Huffman added, the Accord outraces the Chevelle, which was one of the best-loved muscle cars of its time.
So, what does all this have to do with health care? For all the talk about rapid increases in health care costs, we just might be getting better value for it. So say David Wamble, director of health economics at RTI Health Solutions, and colleagues in a recent Health Affairs article, “What’s Been the Bang for the Buck? Cost-Effectiveness of Health Care Spending Across Selected Conditions in the U.S.”
Wamble’s team looked at the degree to which increased spending on seven chronic conditions in the United States has been a good investment. Comparing per-person costs against a fairly simple metric that accounts for poor health and premature death, the researchers found that health status improved among populations with six of the seven conditions. Of those six populations, per-person treatment costs decreased in four.
The bottom line? We often look at spending at an aggregate level, which may not be the most appropriate way to view health care spend, Wamble tells Managed Care. “You really have to take it to another level—each disease has its own trajectory of costs and innovations,” which he says compels policymakers to be strategic about health care spending.
“You can’t necessarily equate increased spending to ‘bad.’ We were able to demonstrate that in many cases we’ve had good value for our spend. In other therapeutic areas, maybe that’s not the case, and we should therefore be more interested in a better understanding of what is driving these numbers.”
To arrive at its findings, Wamble’s team crunched data from the CDC, the Medical Expenditure Panel Survey, and the Global Burden of Disease study. GBD is a World Health Organization-led initiative that quantifies the effects of diseases or injuries every five years. Its last update, for 2015, was published in the Lancet the following year. GBD’s key metric is the disability-adjusted life year (DALY), which combines years of life lost with years lived with disease or disability.
The seven conditions were chosen because they result in substantial morbidity and mortality in the United States. The researchers looked at changes in prevalence, spending, and DALYs over a 20-year period (1996–2015), with spending defined as post-diagnosis out-of-pocket and claims payments. For each condition, Wamble and colleagues funneled down from aggregate data to a per-person view so as to isolate the effect of medical interventions on a disease.
At the head of the class: lung cancer, for which the inflation-adjusted change in per-person treatment cost actually declined by nearly $11,000 over the 20-year period. What’s more, DALYs decreased by 2.44 per person—a remarkable drop in death and disability for a disease that is often caught too late. This degree of decline was second only to HIV/AIDS, which saw a 9-DALY decrease thanks to its commutation from a death sentence to a chronic disease as treatments for it improved.
Cerebrovascular and ischemic heart diseases joined lung cancer and HIV/AIDS in the category of lower inflation-adjusted spending with better outcomes. Together, those four might be thought of as the Honda Accords of spending and value. Breast cancer and diabetes were in a different category; per-person spending increased for them but outcomes also improved.
COPD, for which costs increased and outcomes decreased, might be an Acura ILX, which you can look up for yourself in Consumer Reports.
“One could argue that we’ve done pretty well, that our spend has paid off and that we’ve had good value in our spend,” says Wamble. What’s more, adjusting the model to reduce the proportion of outcomes attributable to medical spending from 100% to 50% (allowing room for the impact of lifestyle modifications and social environment) produced no appreciable change in results.
“Interestingly, we saw the same general theme with four of the seven conditions, still showing improved outcomes and lower costs, which was exciting and, frankly, surprising to us.”
Others argue that the sponsor of this study, the National Pharmaceutical Council—a pharma-funded think tank—has a bias in correlating spending with outcomes. In a February 21 blog post, the Lown Institute, a Boston not-for-profit group critical of the industrialization of health care, called the study a “shrewd move by pharma to use economic analysis to back up the idea of high prices leading to innovation” and pointed out that the relationship between health care spending and health is not always direct.
Wamble replies that conversations around health care spending in the U.S. have long focused on questions of “How much do we spend?” rather than delving deeper to ask, “How well do we spend?”
“What’s driving these numbers is a really important next step in understanding how we approach health care spending and how we consider disease-specific innovation and costs,” he says, “as opposed to applying aggregate-level spending policy.”
In choosing the DALY as the key metric for the study, Wamble and colleagues contribute to the discussion of “value” in health care spending. The DALY was created by the WHO as the antithesis to the quality-adjusted life year (see “QALY vs. DALY,” above), which gets no love from U.S. P&T committees—its embrace by the Institute for Clinical and Economic Review and, more recently, CVS Caremark notwithstanding. The DALY is relatively unknown in this country.
“If you think of health care, it can be beneficial in two ways: You can live longer or live better, and the DALY measure helps us look at both,” says Wamble. DALY was an appropriate measure for this study, he says, in part because it scratches at outcomes that matter to patients.
“In the conditions we included in this analysis, sometimes you find that disability or morbidity is a very important measure to the person receiving care. The DALY allows us to do that across these conditions.”
Can the methods of Wamble & Co. be applied to other diseases—say, by a pharmacy director in Kentucky, where asthma rates are the nation’s highest, or by an employer in heavy industry who is concerned about absenteeism from low back pain? “I think they can,” says Wamble, though he adds a caveat that beneficiary churn may reduce the robustness of a methodology that took a 20-year perspective. The point to remember, he says, is that cost-management strategies should be disease-specific.
“It’s important to recognize that what’s going on in one particular condition—such as asthma or low back pain—is much different than innovations in cost in lung cancer or breast cancer. And as simple as that might sound, I don’t know if that is being strongly considered or applied at a policy level.”