The Impact of High Cost-Sharing on Medical Spending Growth

High deductibles, copayments and coinsurance have become a widespread feature in commercial health plans.  A new analysis carried by the National Bureau of Economic Research attempts to shed light on whether these plans actually reduce medical spending growth.   (NBER Paper)   Most research to date has focussed on whether such plan designs reduce spending in the near term and less on the longer-term effects on spending’s growth rate.  These authors attempted to use variations in the proportion of people with high-deductible plans across states to ascertain impacts on spending growth.  They explore whether changes in levels of insurance coverage affect the introduction of new technologies and/or unit prices, which they view as the primary drivers of spending growth (as opposed to the absolute level of per capita spending).  Not sure I agree that those are the major factors in spending growth–population aging and other factors may be as determinative of spending.  In any event, they have a hypothesis about both the immediate impact on spending, and the impact on spending growth, due to introduction of higher deductibles.  They test that hypothesis using data from 2002 to 2016 from states regarding health spending both for private payers, who have used the high-deductible design, and public payers, who have not, and data from the MEPS for the same time period.

The research confirms that deductibles have increased substantially and become much more common over the study time period.  The average level of deductible in a state does appear correlated both with the level of spending and the rate of spending growth.  The researchers suggest there is more of an effect on spending growth, which they view as related to less use of expensive new technology, particularly since spending growth on drugs, which have a lot of innovation, is more affected than that for hospital services.  I would point out again that I am uncertain about the assumption that new technology is the most important driver of spending growth over longer periods and therefore don’t think that it is accurate to view a lower rate of spending growth as due to less use of new products or services.  Regardless of the cause, the study clearly suggests that while high-deductible plans and other increased cost-sharing may lower spending levels from what they otherwise would be, they appear to have as strong or stronger effect on the rate of growth.  A separate question from whether health spending is reduced is whether the reduction is appropriate; that is are people avoiding needed care or services that weren’t really appropriate.  On this question, the research is also somewhat inconclusive.