Most of the research on geographic variation in health spending has been conducted in regard to the fee-for-service Medicare program. This naturally raises the issue of whether commercial health insurers also experience significant geographic variation in spending and whether that variation is coincident with that seen in Medicare. A recent study examines those issues. (AJMC Article) The researchers examined Medicare data as pulled together by the Dartmouth Atlas project and a commercial insurance database largely composed of large company claims. A broader commercial database would have been more useful.
The researchers found that there was indeed significant geographic variation in private health plan spending, in fact there was about a third greater variation than in Medicare, but that there was a different pattern from Medicare spending variation. High-spending areas in Medicare were not necessarily that for commercial insurance and vice versa. Utilization patterns were somewhat similar, so most of the difference appeared to be price related. Medicare sets prices, subject to some geographic adjustment; private plans must negotiate contracts and prices with providers.
The factor which appeared to best explain commercial spending variation was the degree of hospital concentration, or lack of competitiveness. The fewer hospitals in an area, the higher the spending. This is consistent with the notion that competition among hospitals leads to lower rates for private payers. This is also a major finding of the Massachusetts AG report described in an earlier post. While this may seem like a “duh” to economists, many had actually believed that health care markets were different and some had suggested that more hospitals in an area might mean higher costs because they all would just generate more demand to meet revenue needs. This research suggests that is not the case. Policymakers should seriously consider the possibility of undoing some of the hospital mergers allowed over the last three decades.