Medicare Disease Management Pilots

Rising costs and poor quality have led to one health care management “innovation” after another.  One of the most hyped has been disease management, although that term has begun to fall out of common usage as results have not lived up to the hype.  At its zenith, Medicare decided to engage a number of commercial entities to provide disease management to Medicare beneficiaries under a pilot program.  A recent article in the New England Journal of Medicine gives an overview of the experience with these pilots.   (NEJM Article)   Altogether eight companies were selected for what were to be several-year pilots, although many dropped out when they realized they were going to lose money.  About 163,000  beneficiaries were enrolled in the programs, with another 79,000 in a control group. The companies were paid a monthly management fee, but had targets for cost-saving.

The authors analyzed the results by looking at four process-of-care measures, rates of hospitalization and ER use and overall spending, including the management fee.  The average number of contacts was less than one per month.  Generally the DM companies did well on the process-of-care measures, but not that much better than the control group.  Although a couple of companies had success in limiting utilization, most did not.  And all of the companies had increases in average monthly costs per beneficiary, particularly when the management fee was taken into account.  The authors note that the Medicare population may have presented complexity that the companies weren’t accustomed to dealing with and that the results could be affected by regression to the mean for both sick and healthy study participants.

Disease management companies have claimed success in the commercial world, but as these authors point out, the quality of the studies supporting cost-savings is not always high.  On the other hand, there may have been selection issues regarding the beneficiaries who ended up in the intervention group in the Medicare pilots and the infrequent interaction with those beneficiaries may have also affected the outcomes.  Costs could have been higher in the early years of the pilots because many of these beneficiaries may have needed to catch up on skipped necessary care.  It might be hard to show process of care quality improvement because at the same time Medicare had quality reporting initiatives underway for the whole population.  But overall the results of the pilots can only be read to mean that the way disease management was done in them didn’t work very well and certainly doesn’t appear to have resulted in overall savings.