Bundled payments are one reimbursement approach designed to control spending growth. The assumption is that if there is only a set amount of money available, providers will be more cautious and only order necessary services. There have been suggestions, however, that some of the putative benefit could be offset by a growth in the number of “bundles”, as providers seek to use volume to make up for any loss in the former amount of payments. A report from Altarum, which seems to have been aided and abetted by CMS as a way of defending its “innovation”, suggests that volume does not rise when bundled payment methods are used. (Altarum Report) A JAMA editorial had taken the position that the introduction of the knee and hip joint replacement bundled payments had increased volumes.
These researchers looked at all such procedures for Medicare beneficiaries for years 2010 through 2015, by hospital and with adjustments for various patient, hospital and hospital referral region factors. (Adjusting in an analysis is generally a good thing, but it can be used to help reach certain results.) Hospitals participating in the lower extremity joint replacement bundled payment program (which is now mandatory for all hospitals) were compared with non-participating hospitals. The basic analytic method was a difference-in-differences model that compared changes in volume by hospital. The hospitals participating in the bundled payments model had more joint replacement volume than non-participants before the demonstration’s start, which is what you would expect. These facilities probably believed their greater experience gave them an opportunity to better control costs. There was a greater volume increase in joint replacements at these participating hospitals than at non-participating ones, about 4.8% versus 2.6%, but this was not statistically significant. Other factors which may have been associated with higher volumes were less competition in an HRR and lower growth in FFS beneficiaries and with a higher proportion of non-white beneficiaries in an HRR. I am going to call this one a draw. There may actually be an increase in volume due to participation in the bundled-payments, which would make sense if hospitals feel like they are making a profit on them. On the other hand, the evidence for such an effect isn’t particularly strong at this point.