Two recent Health Affairs articles have focused on different aspects of bundled payment programs, and appear to show different variations on opportunities for success.
Variations in Payment
The first article, Variations in Medicare Inpatient Surgery Payments Strengthen Case for Bundling, is a rigorous analysis of the costs of several selected episodes (“bundles”) of services including hip replacement, coronary artery bypass grafts, back surgery and colectomy. Its objective was to quantify the variations in Medicare payments for these bundles, identifying and eliminating as many contributing factors as possible. The details of these adjustments are described in the full text of the article.
Their analysis found significant variations in payments between the lowest and highest hospital quintile groups; for example the unadjusted cost of back surgery differed by 130% on an unadjusted basis, and 33% after price and case mix adjustments. The other procedures had similar results. Much of the variation was in post-acute care, and variation in physician payments was generally not a major factor, although the ability of physicians to control most of these costs was significant. The study found that hospitals that were higher in cost for one procedure were lower in cost for other procedures; however about 30% of the hospitals were higher costs in all procedures.
This study indicates that there may be significant opportunities under the bundled payment initiative for hospitals and physicians to identify areas in which costs can be reduced, and therefore to achieve savings. However, this opportunity varies significantly between procedure types, even within the same hospital, so careful analysis of data will be critical to identify these opportunities. Fortunately, the data provided by CMS to bundled payment applicants will contain information for the specified DRGs for all hospitals within the specified geographic area, which may provide information to physicians as to practices used by other providers for patients in the same DRG, and the resulting costs. This may provide guidance to physicians seeking to revise care practices to reduce costs.
Much more useful information is provided in this article, which is important reading for providers interested in participating in the bundled payment initiative
Implementation of the Prometheus Model
The other article (The PROMETHEUS Bundled Payment Experiment: Slow Start Shows Problems In Implementing New Payment Models) focused on the implementation of the Prometheus bundled payment model, which is a clinically driven payment model developed to compute bundled payment amounts for a variety of different types of cases. The objective of this model was to be easily implementable by payers and providers, and provide a starting point for the transition from fee for service payment. This model was initially planned to be pilot-tested by three payers and providers in Pennsylvania, Illinois and Michigan.
The article describes the reasons why none of these pilots has been implemented, due to many factors related to the complexity of the model, the difficulty of communication, and the significant change in mindset required by both providers and payers. The article speculates that these factors will slow the adoption of bundled payment by other payers in other markets.
Opportunities still exist
We’re more optimistic about the opportunities for bundled payment. The Medicare bundled payment initiative appears to have received a strong initial response, although it will be interesting to see how many applicants filed letters of intent simply to obtain the CMS data and will not ultimately participate in the initiative. The Medicare initiative is simpler to implement than the Prometheus model, whose complexity appeared to be a major deterrent to its use. The success of participants in the Medicare Heart Bypass and Acute Care Episodes demonstration projects also suggests that this initiative may be advantageous for many providers. But thorough analysis of the data will be required since, as noted in the first article, not all DRGs will offer the same opportunity for savings. Careful selection of DRGs and in-depth analysis of their costs characteristics will be critical for success in this initiative.