Hospitals participating in phase two of a Centers for Medicare & Medicaid Services (CMS) bundled payment model apparently focus only on those conditions that they treat most frequently. According to research published in the March issue of Health Affairs, this means that institutions are enrolled, at most, in two or three conditions out of the possible 48.
Thomas C. Tsai, MD, MPH, of Brigham and Women’s Hospital in Boston, and colleagues reviewed data on hospitals participating in phase one, phase two and not participating in the CMS Bundled Payments for Care Improvement (BPCI) initiative model two. Of the four potential Medicare bundled payment models, model two is the most comprehensive, including all Medicare Parts A and B payments for 48 selected clinical conditions. At the time of the analysis, 225 hospitals participated in BPCI model two, phase one; 107 participated in BPCI model two, phase two; and 3,028 hospitals did not participate.
Phase two hospitals were more frequently urban, major teaching hospitals, large and nonprofit. They were also more likely to be in the Northeast and affiliated with post-acute care providers. Of phase two hospitals, 72 percent utilized BPCI for three or fewer conditions and 50 percent for only one. The most common conditions BPCI was used for in these hospitals included major lower extremity joint replacement (72.9 percent), congestive heart failure (35.5 percent), chronic pulmonary disease, asthma or bronchitis (25.2 percent), simple pneumonia or respiratory infections (20.6 percent), and hip and femur procedures that did not require major joint replacement (18.7 percent).
Patterns emerged about the major enrolled conditions phase two hospitals were participating in, particularly volume of discharges. The average number of discharges among hospitals participating in BPCI for congestive heart failure was nearly twice the national average (140 vs. 72.7 discharges), which was consistent among the other major enrolled conditions. Hospitals enrolled in congestive heart failure payments as part of model two, phase two owed an average of 3.1 percent of total annual Medicare discharges to that condition.
Average 30-day total payments were similar whether participating in phase one, phase two or not participating; however, spending patterns were condition specific. For example, congestive heart failure, obstructive pulmonary disease and pneumonia tended to have high readmission rates (24.1 to 25.3 percent) to account for variation in spending. Meanwhile, post-acute care spending varied between 26.2 percent and 75.3 percent for chronic obstructive pulmonary disease or hip and femur procedures, respectively.
Tsai et al noted that hospitals did not choose their targets randomly; conditions hospitals focused on ones where they had substantial experience. While the small number of specific conditions of focus in phase two might be, as they wrote, disappointing for the overall viability of the BPCI program, it does provide Medicare with the opportunity to look at more tailored approaches to condition-specific, episode-based bundled payments based on case-mix and clinical volumes. They also found that savings opportunities could be had by looking at areas where variation occurred, particularly post-acute care services spending.
However, “because patterns of variation were condition specific, condition-specific delivery innovations are likely needed. For example, savings in a bundled payment program might be maximized by reducing the use of skilled nursing facilities for major joint replacement,” they wrote.