These results appear to be significantly better than the results of the Medicare Physician Group Practice Demonstration in which the best performing (Marshfield) saved an average of $363 per beneficiary per year out of the much larger cost base of full FFS Medicare eligible. See below (note this analysis assumes PY2 beneficiaries were reflective for the full five years).
Public reporting on the BSC pilot over the past six months suggests four sources of savings:
- shortened hospital stays (half a day off the average LOS)
- reduced readmissions (somewhere between 15-22% depending on the report)
- 85 patients which were seen in an out-of-network emergency department and were stabilized and transferred to a Catholic Healthcare West hospital for treatment
- a 13% reduction in elective procedures such as bariatric surgery and elective knee surgery
Taking these reports at face value and applyng some generic assumptions on costs for these types of events, I estimate that of the $480 PMPY savings:
- 20-30% came from shortened stays (assuming per diem charge)
- 10-15% from reduced readmissions
- 8-12% from the transfer to in-network hospitals
Reductions in LOS and readmissions and increased network adherence seem liked solid, sustainable savings. But this leaves 45-60% of the savings to be explained by the 13% reduction in elective procedures (or some other un-discussed savings source). Some of these reductions in elective procedures may really be the result of improved patient counseling, but part is likely tied to the economy. To the extent that financing co-pays or concerns about lost time from work made patients more receptive to counseling in 2010, these savings might not be sustained as the economy turns around.
Even so, a sustained 6-8% reduction in employer costs (that portion of savings not tied with elective procedures) is pretty good for a first time out (2010 was the first year the ACO was in operation).
It is also interesting that of the $20M saved, only about 25% or $5M was shared out to the three partners (compare that vs. the 80% share rate for the Medicare demonstration and the 50-60% share rate top performing providers can achieve under CMS draft rules for ACOs). For Hill Physicians, for example, this implies an incremental $34 per patient. Hardly an inspiring return (even if, as suggested in the public statements, little incremental investment was required by the partners).
These may be partially offset by enhanced share of the care provided the participating patients (i.e. the participating members may have more consistently visited Catholic Healthcare West beyond the 85 transferred patients or have spent more time with Hill Physicians physicians vs. going out of network or pursuing care not offered by Hill). But these are probably also small numbers.
My take: the partnership is not playing a “single stage” game here: the pay-off for the partners will be in share gain within CalPERS (e.g., either in BSC vying for a new CalPERS contract) or in expanding provider share among the membership by being the centerpiece of lower cost coverage options. Aligning incentives around share growth creates an interesting model for rewarding ACO participation.