Health Policy$ense

Not in the Stars: Hospital Readmissions Not Linked to SNF Performance Measures

In a study published today in JAMA, LDI Senior Fellows Mark Neuman and Rachel Werner cast doubt on the strategy of reducing hospital readmissions by preferentially choosing skilled nursing facilities (SNFs) with high quality ratings on Nursing Home Compare, Medicare’s five-star rating system.

Rachel Werner, Mark Newman
LDI Senior Fellows Mark Neuman, MD and Rachel Werner, MD, PhD

The study included 1.5 million Medicare hospital discharges to SNFs over one year, 23.3% of which led to readmission or death within 30 days. The researchers looked at five SNF performance measures: star ratings for staffing intensity and state inspections, and three clinical measures (percentage of postacute residents with delirium, new or worsening pressure ulcers, and reporting moderate or severe pain). In unadjusted analyses, discharges to SNFs with better staffing and inspection ratings, and lower rates of pressure ulcers, had lower risks of 30-day readmission or death. Paradoxically, admission and death rates were lower in SNFs with higher percentages of residents with delirium and pain.

But in analyses that adjusted for patient factors, SNF facility characteristics, and the discharging hospital, quality measures were not consistently associated with lowered risks. SNFs with best inspection ratings (9.8% of all SNFs) had slightly lower risks than SNFs with the worst inspection ratings, but risks did not differ according to staffing ratings or percentages of residents with delirium or in pain. The authors note:

In a setting of an average readmission or death rate of 23%, we found that by choosing an SNF in the highest vs. lowest category of facility inspection ratings, hospitals might expect at most an absolute reduction in their readmission rates of 0.7 percentage points, or a relative reduction of just over 3%. Such a reduction may be meaningful to both the hospital in terms of their finances and the patients whose readmissions would be potential prevented; however, our findings suggest that there is significant variation in rates of readmission that is not explained by currently available performance measures.

Without other strategies, the authors conclude, relying on existing SNF performance measures will not likely lead to large-scale reductions in hospital readmissions.

These results could not be more timely, as Medicare has fined a record number of hospitals in 2014 for excessive readmission rates. And in response to criticisms that its star ratings rely on unverified, self-reported data, in 2015 Medicare will improve its Nursing Home Compare system through focused audits and payroll-based staffing reporting. Whether these changes produce more reliable performance measures, and whether these measures can be used to influence hospital readmission rates, remains unknown.

Click here to read the Penn Medicine News Release.