What's really your rate?
Meanwhile, across the seminar room another administrator fidgets and wears a look of dejection. Her rehospitalization rate at Foster Independence Rehab is 19% — and that is down from 24%! No matter what they do, they cannot seem to reduce it any further. Certainly, referring hospitals will avoid her facility in favor of Mossy Knoll.
Her director of nursing whispers, “But Mossy's acuity is the same as the early bird special at the local diner — we run a mini hospital!”
People are not all the same. Nursing homes often attract specific cohorts by specializing or through reputation. If you work in a nursing home, you understand this basic premise. However, those who don't might not understand the diversity of nursing homes, and the people they serve — and consequently, their outcomes. Acuity, diseases and other conditions all influence one's rehospitalization experience. This reality makes unadjusted comparisons almost meaningless. Currently, published nursing home readmission rates are dated and unadjusted.
We modeled the 30-day rehospitalization risk for individual Medicare admissions using MDS 3.0 data. Using this model to calculate an expected rate, we computed a case-mix adjusted rate: (observed rate/expected rate) x (national average rate). While the adjusted rehospitalization rate changed how one would perceive a nursing home, the rates varied widely; many facilities had adjusted rates over 25% or less than 15%.
Regardless of this adjustment, facility rates consider all people who return to hospital. However, by analyzing specific diseases — pneumonia, for example — one can glean significant insight. This disease-specific, case-mix adjusted analysis distinguishes quality improvement opportunities from marketing challenges. You can be excellent at managing pneumonia, but your case-mix adjusted rates for other types of patients may mask this area of excellence.