Current Criteria May Mislabel Many Hospital Readmissions

Greg D. Sacks, MD, MPH Department of Surgery, David Geffen School of Medicine at UCLA MedicalResearch.com Interview with:
Greg D. Sacks, MD, MPH
Department of Surgery, David Geffen School of Medicine at UCLA

MedicalResearch: What are the main findings of this study?


Dr. Sacks:
This study evaluated the all-cause readmissions measure developed by the Centers for Medicare and Medicaid Services to penalize hospitals for unplanned readmissions. By evaluating readmissions of surgical patients at a single academic medical center, we found that the readmissions measure was able to identify only a third of the planned readmissions and mislabeled the remaining two thirds of planned readmissions as unplanned. This discrepancy was a result of the measure’s reliance on administrative claims data, which disagreed in 31% of cases with clinical data abstracted from the patient’s chart. Also, almost a third (27%) of the readmissions in this study were for reasons unrelated to the original hospitalization.

MedicalResearch: What should patients and clinicians take away from this report?

Dr. Sacks: Hospital readmissions after surgery are complex, are caused by a wide range of clinical and social factors, and do not always represent poor care quality. Policy measures that penalize hospitals should ensure that the data used are reliable and do not penalize hospitals for factors that are beyond their control. Readmissions unrelated to the index hospital stay should be considered as a separate entity and should not be included in measures that aim to grade hospital performance.

MedicalResearch: What future research do you recommend as a result of this work?

Dr. Sacks: Researchers should develop reliable ways to identify and measure unplanned, preventable, and related readmissions. These readmissions are most likely to represent lapses in care for which hospitals can be held accountable and would offer useful feedback to guide hospitals in their quality improvement efforts.

Citation: