Author Interviews, Hospital Readmissions, JAMA, Pediatrics / 11.08.2015
Without Insurance Data, Hospitals Don’t Know Their True Readmission Rate
MedicalResearch.com Interview with:
Alisa Khan, MD
Pediatric hospitalist
Boston Children's Hospital and Instructor of Pediatrics
Harvard Medical School
Medical Research: What is the background for this study?
Dr. Khan: Patients can be readmitted to the same hospital they were discharged from or to a different hospital. In adults, readmissions to different hospitals make up about 20% of all readmissions. We don’t know a lot about how often different-hospital readmissions happen in children.
Insurance companies know hospitals’ true readmission rates (which include when a hospital’s patients are readmitted to the same hospital and when they are readmitted to a different hospital). However, hospitals don’t know their true readmission rates since they don’t have access to the full information that insurance companies have.
If hospitals don’t know their true rates, they may think they are doing better at preventing readmissions than they really (for instance, if all their discharged patients are simply being readmitted to a different hospital). Hospitals may also draw incorrect conclusions when they compare themselves to one another (like through benchmarking), and may not be able to predict whether they will be subject to penalties by insurers for having excessively high readmission rates.
Medical Research: What are the main findings?
Dr. Khan: We found that about 1 in 7 pediatric readmissions in New York over a 5-year period were to a different hospital than the hospital the patient was discharged from. The percentage of different-hospital readmissions varied by hospital and patient characteristics. Patients who were admitted to non-children’s hospitals, lower-volume hospitals, or urban hospitals had a higher chance of being readmitted to a different hospital, as did patients who were younger, white, privately insured, or who had certain chronic conditions (like mental health, neurologic, and circulatory conditions).
We also found a lot of variability in how much individual hospitals would underestimate their true readmission rates if they only used this incomplete same-hospital readmission info. Some hospitals would underestimate their true readmission rates by only 0.6 relative percentage points while others would underestimate them by 68 points.
(more…)