Get access

A meta-analysis of hospital 30-day avoidable readmission rates

Authors

  • Carl van Walraven MD MSc FRCPC,

    Corresponding author
    1. Associate Professor, Faculty of Medicine, University of Ottawa, Ottawa, Ontario, Canada
    2. Senior Scientist
    3. Adjunct Scientist Institute for Clinical Evaluative Sciences, Ottawa, Ontario, Canada
    Search for more papers by this author
  • Alison Jennings MA,

    1. Methodologist, Clinical Epidemiology Program, Ottawa Hospital Research Institute, Ottawa, Ontario, Canada
    Search for more papers by this author
  • Alan J. Forster MD MSc FRCPC

    1. Associate Professor, Faculty of Medicine, University of Ottawa, Ottawa, Ontario, Canada
    2. Senior Scientist
    3. Adjunct Scientist Institute for Clinical Evaluative Sciences, Ottawa, Ontario, Canada
    Search for more papers by this author

  • This study was supported by the Department of Medicine, University of Ottawa. Carl van Walraven had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Dr Carl van Walraven, Ottawa Hospital Research Institute, ASB1-003 1053 Carling Avenue, Ottawa, ON K1Y 4E9, Canada. E-mail: carlv@ohri.ca

Abstract

Rationale and objectives  Urgent readmission to hospital is commonly used to measure hospital quality of care. Hospitals that measure the proportion of urgent readmissions judged avoidable need to know previously published rates for comparison. In this study, we generated a literature-based estimate for the proportion of 30-day urgent readmissions deemed avoidable for hospitals to use to gauge their performance in avoidable readmissions.

Methods  We searched the Medline and Embase databases to identify published studies that reported the proportion of 30-day urgent readmissions deemed avoidable. We then modelled the overall proportion of 30-day urgent readmissions deemed avoidable.

Results  We included 16 studies that used a wide variety of patients and a diverse range of methods to classify readmissions as avoidable. Studies reported a broad range for the proportion of urgent 30-day readmissions deemed avoidable. Overall, 848 of 3669 readmissions (23.1%, 95% confidence interval, 21.7–24.5) of 30-day urgent readmissions were classified as avoidable. This proportion varied significantly based on hospital teaching status and number of reviewers for each case [teaching hospitals: with one reviewer, 9.3% (4.2–19.3); with >1 reviewer, 21.6% (13.2–33.3); non-teaching hospital: with one reviewer, 32.2% (11.4–63.9); with >1 reviewer, 39.9% (37.6–42.2)]. Significant heterogeneity remained between studies even after clustering studies by these covariates.

Conclusions  Less than one in four readmissions were deemed avoidable. Health system planners need to use caution in interpreting all cause readmission statistics as they are only partially influenced by quality of care.

Ancillary