A Validated Method for Identifying Unplanned Pediatric Readmission

J Pediatr. 2016 Mar;170:105-12.e1-2. doi: 10.1016/j.jpeds.2015.11.051. Epub 2015 Dec 30.


Objective: To validate the accuracy of pre-encounter hospital designation as a novel way to identify unplanned pediatric readmissions and describe the most common diagnoses for unplanned readmissions among children.

Study design: We examined all hospital discharges from 2 tertiary care children's hospitals excluding deaths, normal newborn discharges, transfers to other institutions, and discharges to hospice. We performed blinded medical record review on 641 randomly selected readmissions to validate the pre-encounter planned/unplanned hospital designation. We identified the most common discharge diagnoses associated with subsequent 30-day unplanned readmissions.

Results: Among 166,994 discharges (hospital A: n = 55,383; hospital B: n = 111,611), the 30-day unplanned readmission rate was 10.3% (hospital A) and 8.7% (hospital B). The hospital designation of "unplanned" was correct in 98% (hospital A) and 96% (hospital B) of readmissions; the designation of "planned" was correct in 86% (hospital A) and 85% (hospital B) of readmissions. The most common discharge diagnoses for which unplanned 30-day readmissions occurred were oncologic conditions (up to 38%) and nonhypertensive congestive heart failure (about 25%), across both institutions.

Conclusions: Unplanned readmission rates for pediatrics, using a validated, accurate, pre-encounter designation of "unplanned," are higher than previously estimated. For some pediatric conditions, unplanned readmission rates are as high as readmission rates reported for adult conditions. Anticipating unplanned readmissions for high-frequency diagnostic groups may help focus efforts to reduce the burden of readmission for families and facilities. Using timing of hospital registration in administrative records is an accurate, widely available, real-time way to distinguish unplanned vs planned pediatric readmissions.

Publication types

  • Research Support, Non-U.S. Gov't
  • Validation Study

MeSH terms

  • Adolescent
  • Child
  • Child, Preschool
  • Female
  • Hospitals, Pediatric / statistics & numerical data*
  • Humans
  • Infant
  • Male
  • Patient Discharge / statistics & numerical data*
  • Patient Readmission*
  • Risk Factors
  • Tertiary Care Centers / statistics & numerical data*
  • Young Adult