Format

Send to

Choose Destination
J Pediatr. 2017 Jul;186:150-157.e1. doi: 10.1016/j.jpeds.2017.03.056. Epub 2017 May 2.

Adding Social Determinant Data Changes Children's Hospitals' Readmissions Performance.

Author information

1
Department of Pediatrics, University of Colorado School of Medicine, Aurora, CO. Electronic address: Marion.Sills@childrenscolorado.org.
2
Children's Hospital Association, Lenexa, KS.
3
Department of Pediatric Emergency Medicine, Children's Hospitals and Clinics of Minnesota, Minneapolis, MN.
4
Department of Pediatrics, University of Missouri-Kansas City School of Medicine, Kansas City, MO.
5
Department of Family and Community Medicine, University of California-San Francisco, San Francisco, CA.
6
Departments of Emergency Medicine and Pediatrics, University of Michigan, Ann Arbor, MI.
7
Children's Health System of Texas, University of Texas Southwestern, Dallas, TX.
8
Department of Pediatrics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA.
9
Department of Pediatrics, Baylor College of Medicine, Houston, TX.
10
Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH.

Abstract

OBJECTIVES:

To determine whether social determinants of health (SDH) risk adjustment changes hospital-level performance on the 30-day Pediatric All-Condition Readmission (PACR) measure and improves fit and accuracy of discharge-level models.

STUDY DESIGN:

We performed a retrospective cohort study of all hospital discharges meeting criteria for the PACR from 47 hospitals in the Pediatric Health Information database from January to December 2014. We built four nested regression models by sequentially adding risk adjustment factors as follows: chronic condition indicators (CCIs); PACR patient factors (age and sex); electronic health record-derived SDH (race, ethnicity, payer), and zip code-linked SDH (families below poverty level, vacant housing units, adults without a high school diploma, single-parent households, median household income, unemployment rate). For each model, we measured the change in hospitals' readmission decile-rank and assessed model fit and accuracy.

RESULTS:

For the 458 686 discharges meeting PACR inclusion criteria, in multivariable models, factors associated with higher discharge-level PACR measure included age <1 year, female sex, 1 of 17 CCIs, higher CCI count, Medicaid insurance, higher median household income, and higher percentage of single-parent households. Adjustment for SDH made small but significant improvements in fit and accuracy of discharge-level PACR models, with larger effect at the hospital level, changing decile-rank for 17 of 47 hospitals.

CONCLUSIONS:

We found that risk adjustment for SDH changed hospitals' readmissions rate rank order. Hospital-level changes in relative readmissions performance can have considerable financial implications; thus, for pay for performance measures calculated at the hospital level, and for research associated therewith, our findings support the inclusion of SDH variables in risk adjustment.

KEYWORDS:

healthcare quality measurement; pay for performance; performance measure; predictive model; public reporting; risk adjustment

PMID:
28476461
DOI:
10.1016/j.jpeds.2017.03.056
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center