Title

Hospital Readmission Risk Prediction Models: A Literature Review

Student Author(s)

Kathryn Shanklin

Faculty Mentor(s)

Rob Schwartz, MBA, MHA, MS/MIT, Tanya Young, MBA, BSN, RN, and Emilie Dykstra Goris, PhD, RN

Document Type

Poster

Event Date

4-15-2016

Abstract

Limiting 30-day hospital readmissions, a major cost factor for healthcare organizations, requires identifying high-risk patients so that nursing interventions aimed at reducing readmission may be implemented. The identification of key variables contributing to readmission is essential for the development of effective 30-day readmission risk prediction models. The purpose of this study was to complete a comprehensive literature review to examine existing published readmission risk-prediction models and the unique variables utilized in each model. Orem’s Self-Care Deficit Nursing Theory served as a foundation for this research as it demonstrates the importance of identifying self-care deficits so that specialized nursing assistance can be provided where most needed. An extensive review of the literature from September, 2010 through September, 2015 was completed with Cochrane Library, MEDLINE, and CINAHL databases using the keywords readmission prediction model, readmission algorithm, and risk assessment tool. The literature search yielded 16 articles that met the inclusion criteria. The most common variable categories included in the algorithms were demographics, diagnosis, number of admissions, procedures, laboratory values, length of stay, comorbidities, and socioeconomic indicators. Demographics and health literacy were the most promising readmission risk prediction variable categories based on this review. Limitations for this study include the use of published literature when many hospitals use unpublished hospital-specific algorithms. Additionally, specific variables are rarely included in the literature so this study evaluated broad variable categories. A comprehensive understanding of readmission prediction models and variables can direct future nursing research by contributing to the development of increasingly effective models as well as identifying pertinent variables indicating patient readmission risk.

This document is currently not available here.

Share

COinS