Can Braden Score Predict Outcomes for Hospitalized Heart Failure Patients?

Abstract

Background: Braden score is a routine assessment of pressure ulcer risk hypothesized to identify the frail phenotype.

Objectives: To investigate the predictive utility of the Braden score on outcomes of inpatients with heart failure (HF).

Design: Retrospective cohort study.

Setting: An academic medical center between January 1, 2012 and June 30, 2013.

Participants: 642 inpatients with a primary diagnosis of HF (ICD-9 428).

Measurements: The primary predictor was Braden score. Primary outcome was 30-day mortality. Additional outcomes included 30-day readmission, length of stay (LOS), and discharge destination. Multivariable methods were used to determine the association between the primary predictor and each outcome adjusted for patient demographics and clinical variables.

Results: Mean admission and discharge Braden scores were 19.5 ± 2.3 (SD) (range = 9-23) and 20.0 ± 1.9 (range = 11-23), respectively (P < .0001). Mean age was 61.8 ± 16.2 years (range = 19-101). The 30-day mortality rate was 4.4%, 30-day readmission rate was 16.2%, mean LOS was 7.0 ± 8.7 days, and 78.2% were discharged home. After adjustment, higher (better) Braden score was significantly associated with decreased 30-day mortality (discharge Braden AOR 0.81 (95% CI 0.66-0.996)), and decreased average LOS (admission Braden β -0.52 days (P = .0002)). Higher discharge Braden score was significantly associated with discharge to home (AOR 1.66 (95% CI 1.42-1.95)). Braden score was not significantly associated with 30-day readmission.

Conclusion: Braden score is an independent predictor of mortality, LOS, and discharge destination among inpatients with HF. Further exploration of the use of Braden scores to identify inpatients who might benefit from specialized intervention is warranted.

Keywords: Braden Scale; frailty; heart failure; mortality; risk prediction.