Predictive Factors For Adverse Outcomes In Patients With A Pressure Ulcer

Purposes/Aims:  The purpose of this study is to identify factors that predict adverse outcomes in the patient with a pressure ulcer (PrU).  A retrospective observational cohort study design was used, including data from a large abstract database of patients discharged from the hospital in Washington State between from 2012 and 2013.  This study addresses the following aims for patients hospitalized with sepsis, pneumonia, COPD, heart failure, AMI, CVA, and AKI:  1) To compare rates of comorbidities between PrU and PrU-free patients by PrU stage; (2) To compare rates of adverse outcomes between PrU and PrU-free patients by PrU stage, and (3) To identify predictive factors for adverse outcomes in the hospitalized patient with a PrU.

Rationale/Conceptual Basis/Background: Pressure ulcers (PrU) are a significant safety threat in the hospitalized patient, particularly those who are debilitated with multiple comorbidities.  Despite the current use of risk assessment scales, PrUs occur in 4.5% of all hospitalized patients and 8-22% of ICU patients.  PrUs increase the risk for poor outcomes including in-hospital mortality, longer hospital lengths of stay, discharge to a setting other than home, greater costs of care, decreased quality or life, and greater post hospital stay morbidity. However, it is not known how the outcomes in the patient with a PrU are affected by interactions of key comorbidities or PrU stage. Additionally, the National Pressure Ulcer Advisory Panel (NPUAP) includes factors predictive of pressure ulcer outcomes as a research priority.

Methods: Using a retrospective observational cohort study design, data will be extracted from the Comprehensive Hospital Abstract Reporting System (CHARS) database in Washington State. The CHARS dataset includes electronically abstracted encounter information from each acute care hospital discharge in Washington State.  In order to assemble representative patients at a common point in the course of their diseases, the most common primary medical diagnoses for patients 45+ years of age were identified. The study population will be patients 45+ years of age discharged from the hospital during 2012 and 2013 with a primary ICD-9 discharge diagnosis of sepsis, pneumonia, COPD, heart failure, acute myocardial infarction (AMI), CVA, or acute kidney injury (AKI).  Binary logistic regression models will be used to examine differences in risks of in-hospital death and extended length of stay between the PrU and PrU-free groups controlling for age, sex, payer, comorbidities, and length of stay. Comorbid conditions will be derived using the Elixhauser method.

Results:  Expected results will provide insight as to the outcomes of the patient with a PrU stratified by PrU stage and comorbidity.

Implications: This project will provide a deeper understanding of PrU risk factors by stage and by comorbidity. The expected outcome of this research is to increase the precision of PrU risk assessment and ultimately reduce the occurrence of HAPUs in at risk patient populations.