Potentially Reversible Resident Factors Associated with Rejection of Care Behaviors

Abstract

To identify the potentially modifiable resident-level factors associated with rejection of care in nursing home (NH) residents. Secondary analysis of a 3.0 national field test to revise the Minimum Data Set (MDS). Seventy-one NHs in eight states. Three thousand two hundred thirty NH residents scheduled for MDS assessments from September 2006 through February 2007. The potentially mutable characteristics assessed were mood (Patient Health Questionnaire-9), delirium (Confusion Assessment Method), delusions, hallucinations or illusions, hearing impairment, vision impairment, pain severity, and infection diagnoses. Characteristics considered as covariates were cognition, communication abilities, and impairment in activities of daily living. Of 3,230 residents assessed, 312 (9.7%) had demonstrated rejection of care in the preceding 5 days. In multiple regression analysis adjusted for covariates, rejection of care was associated with delusions (odds ratio (OR)=3.9; 95% confidence interval (CI)=2.5-6.0), delirium (OR=1.8, 95% CI=1.3-2.4), minor (OR=2.1, 95% CI=1.5-2.8) and major (OR=2.3, 95% CI=1.5-3.4) depression, and severe to horrible pain (OR=1.6, 95% CI=1.1-2.3). Infection diagnoses were not significant in bivariate analysis. Hallucinations or illusions, mild to moderate pain, and hearing and vision impairment were not significant in multiple regression analysis. In this population, delirium, delusions, depression, and severe pain were associated with rejection of care, suggesting that some care rejection behaviors may resolve with appropriate interventions for the identified target conditions if the associations observed are causal.