The effect of resident characteristics on satisfaction with care and life in skilled nursing facilities: A hierarchical mode

In examining the relationship between resident and facility characteristics and the satisfaction of nursing home residents, many previous satisfaction studies have typically considered only a small number of variables, when it is likely that a variety of variables impact satisfaction scores. With few exceptions, published studies of quality and patient satisfaction have relied on conventional statistical analysis procedures and have ignored the effects of the inter-correlation of data typical in organizational analyses. This study developed a conceptual framework that explains the differences within persons and between facilities of resident total satisfaction scores in nursing homes and explored the relationship between resident characteristics and their effect on satisfaction scores while accounting appropriately for facility effects. The data set included demographic and satisfaction data collected from 834 residents in 32 New Jersey skilled nursing facilities.;Conventional statistical techniques showed that resident choice/autonomy, advancing age, and female gender were significantly and positively related to resident total satisfaction while resident self-reported health and cognitive ability were significantly and negatively related to resident total satisfaction. Although Student t-statistic analyses revealed differences between the resident total satisfaction means for non-Hispanic white race (versus non-white), a higher level of education completed, greater frequency of pain, greater intensity/severity of pain, and a higher level of verbal and mental distress, these findings were not confirmed in the regression analyses.;Hierarchical linear modeling techniques identified differences between the conventional and multilevel bivariate analyses and the model testing analyses for non-Hispanic white race. Comparison of five imputation methods revealed differences between the data sets in the significance of frequency of pain. The reduction of the conceptual model to a single focal relationship demonstrated the ability of the model to account for the majority of the variance in resident total satisfaction.;These findings indicate that the perspectives of nursing home residents, nursing home professionals, and state regulatory bodies may differ considerably in the identification of which aspects of care and life are most important to resident satisfaction. Facility policies and staff training that maximize resident participation in daily decision-making should be encouraged in federal-, state- and facility-directed quality improvement efforts.