Correlates of care quality in long-term care facilities: a multilevel analysis.

Hierarchical modeling was employed to explore correlates of the quality of care provided in long-term care facilities. For this purpose, a multilevel analysis offers two advantages over traditional analytical approaches. First, it accounts for the correlated nature of data recorded on multiple residents from the same facility. Second, it enables the investigators to study the influence of both resident and facility characteristics on care quality. The analysis was performed on data from 301 residents randomly sampled from 88 facilities located in the Province of Quebec, Canada. Results revealed that the presence of cognitive deficits was the strongest correlate of the quality of care provided to a resident. However, this relationship was found to vary significantly across facilities. Four facility-level variables were found to influence the relationship between cognitive functioning and care quality: the number of external collaborators the facility has, the type of training the facility manager has, the size of the facility, and the age distribution of its clientele. From these results, we suggest means to improve the quality of care provided to cognitively impaired older adults living in long-term care facilities.