Does quality of care matter in choosing a nursing home? Evidence from California

In choosing a nursing home (NH) consumers evaluate several quality of care dimensions. Some dimensions, such as hotel services, are observable, while clinical quality is mostly unobservable. We hypothesize that NH choice will be driven by observed rather than unobserved quality.

We examine nursing home choices in a period prior to publication of the federal NH Compare report card, i.e. when clinical quality was unobservable. We estimate the impact of facility and individual characteristics on consumers' choice, and assess the independent role of observable and unobservable quality. Observable quality is measured by hotel service-related deficiencies. Unobservable quality includes CMS clinical quality measures (QMs): decline in function, infections, pressure ulcers, physical restrains. MDS, OSCAR, NH cost reports and census data are used. The analytical sample includes 22,987 long-term care NH admissions in 2001 in California.

We find that Private pay consumers are much less likely to choose NHs with lower hotel quality (OR=0.83, p<0.01), than lower clinical quality (some QMs had no significant association with the choice and those with a significant association had odds ratios much closer to 1. Consumers' choice was also significantly (p=0.01) associated with: distance (OR=0.007); for-profit status (OR=0.87); and size (OR=1.27). Compared to private-pay, Medicaid consumers have a very limited NH choice.

Our findings suggest that in the absence of report cards, NH choice depends mostly on observable rather than clinical quality. Report cards can play an important role by revealing previously unobserved quality and potentially increasing the sensitivity of demand to clinical quality as well.