A longitudinal analysis of nursing home outcomes

Abstract

Objective: To investigate resident and facility attributes associated with long-term care health outcomes in nursing homes.

Data sources: Quarterly Management Minutes Questionnaire (MMQ) survey data for Medicaid case-mix reimbursement of nursing homes in Massachusetts from 1991 to 1994, for specification of outcomes and resident attributes. Facility attributes are specified from cost report data.

Study design: Multivariate logistic and "state-dependence" regression models are estimated for survival, ADL functional status, incontinence status, and mental status outcomes from longitudinal residence histories of Medicaid residents spanning 3 to 36 months in length. Outcomes are specified to be a function of resident demographic and diagnostic attributes and facility-level operating and nurse staffing attributes.

Principal findings: The estimated parameters for resident demographic and diagnostic attributes showed a great deal of construct validity with respect to clinical expectations regarding risk factors for adverse outcomes. Few facility attributes were associated with outcomes generally, and none was significantly associated with all four outcomes.

Conclusions: The absence of uniform associations between facility attributes and the various long-term care health outcomes studied suggests that strong facility performance on one health outcome may coexist with much weaker performance on other outcomes. This has implications for the aggregation of individual facility performance measures on multiple outcomes and the development of overall outcome performance measures.