The Relationships Between the Geriatric Practice Environment, Nursing Practice, and the Quality of Hospitalized Older Adults’ Care

PURPOSE: To test the relationships between the geriatric practice environment, geriatric nursing practice, and overall quality of care for older adults and their families as reported by nurses working in hospitals, while controlling for nurse and hospital characteristics. DESIGN: A cross-sectional tailored survey design was employed. A questionnaire was mailed to a randomly selected sample of nurses whose primary practice area was medicine, surgery, geriatrics, emergency, or critical care in acute care hospitals in Ontario, Canada. METHODS: Participants (N = 2,005) working in 148 hospitals responded to validated measures of the geriatric practice environment, geriatric nursing practice, overall quality of care for older adults and their families, and nurse and hospital characteristics. The relationships were tested using structural equation modeling. FINDINGS: Controlling for nurse and hospital characteristics, the geriatric practice environment had a statistically significant positive relationship of large magnitude with both geriatric nursing practice (β = 0.52) and overall quality of care (β = 0.92); however, the indirect relationship between the geriatric practice environment and overall quality of care, mediated by geriatric nursing practice, was not significant (β = -0.02). Final model fit was acceptable, with the root mean square error of approximation = 0.07, comparative fit index = 0.93, and Tucker-Lewis Index = 0.87. CONCLUSIONS: A strong geriatric practice environment positively and directly influences geriatric nursing practice and overall quality of care for older adults and their families but does not appear to influence overall quality of care indirectly through geriatric nursing practice. CLINICAL RELEVANCE: The results can be used as the basis for promoting practice environments that support overall quality of care and geriatric nursing practice in acute care hospitals.