Hospice care in nursing homes: Is site of care associated with visit volume?

OBJECTIVES: To determine factors associated with hospice visit volume and to examine whether visit volume differs by nursing home (NH) versus non-NH setting. DESIGN: Retrospective cohort study. SETTING: Twenty-one hospices across seven states under the ownership of one parent provider. PARTICIPANTS: Hospice patients from October 1998 through September 1999 in NH (n = 9,460) and non-NH (n = 15,484) settings. MEASUREMENTS: Data were from the provider's centralized information system. Average daily visit volume was the number of visits divided by the number of hospice routine home care days (days not in hospice inpatient or continuous home care). Multivariate logistic regression tested the association between site of care and an individual's probability of having average daily visits above the sample median. RESULTS: Average daily visits standard deviation were 1.1 +/- 1.1 for NH and 1.2 +/- 1.3 for non-NH hospice patients. Site of care was not significantly associated with having an average daily visit volume above the sample median, but patients in NH settings had a lower probability of having a nurse average daily visit volume above the median (adjusted odds ratio (AOR) = 0,59, 95% confidence interval (CI) = 0.46-0.74) and a greater probability of having social worker (AOR = 2.46, 95% CI = 1.87-3.24), aide (AOR = 1.97; 95% CI = 1.11-3.48), and clergy (AOR = 3.23, 95% CI = 2.21-4.44) average daily visits above the median than those in non-NH settings. CONCLUSION: A different mix, not volume, of services appears to be used to address the physical, psychosocial, and spiritual needs of hospice patients/families who reside in NH settings than of those in non-NH settings.