Impact of availability of an inpatient hospice unit on the parent hospital's quality of palliative care for Taiwanese cancer decedents, 2001-2006.

Hospice care has increasingly been shown to affect quality of palliative care at both the individual and institutional levels. However, an institutional effect has only been addressed in single comprehensive cancer centers/selected community hospitals.To investigate the impact of an inpatient hospice unit on the parent hospital's quality of palliative care.This was a retrospective cohort study using administrative data from the entire population of 204,850 Taiwanese pediatric and adult cancer patients who died in 2001-2006. Outcome variables were adjusted by multivariate logistic regression for five groups of confounding variables: 1) patient demographics and disease characteristics, 2) primary hospital characteristics, 3) primary physician specialty, 4) health care resources at the hospital and regional levels, and 5) historical trend.Taiwanese cancer patients who received primary care in a hospital with an inpatient hospice unit (whether or not they received hospice care) were significantly less likely to be intubated (adjusted odds ratio [AOR]: 0.71; 95% confidence interval [CI]: 0.58, 0.86) and use mechanical ventilation support (AOR: 0.70; 95% CI: 0.56, 0.87) in their last month of life. They also were more likely to use hospice care before death (AOR: 3.51; 95% CI: 1.57, 7.86). Furthermore, if they used hospice care, they tended to be referred earlier than cancer patients being cared for in a hospital without an inpatient hospice unit.Integrating both acute care and palliative care approaches to caring for terminally ill cancer patients in the same hospital may influence the quality of palliative care throughout the hospital as evidenced by our findings that these patients have lower likelihood of being intubated with mechanical ventilation support in the last month of life, greater propensity to receive hospice care in the last year of life, and a trend toward earlier referral to hospice care. The generalizability of these results may be limited to patients who died of a noncancer cause and by the two groups not being exactly matched for patients' characteristics.