Home » PHIRNET Webinar » Peter Tanuseputro

Peter Tanuseputro

A population-based examination of interventions near the end-of-life and their effect on location of death and places of care

Date: April 10, 2015
Times: 13h00 to 14h15 Eastern Time (Montréal)
Presenter: Peter Tanuseputro
Relevance to PHIR: We use health administrative data to examine the effectiveness of population interventions provided near the end of life; namely, the effect of palliative home care visits and physician home visits. These two services form the foundation of home-based end-of-life care, yet we have little evidence if they improve care and/or reduce cost. We seek out to examine if these population interventions have an effect on the health and end-of-life experience in the Ontario population.
Objectives: First, we describe where Ontarians died and the number of days they spent in health care institutions in the last 90 and 30 days prior to death. Second, we examine the factors that affect where people died and their places of care, including an examination of palliative home care and physician home visits. Last, we plan to conduct a study to improve the identification of frail individuals nearing the end-of-life in the home care and long-term care settings.
Methods: We observed all deaths in the Ontario Registered Persons Database between April 1, 2010 and March 31, 2013.  For each decedent, we determined if they died in an acute care hospital, complex continuing care (CCC), rehabilitation, or long-term care (LTC) facility. The rest of the population were assumed to die in the community. We also determined how many days were spent in any of these institutions in the last 30 and 90 days prior to life.
Findings: There were 264,755 deaths in the study period, of whom 45.7% died in an acute care hospital, 7.7% in CCC, 0.2% in rehab, and 17.4% in LTC. Twenty-nine percent of the population died outside of these institutions. In the last 30 and 90 days of life, not including LTC, a total of 9 and 16 days, on average, were spent in these institutions, respectively. In multivariable models that adjusted for age, sex, income quintile, rurality, number of chronic conditions, and rostering to a family physician, receiving palliative home care reduces the risk of dying in an institution (not including LTC) by 46%. There was a dose response relationship between the number of physician home visits and the lowering of risk of dying in an institution, ranging from a reduction of 55% for one home visits to 75% for 7 or more visits. There was a similar relationship for number of days spent in institutions in the last 90 days for these interventions, with palliative home care reducing institution days by 7 days and physician visits by 3 to 7 days (for 1 and 7+ home visits, respectively). Less than one-third of the population, however, ever received palliative home care of a physician visit in their last year of life.
PHIR Questions: 1) What is the effect of current interventions provided to populations nearing the end of life in improving outcomes such as location of death and places of care? 2) Can we improve the how front line workers in the home care and long-term care settings identify individuals who may be nearing the end of life, and provide services that match their needs and values?