Residents in a Remote Island Having Family Members in Distant Areas Showed Higher Preference for Place of End-of-Life Care: The Ajishima Study.

Introduction: To investigate the proportion of those having preferred place for end-of-life care among residents in a remote island and its association with family composition.

Methods: Cross-sectional questionnaire survey was conducted in Ajishima, an island 23 km away from the coast of Ishinomaki City, northeast of Japan. Between October 2017 and February 2018, the questionnaire was distributed to 288 eligible residents and 113 valid responses were analyzed. Primary outcome was whether the subjects had preferred place for end-of-life care. The explanatory variable was family composition defined as whether having family members inside or outside the island [none (In-/Out-), only inside the island (In+/Out-), only outside the island (In-/Out+), and both inside and outside (In+/Out+)]. Poisson regression analysis was used to calculate the prevalence ratios (PRs) and 95% confidence intervals (CIs) of showing preferred place in each group.

Results: The proportion of those having preferred place for end-of-life care was 72.6% in total. This rate significantly differed across family composition groups: 67.6%, 40.0%, and 82.9% for In+/Out+, In+/Out-, and In-/Out+ groups, respectively. The PR (95%CI) of having preferred place was 0.66 (0.33, 1.36) and 1.26 (1.01, 1.56) for In+/Out- and In-/Out+ groups, respectively, compared with In+/Out+ group.

Conclusions: This study showed that significantly higher preference for place of end-of-life care was seen among residents who had family members only outside the island compared with those who had families both inside and outside. Health care professionals should consider family compositions when initiating end-of-life discussion to residents in remote areas.