PLoS ONE (Jan 2017)
Changes in the living arrangement and risk of stroke in Japan; does it matter who lives in the household? Who among the family matters?
Abstract
Previous studies have suggested associations of family composition with morbidity and mortality; however, the evidence of associations with risk of stroke is limited. We sought to examine the impact of changes in the household composition on risk of stroke and its types in Japanese population. Cox proportional hazard modelling was used to assess the risk of incident stroke and stroke types within a cohort of 77,001 Japanese men and women aged 45-74 years who experienced addition and/or loss of family members [spouse, child(ren), parent(s) and others] to their households over a five years interval (between 1990-1993 and 1995-1998). During 1,043,446 person-years of the follow-up for 35,247 men and 41,758 women, a total of 3,858 cases of incident stroke (1485 hemorrhagic and 2373 ischemic) were documented. When compared with a stable family composition, losing at least one family member was associated with 11-15% increased risk of stroke in women and men; hazard ratios (95% confidence interval) were 1.11 (1.01-1.22) and 1.15 (1.05-1.26), respectively. The increased risk was associated with the loss of a spouse, and was evident for ischemic stroke in men and hemorrhagic stroke in women. The addition of any family members to the household was not associated with risk of stroke in men, whereas the addition of a parent (s) to the household was associated with increased risk in women: 1.49 (1.09-2.28). When the loss of a spouse was accompanied by the addition of other family members to the household, the increased risk of stroke disappeared in men: 1.18 (0.85-1.63), but exacerbated in women: 1.58 (1.19-2.10). In conclusion, men who have lost family members, specifically a spouse have higher risk of ischemic stroke, and women who gained family members; specifically a parent (s) had the higher risk of hemorrhagic stroke than those with a stable family composition.