Age and gender variation in the impact of household structure on elderly mortality

Background There is little information about the impact of household structure and composition on elderly mortality in developing countries. This study examines the impact of relationship to head of household, and the presence of co-resident spouses and sons on elderly mortality in rural Bangladesh with a particular focus on age and gender differences. Methods A total of 9365 individuals aged greater than or equal to 60 at baseline (5128 males and 4237 females) in the Matlab Surveillance area in rural Bangladesh were followed for a period of 8 years (1974-1982) with all predictors (the presence of a spouse, one or more co-resident adult sons, relationship to head of household, household economic status, age and disability status) being measured at the beginning of follow-up. Cox proportional hazard models were used in the analysis. Results Being the head of household had a significant impact on reducing elderly mortality for both men and women. The presence of a spouse reduced mortality for all elderly men but had a significant beneficial impact only on women whose husbands were heads of households. Finally the presence of one or more coresident adult sons reduced mortality for elderly women but not for elderly men. For all three of the above predictors there was a decline in effect with the age of the elderly. Conclusions Relationship to head of household and the presence of spouses and sons have powerful impacts on reducing mortality for elderly men and women in rural Bangladesh with the effects varying significantly by gender and age. Furthermore, individual rather than joint access to material resources is an important determinant of elderly mortality.