Objective: To investigate the association between cooking behaviour and long-term survival among elderly Taiwanese.
Design: Cohort study. The duration of follow-up was the interval between the date of interview and the date of death or 31 December 2008, when censored for survivors. Information used included demographics, socio-economic status, health behaviours, cooking frequencies, physical function, cognitive function, nutrition knowledge awareness, eating out habits and food and nutrient intakes. These data were linked to death records. Cox proportional-hazards models were used to evaluate cooking frequency on death from 1999 to 2008 with related covariate adjustments.
Setting: Elderly Nutrition and Health Survey in Taiwan, 1999-2000.
Subjects: Nationally representative free-living elderly people aged ≥65 years (n 1888).
Results: During a 10-year follow-up, 695 participants died. Those who cooked most frequently were younger, women, unmarried, less educated, non-drinkers of alcohol, non-smokers, without chewing difficulty, had spouse as dinner companion, normal cognition, who walked or shopped more than twice weekly, who ate less meat and more vegetables. Highly frequent cooking (>5 times/week, compared with never) predicted survival (hazard ratio (HR) = 0·47; 95 % CI, 0·36, 0·61); with adjustment for physical function, cognitive function, nutrition knowledge awareness and other covariates, HR was 0·59 (95 % CI, 0·41, 0·86). Women benefited more from cooking more frequently than did men, with decreased HR, 51 % v. 24 %, when most was compared with least. A 2-year delay in the assessment of survivorship led to similar findings.
Conclusions: Cooking behaviour favourably predicts survivorship. Highly frequent cooking may favour women more than men.