Happiness and unhappiness have no direct effect on mortality

822 www.thelancet.com Vol 387 February 27, 2016 What defi nes a good life? If in answering this question you included happiness in your list, you are not alone. Indeed, the pursuit and enjoyment of happiness is a common goal and desire in life for most people. Adults of all ages, including those in old age, frequently report the experience of happiness as a determinant of a good life. Since both happiness and health are crucial aspects of quality of life, medical work about the potential positive eff ects of happiness on a person’s health and longevity is a growing area that has received increasing attention in the past decade. In The Lancet, Bette Liu and colleagues use data from a cohort of women in the UK Million Women Study, mean age 60 years, to examine whether happiness was associated with good health and with reduced mortality risk after an average follow-up of about 10 years. The strongest correlates of unhappiness were treatment for depression and anxiety (odds ratio [OR] 0·224 [99% group-specifi c CI 0·218–0·229]) and self-reported poor health (OR 0·298 [0·293–0·303]). In crude analyses of 719 671 women without chronic health disorders from the cohort, unhappiness was signifi cantly associated with an increased risk of all-cause mortality (age-adjusted rate ratio [RR] 1·29, 95% CI 1·25–1·33). However, in multivariate regression models adjusted for age, personal characteristics, treatment for illnesses, and self-reported health (the key factor) there was no signifi cant association (0·98, 0·94–1·01). Other researchers have found that hedonic wellbeing (ie, a viewpoint that defi nes wellbeing through experiences of pleasure vs displeasure and that can be roughly summarised as happiness) is not a good predictor of mortality in women, when baseline levels of health and health behaviours had been taken into account. A Japanese cohort study (n=88 175) showed that high levels of enjoyment in life appeared to protect against cardiovascular mortality in men but not in women aged 40–69 years. However, another study with 97 253 women from the USA aged 50–79 years showed that high levels of optimism were associated with reduced mortality risk. Reviews and a meta-analysis on the associations between happiness (hedonic wellbeing, subjective wellbeing, or positive psychological wellbeing) and longevity support the notion that happy people live longer. Although most studies on this topic did not adjust their analyses for self-reported health, they suggest that the associations between happiness and longevity are strongest among healthy individuals. Liu and colleagues’ main fi nding does not support this association, since no signifi cant relation was shown between happiness and mortality risk within the subgroup of people reporting good or excellent health (indeed, Liu and colleagues’ analysis of only people reporting excellent health showed all-cause mortality risk to be slightly higher in the unhappiest individuals compared with the happiest). The happiness–mortality association seems to be, to some extent, sex-specifi c, with higher positive eff ects reported in men than in women. Previous research shows diff erent profi les of psychological wellbeing according to sex; for instance, women’s wellbeing would rely more on positive interpersonal relations than men’s. Moreover, in older adults, the diff erence between men and women on reported wellbeing and happiness increases with advancing age. Therefore, men and women probably defi ne happiness diff erently, which might explain, at least partially, sex-specifi c diff erences for the associations between happiness and medical outcomes, including mortality. Further qualitative research that allows separate content analysis for men and women about happiness across diff erent age ranges is needed to improve understanding of the complex concept of happiness. Although mortality is one of the most well documented outcomes in reports on happiness, happiness has also been shown to be associated with other medical disorders, particularly a reduced risk of both incident cardiovascular diseases and disability levels (see Diener and Chan for a review). Research about happiness of older adults should focus on medical outcomes other than mortality, particularly the incidence and prognostic value of disabling diseases and disability, the most important clinical issues faced by elderly populations. For example, an important gap exists in knowledge about the potential associations of happiness with the incidence of cognitive decline and dementia; indeed, happiness is associated with healthy lifestyles, such as exercise and abstinence from smoking, which, in turn, are protective factors against dementia. Therefore, it is plausible to suggest that happiness could be associated with a reduced risk of incident dementia. Moreover, further research from a Happiness and unhappiness have no direct eff ect on mortality