Reaction time, cardiorespiratory fitness and mortality in UK Biobank: an observational study

Intelligence has previously been associated with mortality, although it is unclear whether associations are independent of related factors such as information processing or cardiorespiratory fitness. We investigate whether fluid intelligence, reaction time and cardiorespiratory fitness are independently associated with mortality within the general population. UK Biobank recruited adults across England, Scotland and Wales, March 2006 to July 2010. 54019 participants (women 52%) with complete data were included. Those who died in the first year of follow-up (n=58) were excluded. Fluid intelligence was measured as the number of correct answers to a two minute logic/reasoning-test, reaction time was measured as average time taken to respond to matching symbols on a computer screen and cardiorespiratory fitness was measured through a sub-maximal exercise test. Associations with mortality were assessed by Cox-proportional hazard models adjusted for age, sex, ethnicity, social deprivation, cancer and non-cancer illnesses, medications, employment, education, smoking, BMI, diet, sleep, and physical activity. Over 5.8 years of follow-up there were 810 deaths. Higher intelligence (hazard ratio [HR] per SD = 0.91; 95% CI 0.84, 0.99), faster reaction time (HR per SD = 0.92; 0.85, 0.98)) and higher fitness (HR per SD = 0.85; 0.78, 0.93) were associated with a higher risk of mortality after adjustment for each other and other covariates. No interaction was observed between fluid intelligence and reaction time (p=0.147) or between fluid intelligence and cardiorespiratory fitness (p=0.238). In conclusion, fluid intelligence, reaction time and cardiorespiratory fitness were associated with mortality with estimates persisting after adjustment for each other and multiple risk factors.