PT - JOURNAL ARTICLE AU - L Smith AU - B Gardner AU - M Hamer TI - Childhood correlates of adult TV viewing time: a 32-year follow-up of the 1970 British Cohort Study AID - 10.1136/jech-2014-204365 DP - 2015 Apr 01 TA - Journal of Epidemiology and Community Health PG - 309--313 VI - 69 IP - 4 4099 - http://jech.bmj.com/content/69/4/309.short 4100 - http://jech.bmj.com/content/69/4/309.full SO - J Epidemiol Community Health2015 Apr 01; 69 AB - Background To identify, using a longitudinal data set, parental and childhood correlates of adult television (TV) viewing time at 32-year follow-up. Method Data were derived from the 1970 British Cohort Study, a longitudinal observational study of 17 248 British people born in a single week of 1970. The present analyses incorporated data from the age 10 and 42-year surveys. When participants were aged 10 years, their mothers provided information on how often participants watched TV and played sports (never/sometimes/often), and parents’ own occupation, as well as height and weight. A health visitor objectively assessed participants’ height and weight at age 10. Thirty-two years later, when participants were aged 42 years, they reported their daily TV viewing hours (none/0≤1/1<3/3<5/≥5), physical activity and health status. Associations between putative childhood and parental correlates and adult TV viewing time were investigated using logistic regression. Results Valid data at both time points were available for 6188 participants. Logistic regression models showed that those who reported ‘often’ watching TV at baseline were significantly more likely to watch >3 h/days of TV at follow-up (OR 1.42, 95% CI 1.21 to 1.65), as were those whose father was from a lower socio-occupational class (intermediate, routine/manual) compared with managerial (OR 1.55, 95% CI 1.14 to 2.11; OR 2.05, 95% CI 1.47 to 2.87). Body mass index (BMI) at age 10 was inversely associated with high TV in adulthood (per unit increase; OR 0.93, 95% CI 0.90 to 0.96) although fathers BMI when the child was aged 10 was positively associated with high TV in adulthood (per unit increase; OR 1.04, 95% CI 1.02 to 1.06). Conclusions Findings suggest that childhood TV viewing time tracks into adulthood. Parents’ health behaviours and social position appear to be associated with their children's viewing habits, which may have important implications for the direction of future policy and practice. Specifically, findings support the case for early life interventions, particularly on socioeconomic inequalities, as a way of preventing sedentary behaviour in later life.