Background Research suggests television viewing time may be associated with incident obesity and central obesity in young adults. No study has investigated these associations in older English adults. The aim of this study was to investigate longitudinal associations between television viewing time and incident obesity and central obesity in a sample of older English adults. Analyses of data from the English Longitudinal Study of Ageing. At baseline (2008), participants reported their television viewing time. Research nurses recorded obesity and central obesity by body mass index and waist circumference, respectively, at four year follow-up. Associations between television viewing time and incident obesity (BMI > 30 kg/m2) and central obesity (waist >102 cm men; > 88 cm women) at four year follow-up were examined using adjusted logistic regression. Participants gave full written informed consent to participate in the study and ethical approval was obtained from the London Multicentre Research Ethics Committee. Results A total of 3777 initially non-obese participants (aged 64.8 ± 8.6 yrs, 46.4% male) were included in the analyses using BMI as an outcome and 2947 for the analyses using waist circumference. No significant associations were found between television viewing time and incident obesity. A significant association was found between watching ≥6 hrs/d of television (compared to <2 hrs/d) and central obesity (Odds Ratio 1.48; 95% confidence interval 1.07 to 2.03) after adjustment for covariables including physical activity. Conclusions In this sample of older community dwelling English adults greater television viewing time was associated with incident central obesity, but not total obesity when measured by BMI. Interventions to reduce the incidence of central obesity in this age group that focus on reducing TV time, as well as targeting other health behaviours (eg, increasing physical activity levels, improving dietary intake) might prove useful.