Background: The mechanisms explaining the positive relationship between television (TV) viewing and body mass index (BMI) are unclear. ‘Mindless eating’ and ‘physical activity displacement’ theories have been suggested, but have not been tested longitudinally among young adults. This study aimed to determine whether longitudinal associations between young adults’ TV viewing and BMI are explained by changes in TV-related food and beverage consumption (FBC) and/or leisure-time physical activity (LTPA) over 5 years among young adults.

Methods: A cohort of young Australian adults (n = 1068) was assessed in 2004–6 (T1) and 2009–2011 (T2), height and weight were measured (T1) or self-reported (T2), and participants self-reported TV viewing time (hours/day), weekly TVrelated FBC and LTPA (mins/week). Linear regression was used to examine direct pathways between TV viewing and BMI, adjusting for TV-related FBC and LTPA to examine indirect pathways.

Results: The association between TV viewing time and BMI (β: 0.41, 95% CI 0.03, 0.78 for > 1-h increase in TV viewing/ day) was not explained by TV-related FBC (β: 0.37, 95% CI -0.18, 0.91) or LTPA (β: 0.38, 95% CI -0.17, 0.93) hypotheses. Increased TV-related FBC was associated with increased TV viewing (0.39 ± 1.54 h/day) and greater increases in BMI (0. 92 ± 2.28 kg/m² , p= 0.16). LTPA increases were not associated with changes in TV viewing (− 0.07 ± 1.42 h/day), and increases in BMI were smallest when LTPA increased (0.44 ± 2.25 kg/m² ) and greatest when LTPA decreased (0.82 ± 2. 30 kg/m²) (p = 0.13).

Conclusions: Factors other than changes in TV-related FBC or LTPA may explain the longitudinal relationship between TV viewing and increasing BMI among young adults. Findings confirm that TV viewing is a risk factor for weight gain in young adults but the underlying causal mechanisms remain unclear.