Patterns of sitting and mortality in the Nord-Trøndelag health study (HUNT)

BACKGROUND: Current evidence concerning sedentary behaviour and mortality risk has used single time point assessments of sitting. Little is known about how changes in sitting levels over time affect subsequent mortality risk. AIM: To examine the associations between patterns of sitting time assessed at two time points 11 years apart and risk of all-cause and cardio-metabolic disease mortality. METHODS: Participants were 25,651 adults aged > =20 years old from the Nord-Trøndelag Health Study with self-reported total sitting time in 1995-1997 (HUNT2) and 2006-2008 (HUNT3). Four categories characterised patterns of sitting: (1) low at HUNT2/ low at HUNT3, ‘consistently low sitting’; (2) low at HUNT2/high at HUNT3, ‘increased sitting’; (3) high at HUNT2/low at HUNT3, ‘reduced sitting’; and (4) high at HUNT2 /high at HUNT3, ‘consistently high sitting’. Associations of sitting pattern with all-cause and cardio-metabolic disease mortality were analysed using Cox regression adjusted for confounders. RESULTS: Mean follow-up was 6.2 years (158880 person-years); 1212 participants died. Compared to ‘consistently low sitting’, adjusted hazard ratios for all-cause mortality were 1.51 (95% CI: 1.28–2.78), 1.03 (95% CI: 0.88–1.20), and 1.26 (95% CI: 1.06–1.51) for ‘increased sitting’, ‘reduced sitting’ and ‘consistently high sitting’ respectively. CONCLUSIONS: Examining patterns of sitting over time augments single time-point analyses of risk exposures associated with high sitting time. Whilst sitting habits can be stable over a long period, life events (e.g., changing jobs, retiring or illness) may influence sitting trajectories and therefore sitting-attributable risk. Reducing sitting may yield mortality risks comparable to a stable low-sitting pattern. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12966-016-0457-8) contains supplementary material, which is available to authorized users.