Change in lifestyle behaviors and diabetes risk: evidence from a population-based cohort study with 10 year follow-up

$\textbf{Background:}$ Promoting positive changes in lifestyle behavior in the whole population may be a feasible and effective approach to reducing type 2 diabetes burden, but the impact of population shifts of modifiable risk factors remains unclear. Currently most of the evidence on modifiable lifestyle behavior and type 2 diabetes risk on a population level comes from studies of between-individual differences. The objective of the study was to investigate the association and potential impact on disease burden for within-individual change in lifestyle behavior and diabetes risk. $\textbf{Methods:}$ Population-based prospective cohort study of 35,680 participants aged 30–50 at baseline in 1990–2003 in Västerbotten County, Sweden (follow-up until 2013). Five self-reported modifiable lifestyle behaviors (tobacco use, physical activity, alcohol intake, dietary fiber intake and dietary fat intake) were measured at baseline and 10 year follow-up. Lifestyle behaviors were studied separately, and combined in a score. Incident diabetes was detected by oral glucose tolerance tests. Multivariate logistic regression models and population attributable fractions (PAF) were used to analyze the association between change in lifestyle behavior between baseline and 10 year follow-up, and risk of incident diabetes. $\textbf{Results:}$ Incident diabetes was detected in 1,184 (3.3%) participants at 10 year follow-up. There was a reduced diabetes risk associated with increase in dietary fiber intake, odds ratio (OR) 0.79 (95% confidence interval (CI) 0.66, 0.96) for increase of at least one unit standard deviation (3.0 g/1,000 kcal) of the baseline distribution, PAF 16.0% (95% CI 4.2, 26.4%). Increase in the lifestyle behavior score was associated with reduced diabetes risk, OR 0.92 (95% CI 0.85, 0.99) per unit increase of the score. $\textbf{Conclusions:}$ These results support a causal link between lifestyle behavior and type 2 diabetes incidence. A small shift in lifestyle behaviors, in particular intake of dietary fiber, has the potential to reduce diabetes burden in the population and might be a suitable target for public health intervention.This work was supported by the Medical Research Council [MC_UU_12015/4], the Swedish Council for Working Life and Social Research [FAS 2006_1512], and the Swedish Research Council [2006-21576-36119-666]. The Västerbotten Intervention Programme is financed by Västerbotten County Council. Dr. Feldman is supported by the Raymond and Beverly Sackler Foundation through Churchill College, Cambridge