Résumé
We examined whether psychological distress predicts incident type 2 diabetes and if the association differs between populations at higher or lower risk of type 2 diabetes. This was a prospective cohort of 5,932 diabetes-free adults (4,189 men and 1,743 women, mean age 54.6 years) with three 5-year data cycles (1991-2009): a total of 13,207 person-observations. Participants were classified into four groups according to their prediabetes status and Framingham Offspring Type 2 Diabetes Risk Score: normoglycemia with a risk score of 0-9, normoglycemia with a risk score of 10-19, prediabetes with a risk score of 10-19, and prediabetes with a risk score of >19. Psychological distress was assessed by the General Health Questionnaire. Incident type 2 diabetes was ascertained by 2-h oral glucose tolerance test, doctor diagnosis, or use of antihyperglycemic medication at the 5-year follow-up for each data cycle. Adjustments were made for age, sex, ethnicity, socioeconomic status, antidepressant use, smoking, and physical activity. Among participants with normoglycemia and among those with prediabetes combined with a low risk score, psychological distress did not predict type 2 diabetes. Diabetes incidence in these groups varied between 1.6 and 15.6%. Among participants with prediabetes and a high risk score, 40.9% of those with psychological distress compared with 28.5% of those without distress developed diabetes during the follow-up. The corresponding adjusted odds ratio for psychological distress was 2.07 (95% CI 1.19-3.62). These data suggest that psychological distress is associated with an accelerated progression to manifest diabetes in a subpopulation with advanced prediabetes.