Table 1

Severe depression and use of antidepressant medication during the 4 years preceding diabetes diagnosis among participants with incident type 2 diabetes and individually matched control subjects

Severe depressionAntidepressant use*No. participants (no. incident diabetes cases)ORs (95% CI) for incident diabetes
Comparison AComparison BComparison C
NoNo4,530 (696)1 (reference)0.52 (0.40–0.67)0.84 (0.44–1.57)
NoYes331 (85)1.93 (1.48–2.51)1 (reference)1.61 (0.82–3.16)
YesNo65 (12)1.20 (0.64–2.25)0.62 (0.32–1.22)1 (reference)
YesYes159 (58)3.17 (2.27–4.45)1.65 (1.09–2.48)2.65 (1.31–5.39)
Additional adjustment for prevalent physical disease
NoNo4,530 (696)1 (reference)0.60 (0.45–0.79)0.95 (0.49–1.84)
NoYes331 (85)1.68 (1.27–2.21)1 (reference)1.59 (0.79–3.22)
YesNo65 (12)1.05 (0.55–2.04)0.63 (0.31–1.27)1 (reference)
YesYes159 (58)2.76 (1.93–3.94)1.64 (1.06–2.54)2.61 (1.25–5.49)
  • Results are based on conditional logistic regression analysis.

  • *“Yes” refers to a minimum use of 200 defined daily doses of antidepressants and “No” to a use of 0–199 defined daily doses during 4 years.

  • †Hypertension, coronary heart disease, cerebrovascular disease, and cancer. Prevalence of cardiovascular disease 4 years before the diabetes diagnosis was higher among diabetic case subjects than among control subjects (hypertension 28.0 vs. 9.3%, P < 0.0001; coronary heart disease 3.2 vs. 1.4%, P = 0.0002). There was no difference in cerebrovascular disease (0.7 vs. 0.8, P = 0.88) or cancer prevalence (2.7 vs. 2.4%, P = 0.61) between the groups.