Table 1—

Logistic regression models with drug-treated diabetes during follow-up as the dependent variable

Concise model: n = 4,595 (194 of whom developed diabetes)
Full model: n = 4,435 (182 of whom developed diabetes)
Score
OR (95% CI)CoefficientOR (95% CI)Coefficient
Intercept−5.514−5.658
Age (years)
 45–541.87 (1.12–3.13)0.6281.92 (1.13–3.25)0.6502
 55–642.44 (1.48–4.01)0.8922.56 (1.53–4.28)0.9403
BMI (kg/m2)
 >25 to 301.18 (0.57–2.45)0.1651.02 (0.48–2.15)0.0151
 >302.99 (1.31–6.81)1.0962.55 (1.10–5.92)0.9383
Waist circumference (cm)
 Men, 94 to <102; women, 80 to <882.36 (1.25–4.45)0.8572.78 (1.43–5.40)1.0213
 Men, ≥102; women, ≥883.86 (1.93–7.71)1.3504.16 (2.00–8.63)1.4244
Use of blood pressure medication*2.04 (1.46–2.83)0.7112.04 (1.45–2.88)0.7142
History of high blood glucose8.49 (5.66–12.73)2.1399.61 (6.31–14.63)2.2635
Physical activity <4 h/week1.31 (0.88–1.95)0.2682
Daily consumption of vegetables, fruits, or berries1.18 (0.85–1.64)0.1651
Area under the ROC curve0.8570.8600.852
  • Only subjects with no missing baseline risk factor data were included into the specific model. The score values were estimated based on the β coefficients of the logistic regression model and are presented for the full model. The concise model includes only these statistically significant variables. The full model includes also physical activity and fruit and vegetable consumption.

  • *

    * Question “Have you ever used drugs for high blood pressure?: No/Yes” in the questionnaire;

  • question “Have you ever been told by a health-care professional that you have diabetes or latent diabetes?: No/Latent diabetes/Diabetes” in the questionnaire;

  • individuals who, in their spare time, “read, watch TV, and work in the household with tasks that don’t strain physically” and whose “work is mainly done sitting and does not require much walking.” The next category was “physical activity at least 4 hours per week.”