Logistic regression analysis of the effect of average annual weight change over 10 years and baseline characteristics on risk of subsequent diabetes in a cohort of overweight adults, First National Health and Nutrition Examination Survey, Epidemiologic Follow up Study, 1971–1992
| Variable | Model I | Model II | |||||
|---|---|---|---|---|---|---|---|
| β | Odds ratio | 95% confidence intervals | β | Odds ratio | 95% confidence intervals | ||
| Annual weight change (per kg/y) | 0.380 | 1.46 | 1.26, 1.69 | 0.399 | 1.49 | 1.29, 1.73 | |
| Age (per 10y) | 0.033 | 1.23 | 1.09, 1.38 | 0.028 | 1.16 | 1.03, 1.31 | |
| Age2 3-150 | −0.001 | −0.001 | |||||
| BMI (per kg/m2) | 0.117 | 1.13 | 1.09, 1.16 | 0.114 | 1.12 | 1.08, 1.16 | |
| Skinfold Ratio (per 0.2 unit) | 0.423 | 1.12 | 1.03, 1.15 | ||||
| Systolic blood pressure (per 5mm Hg) | 0.008 | 1.04 | 1.01, 1.08 | ||||
↵3-150 The significant non-linear association between age and diabetes risk requires calculation of the odds ratio for a specified change in age. The odds ratio in this table represents the change from age 40 to age 50.