Table 2. Association model of variables for gestational diabetes in a conditional logistic regression model.
| Variable | Initial adjustment modela | p-value | Final adjustment modelb | p-value | ||
|---|---|---|---|---|---|---|
| OR | 95%CI | OR | 95%CI | |||
| Age ≥ 35 years | 3.690 | 1.520–8.950 | 0.004 | 3.87 | 1.651–9.074 | 0.002 |
| Overweightness | 1.430 | 0.543–3.767 | 0.469 | − | − | − |
| Obesity | 1.710 | 0.645–4.555 | 0.280 | 2.08 | 0.920–4.707 | 0.079 |
| FH T2DM | 2.010 | 0.910–4.457 | 0.084 | 2.09 | 0.962–4.550 | 0.063 |
| PH GD | 2.880 | 0.384–21.614 | 0.304 | − | − | − |
| Weight gain | 0.981 | 0.868–1.108 | 0.754 | − | − | − |
Abbreviations: 95%CI, 95% confidence interval; FH T2DM, family history of type 2 diabetes mellitus; OR, odds ratio; PH GD, previous history of gestational diabetes.
Notes:aInitial model: step 1 of backward conditional logistic regression with all variables inserted. bFinal model: stage 4 of the regression process, after the removal of the variables weight gain, excess weight, and PH GD (in that order).