Skip to main content
. 2022 Sep 23;20:318. doi: 10.1186/s12916-022-02515-w

Table 2.

Logistic regression analysis for actionable ADA-HbA1c (defined as prediabetes or diabetes) after adjusting for covariates

Risk factors Estimated coefficients P-value OR OR (95% CI)
Lower Upper
Age (years) 0.11 0.015 1.11 1.02 1.21
Ethncity 1.95 < 0.001 7.05 2.94 16.92
Glucose-lowering drugs (yes) 1.06 0.008 2.87 1.31 6.30
FPG at OGTT in pregnancy (mmol/l) 0.67 0.011 1.96 1.16 3.29
2-h OGTT glucose in pregnancy (mmol/l) 0.12 0.246 1.13 0.92 1.37
Pre-pregnancy BMI (kg) 0.04 0.255 1.04 0.97 1.12
GWG (kg) − 0.03 0.366 0.98 0.92 1.03
> 3 children (yes) 0.07 0.879 1.07 0.45 2.54
Low education (yes) 0.25 0.476 1.28 0.65 2.55
1.degree relatives w/diabetes (yes) 0.66 0.070 1.93 0.95 3.91
GDM before index pregn (yes) 1.18 0.031 3.24 1.11 9.45
Glucose-lowering drugs * GDM before index pregn − 1.96 0.009 0.14 0.03 0.61

The model includes the covariates (odds ratio (OR) with 95% CI): age, ethnicity, glucose-lowering drugs, fasting plasma glucose (FPG) and 2-h oral glucose tolerance test (OGTT) values in pregnancy, pre-pregnancy BMI, gestational weight gain (GWG), parity, education, first-degree relatives with diabetes, gestational diabetes mellitus (GDM) before index pregnancy, and the interaction term: Glucose-lowering drugs * GDM before index pregnancy