Skip to main content
. 2020 Apr 13;3(4):e202605. doi: 10.1001/jamanetworkopen.2020.2605

Table 3. Association Between Mode of Delivery With the Risk of Type 2 Diabetes.

Characteristic HR (95%CI) P value
Vaginal delivery Cesarean delivery
Cases of type 2 diabetes, No. 1927 87 NA
Person-time, Person-years 1 852 102 61 876 NA
Incidence rate of type 2 diabetes, per 10 000 person-years 10.4 14.1 NA
Crude 1 [Reference] 1.42 (1.14-1.76) .002
Adjusteda 1 [Reference] 1.46 (1.18-1.81) .001
Additional analyses
Marginal structural model estimate 1 [Reference] 1.19 (0.96-1.50) .12
Treating maternal BMI as continuous variableb 1 [Reference] 1.47 (1.18-1.82) .001
Log-binomial model for risk ratio 1 [Reference] 1.49 (1.19-1.87) <.001
Additional adjustment for offspring characteristics
Breastfeeding 1 [Reference] 1.45 (1.17-1.80) .001
Updated BMI 1 [Reference] 1.34 (1.08-1.67) .008

Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); HR, hazard ratio; NA, not applicable.

a

Adjusted models included terms for maternal age at delivery, race/ethnicity (white or other), maternal educational level (≥high school or ≥college), maternal prepregnancy BMI group (<18.5, 18.5-24.99, 25-29.99, or ≥30), gestational weight gain (<9.1 kg or ≥9.1 kg), maternal height, gestational diabetes (yes or no), preeclampsia (yes or no), pregnancy-induced hypertension (yes or no), year of birth (1946-1951, 1952-1961, or 1962-1964), gestational age at delivery (<37, 37-39, 40-42, or ≥43 weeks), birth weight group (<2.3, 2.3-3.1, 3.2-3.8, 3.9-4.4, or ≥4.5 kg), smoking during pregnancy (no, first trimester, second and third trimesters, or current), and region of residence at birth (Northeast, Midwest, West, or South).

b

Adjusted model modeling prepregnancy BMI as a continuous variable with a linear term and a quadratic term.