Table 3.

Table for evidence grading: risk of gestational diabetes and risk of delivering a large-for-gestational-age infant

Reference details	Glazer N et al. (60), USA	Villamor and Cnattingius (5), Sweden	Villamor and Cnattingius (5), Sweden
Study design	Prospective cohort study	Prospective cohort study	Prospective cohort study
Population/subject characteristics	Obese women (heavier than 200 lbs or 90.7 kg) of mixed ethnicity with 2 singleton births who were nondiabetic at the first pregnancy	Women in Sweden giving birth to two consecutive singletons between 1992 and 2001	Women in Sweden giving birth to two consecutive singletons between 1992 and 2001
No of subjects analysed	4,012	313 (from 151,025) for risk of gestational diabetes	2,350 (from 151,025) for risk of delivering a large-for-gestational-age infant
Outcome measures	Risk of gestational diabetes at the second pregnancy	Risk of gestational diabetes at the second pregnancy	Risk of delivering a large-for gestational-age (LGA) infant at the second pregnancy
Exposure	Prepregnancy weight at an index pregnancy minus the corresponding weight at the previous pregnancy	Difference between the two pregnancies with respect to BMI recorded at the first antenatal visit	Difference between the two pregnancies with respect to BMI recorded at the first antenatal visit
Follow-up period, drop-out rate	Nine-months follow-up, no drop-outs	Nine-months follow-up, no drop-outs	Nine-months follow-up, no drop-outs
Dietary assessment method	No dietary assessment	No dietary assessment	No dietary assessment
Results	Women who lost at least 10 lbs (4.54 kg) between pregnancies had a decreased risk of gestational diabetes relative to women who lost less weight during this period (relative risk=0.63, 95% CI, 0.38–1.02)	Overweight and obese women who decreased their BMI more than one unit between pregnancies had no significant reduction in the risk of gestational diabetes (OR 0.96, 95% CI, 0.66–1.37)	Overweight and obese women who decreased their BMI more than one BMI-unit between pregnancies had a significant reduction in the risk of giving birth to a LGA-infant (OR 0.82, 95% CI, 0.72–0.95)
Confounders adjusted for	Age and weight gain during each pregnancy	Height, interpregnancy interval, age, country of origin, years of education, year of delivery and smoking	Height, interpregnancy interval, age, country of origin, years of education, year of delivery and smoking
Study quality and relevance	Study quality: B. The study is not quite relevant since there is no information that the women received dietary advice and we do not know why they lost weight	Study quality: A. The study is not quite relevant since there is no information that the women received dietary advice and we do not know why they lost weight	Study quality: A. The study is not quite relevant since there is no information that the women received dietary advice and we do not know why they lost weight