Table 1 Summary of evidence grading for meta-analyses associating diabetes and anti-diabetic interventions with risk of obstetric and gynaecological morbidity—cohort studies only

Strong

P < 10^−6||; > 1000 cases; I² < 50%; no small study effects^¶; prediction interval excludes the null value; no excess significance bias^† survives 10% credibility ceiling

n = 2

Maternal outcomes

⦁ Caesarean section (GDM vs non-GDM)

Fetal outcomes

⦁ Large for gestational age (GDM vs non-GDM)

Highly suggestive

P < 10^−6||; > 1000 cases; P < 0.05 of the largest study in a meta-analysis

n = 3

Gynaecological outcomes

⦁ Ovarian cancer occurrence (Metformin vs non-metformin, DM2)

Fetal outcomes

⦁ Major congenital malformations (unspecified) (PGDM vs non-DM)

⦁ Congenital heart defects (PGDM vs non-DM)

Suggestive

P < 10^−3||; > 1000 cases

n = 10

Maternal outcomes

⦁ Preeclampsia (GDM vs non-GDM, IADPSG criteria)

⦁ Preeclampsia (GDM vs non-GDM, WHO criteria)

⦁ Postnatal depression (GDM vs non-GDM)

Fetal outcomes

⦁ Stillbirth (> 20 weeks or > 400 g) (PGDM vs non-DM)

⦁ Major congenital malformations (unspecified) (GDM vs non-GDM)

⦁ Large for gestational age (GDM vs non-GDM, IADPSG criteria)

⦁ Respiratory distress syndrome (DM vs non-DM)

⦁ Introduction of formula milk/breastmilk substitute before hospital discharge (GDM vs non-GDM)

Gynaecological outcomes

⦁ Endometrial cancer incidence (DM vs non-DM)

⦁ Improved endometrial cancer survival (metformin vs other anti-diabetics)

Weak

P < 0.05

n = 28

Maternal outcomes

⦁ Miscarriage (poor vs optimal glycaemic control of DM1/2)

⦁ Antenatal depressive symptoms ⦁ (GDM vs non-GDM)

⦁ Decreased duration of ⦁ breastfeeding (GDM vs non-GDM)

⦁ Low breastmilk protein content (GDM vs non-GDM)

⦁ Miscarriage (continuous sc Ins infusion vs multiple daily inj, DM1)

Fetal outcomes

⦁ Congenital malformations (preconception vs no preconception care, PGDM)

⦁ Perinatal mortality (preconception vs no preconception care, PGDM)

⦁ Preterm delivery (preconception vs no preconception care, PGDM)

Fetal outcomes

⦁ Macrosomia (GDM vs non-GDM, WHO criteria)

⦁ Macrosomia in Iranian women (GDM vs non-GDM)

⦁ Congenital malformations (unspecified) (poor vs optimal glycaemic control of DM1/2)

⦁ Major congenital malformations (poor vs optimal glycaemic control of DM1/2)

⦁ Perinatal mortality (poor vs optimal glycaemic control of DM1/2)

⦁ Perinatal mortality (DM2 vs DM1)

⦁ Brachial plexus palsy (GDM vs non-GDM)

⦁ Respiratory distress syndrome (GDM vs non-GDM)

⦁ LGA (Lispro vs Regular Ins or NPH, GDM, DM1/2)

⦁ LGA (Lispro vs Regular Ins, DM1)

⦁ Macrosomia > 4.5 kg (continuous sc Ins infusion vs multiple daily inj, DM1)

⦁ Higher birth weight (Lispro vs Regular Ins or NPH, GDM/ DM1/2)

Gynaecological outcomes

⦁ Cervical cancer occurrence (metformin vs non-metformin, T2DM)

Gynaecological outcomes

⦁ Endometrial cancer mortality (disease-specific) (DM1/2 vs non-DM)

⦁ Premalignant/malignant endometrial polyps (DM vs non-DM)

⦁ Ovarian cancer incidence (DM vs non-DM)

⦁ Ovarian cancer mortality (disease-specific) (DM1/2 vs non-DM)

⦁ Infrequent cervical cancer screening (DM vs non-DM)

⦁ Ovarian cancer incidence (DM1 vs non-DM)

⦁ Ovarian cancer incidence (DM2 vs non-DM)