El‐Bastawissi 1999.

Methods	Retrospective cohort study Comparison groups: A) External from general population ‐ matching for age and country of origin (foreign vs USA) B) Women with Carcinoma in situ (CIS) but no treatment ‐ unmatched Both had regressional analysis for parity, race, maternal smoking, marital status and history of TOPs. Information source ‐ Cancer Surveillance System (a population‐based cancer registry covering 13 counties of western Washington) at the Fred Hutchinson Cancer Research Center in Seattle, Washington, and Birth Certificates from the Department of Health in Washington state
Participants	Treated group ‐ 1096 women who were less than 50 years old with CIS, were diagnosed between 1984 and 1992, were treated with excisional or ablative therapy and subsequently delivered live singletons between 1984 and 1995 (the women were identified by the Cancer Surveillance System) Untreated group ‐ A) 9201 women (random sample selected from birth certificates, but frequency‐matched for age and the country of origin) without cervical cancer who gave birth during the same years without previous treatment. B) 330 women with untreated CIS Only women (for both the treated and untreated group) residing in the 13 counties of western Washington covered by the Cancer Survellance system were included. Only women who indicated the same father of the index infant and previous children were included.
Interventions	Excision NOS (CKC, LC, LLETZ); Ablation NOS (LA, CT)
Outcomes	PTB (< 37 weeks); PTB (< 37 weeks) (singleton pregnancies); CS; LBW (< 2500 g)
Notes	From the 1851 women with CIS, 1539 women had a pregnancy after the CIS diagnosis. Of these women, 212 had no surgical procedure before pregnancy, 227 had D&C or ECC before pregnancy, 85 had cryosurgery or LA before pregnancy, and 1011 had conisation before pregnancy. For 4 women, the procedure (if any) before pregnancy was unknown. From the 1851 women with CIS, 312 were pregnant at the time of the diagnosis. Of these women, 118 had no surgical procedure during pregnancy, 33 had D&C or ECC during pregnancy, 6 had cryosurgery or LA during pregnancy, and 142 had conisation during pregnancy. For 13 women, the procedure (if any) during pregnancy was unknown. It is possible to make the following comparisons in our meta‐analysis: a) Women with CIS and treatment before pregnancy versus women with CIS but no treatment (diagnosis of CIS before pregnancy) b) Women with CIS and treatment before pregnancy versus women with CIS but no treatment (diagnosis of CIS during pregnancy) c) Women with CIS and treatment before pregnancy versus women with CIS but no treatment (diagnosis of CIS before or during pregnancy) d) Women with CIS and treatment before pregnancy versus general population Women that had treatment during pregnancy were excluded according to the exclusion criteria of the systematic review.
Risk of bias
Bias	Authors' judgement	Support for judgement
Incomplete outcome data (attrition bias) All outcomes	Low risk	Information on birth weight, gestation length, and delivery method was complete for 98.8%, 83.2%, and 93.8% of women with CIS versus 99.7%, 86.7%, and 94.7% for comparison women, respectively.
Selective reporting (reporting bias)	Low risk	No reporting bias is obvious
Other bias	Low risk	No other obvious source of bias
Relevant assignment described?	Low risk	Yes, treatment performed on clinical grounds
Representative intervention group?	Low risk	All eligible for the study women from 13 counties of western Washington (a population‐based study)
Representative comparison group?	Low risk	Both untreated groups were drawn from the same source as the treated group
Comparability of treatment groups?	Low risk	Matching for age and country of origin (foreign vs USA). Regressional analysis for parity, race, maternal smoking, marital status and history of TOPs.