Results
Our search strategy identified 978 studies. We selected 33 papers for assessment of the full paper (Figure S1). We excluded 13 studies: 5 studies without CL measurement15-19, 2 studies with the wrong population20,21, 2 studies considered duplicate patient population22,23, and 1 study included patients with a previous episode of threatened preterm labor24. For three studies, there was concern about trustworthiness of the data25,26,27. Two studies25,26 had not shared data in an individual participant data meta-analysis5, of which one study had been retracted by the journal as it was not performed under the appropriate ethics 25 . A third author did not respond on our request for additional information26 (Table S1). Two studies involving cerclage presented combined results for singleton and twins only and were excluded because the authors could not be contacted to assess only twins data28,29. Thus, we included 20 studies (Table S2).