Statistical analysis
Following the linkage between nulliparous women with previous sPTB and primiparous women with an ongoing pregnancy after 16 weeks of gestation, the baseline characteristics of the first pregnancy were described and tested for the three spontaneous PTB groups of the index population; chi-square for categorical variables and anova for continues variables.
Logistic regression analysis was performed to assess the association between the gestational age at prior birth and the PTB pregnancy outcomes of the subsequent pregnancy. PTB rates are expressed per 100 births. Odds ratios (OR) and 95% confidence intervals (C.I.) were calculated separately for the three outcome groups (PTB <24 weeks, PTB < 32 weeks and PTB <37 weeks) in the subsequent pregnancy, with prior birth between 24+0 – 27+6 weeks as the reference. Analysis was adjusted for maternal characteristics (age, ethnicity, low SES), neonatal characteristics (fetal sex) and obstetrics characteristics (interpregnancy interval).
A separate analysis was conducted to assess the association between the interpregnancy interval, gestational age at prior birth and the recurrent risk. The interpregnancy interval was subdivided into groups of 0-3, 4-6, 7-12, 13-36 and >37 months, with 13-36 months as the reference.
Linkage and analysis were performed using SAS, version 9.4.