Background Premature birth (<37 weeks gestation) is linked to maternal weight. However, the impact of maternal interpregnancy weight change on premature birth is unclear. The objective of this study was to examine the association between maternal interpregnancy weight change and premature birth, in a population-based cohort.

Methods Data from 2003 to 2018 from one maternity centre in England was used to produce two cohorts. The primary cohort (n=14,961 women) consisted of first and second live-birth pregnancies. The secondary cohort (n=5,108 women) consisted of second and third live-birth pregnancies. Logistic regression was used to look for associations between interpregnancy weight change and premature birth, whilst adjusting for covariates at second or third pregnancy (maternal age, ethnicity, educational achievement, employment status, co-morbidities, pregnancy complications, smoking status, previous premature birth, sex of the infant, fertility treatment), BMI and gestational age at initial pregnancy booking appointment and interpregnancy interval. Subgroup analyses were performed by booking BMI group at initial pregnancy.

Results More women gained weight (>1 kg/m², primary cohort: 47.1%, secondary cohort: 46.5%) than lost weight (>1 kg/m², primary cohort: 15.8%, secondary cohort: 18.1%) between pregnancies. The median weight change was 0.8 kg/m² between pregnancies in both primary (IQR -0.4 to 2.4 kg/m²) and secondary (IQR -0.5 to 2.4 kg/m²) cohorts.

Primary cohort 3.4% (n=514) of second pregnancy births were premature births. Weight loss (>3 kg/m²) was associated with increased risk of premature birth (adjusted odds ratio (aOR) 3.50, 95% Confidence Interval (CI) 1.78–6.88), in women who were normal weight (BMI 18.5–25 kg/m²) at the start of their first pregnancy. Weight gain >1 kg/m² was not associated with the outcome in any of the initial BMI categories.

Secondary cohort 4.2% (n=212) of third pregnancy births were premature births. Losing >3 kg/m² was associated with increased risk of premature birth (aOR 2.01, 95% CI 1.05–3.87) when analysing the whole sample, but not when restricting the analysis to overweight or obese women at the start of the second pregnancy.

Conclusion Interpregnancy weight loss in women who were normal weight in their first pregnancy was associated with increased risk of premature birth. No such association was found when the analysis was restricted to women who were overweight or obese in their previous pregnancy whether it was first or second. Interpregnancy weight gain was not associated with premature birth. Current national guidelines encouraging obese women to lose weight before a pregnancy should be followed, in order to reduce other maternal and offspring adverse outcomes of maternal obesity.