Length of labour in mothers and their daughters: A matched cohort study

Objective: Physiological length of labour is highly variable and population norms have low sensitivity and specificity for individuals. The birth history of mothers may provide a basis for personalized assessment of labour progress in their nulliparous daughters. This study was designed to investigate the relationship between length of labour in nulliparous daughters and in their mother's first birth, as a basis for constructing individualised labour prediction models in future. Study design: A mother-daughter matched cohort study was conducted in two Israeli maternity hospitals. Recruitment took place between September 2014 and June 2015 via antenatal clinics. Inclusion criteria were nulliparous daughters with singleton pregnancies at ≥32 weeks' gestation and mothers of included daughters who had a first birth in hospital prior to 1997. Data were collected prospectively for daughters by questionnaire and from electronic hospital records, and through retrospective recall questionnaires for mothers. Mother-daughter length of labour data were analysed using parametric and non-parametric tests and logistic regression. Length of labour was categorized as ≤10 h and >10 h. Other factors influencing daughters' length of labour were also examined. Results: Data from 323 mother-daughter pairs were analysed. Univariate logistic regression analysis showed that daughters of mothers who were in active labour for more than 10 h showed increased likelihood of having a longer labour [OR1.91 (95 % CI 1.19, 3.05, P = 0.007)]. Controlling for infant gender increased the effect size [OR3.23 (95 % CI 1.55, 6.74, P = 0.002)]. Multivariable logistic regression indicated that mothers' length of labour [OR1.88 (95 % CI 1.12, 3.17)] and daughters' age [OR1.08 (95 % CI 1.02, 1.14)], weight gain in pregnancy [OR1.10 (95 % CI 1.04, 1.16)] and use of anesthesia, were statistically significant factors for daughters' length of labour, with sensitivity, specificity, and positive and negative predictive values of 74 %, 56 %, 66 %, and 64 %, respectively. Conclusions: A strong positive association between mother and daughter lengths of labour was found. A model that includes length of labour in their mother's first birth might be useful for labour progress prediction for nulliparous women. Practitioners could inquire about maternal first birth patterns as an additional heuristic to guide practice and increase precision in the clinical management of nullipara women's labour and delivery.