Abstract
OBJECTIVE: To investigate the association between prenatal maternal perceived stress, salivary cortisol, and the odds of recurrent preterm birth (rPTB). DESIGN: Prospective cohort study. SETTING: Soroka University Medical Center and Clalit Health Maintenance Organization clinics in Israel. POPULATION: A total of 245 pregnant women with a history of spontaneous singleton PTB, all carrying singleton pregnancies in the current pregnancy. Data source were on-line self-completed computerised REDCap questionnaires and summary of computerised medical records including data on medications and laboratory results and cortisol in the saliva. METHODS: Participants provided saliva samples and completed stress questionnaires twice during pregnancy, with two assessments separated by 4-8 weeks. Recurrent PTB incidence was compared between women with high vs. low cortisol and perceived stress (> 75th percentile). Multivariable logistic regression models were used to analyse the odds of rPTB. MAIN OUTCOME MEASURES: Recurrent PTB incidence, defined as delivery < 37 weeks in the current pregnancy, identified from hospital records. RESULTS: Among 245 women, 22.4% experienced rPTB. High perceived stress levels later in pregnancy were associated with a higher incidence of rPTB (38.9% vs. 19.2%; p = 0.011, adj. OR = 2.44, 95% CI: 1.053-5.639; p = 0.038), while adjusting for maternal age, number of prior PTBs and parity. Women with consistently high stress levels had the highest incidence of rPTB (40%), followed by women with increased stress later in pregnancy (37.5%). The multivariable regression analyses identified consistently high stress (adj. OR = 3.82, 95% CI: 1.37-10.64; p = 0.010) and high stress only in later pregnancy (adj. OR = 3.18, 95% CI: 1.04-9.73; p = 0.042) as significant risk factors of rPTB. No association was observed between cortisol levels, high perceived stress in the first trimester and rPTB. No differences were observed between women with versus without rPTB in terms of demographic, health, or lifestyle characteristics. No association was observed between cortisol levels, high perceived stress in the first trimester and rPTB. No differences were observed between women with versus without rPTB in terms of demographic, health, or lifestyle characteristics. CONCLUSIONS: Persistent and late-pregnancy high perceived stress are significant risk factors, associated with increased odds of rPTB.