ObjectiveTo estimate the prevalence of pelvic pain and model associations with potential demographic, obstetric, gynaecological and psychosocial determinants. Design, setting and sampleA cohort study of women born between 1972 and 1973 in Dunedin, New Zealand, most recently assessed when aged 38 years (95% of survivors retained); 429 women were eligible for analysis. MethodsWomen self-completed reproductive health questionnaires at ages 21, 26, 32 and 38 years, with questions on dysmenorrhoea at ages 13 and 15, and on all pelvic pain at age 38. Prevalence and 95% confidence intervals (CI) were calculated and Poisson regression used to model associations. Main outcome measuresThe prevalence of pain and adjusted relative risks (ARR) for potential explanatory factors. ResultsOver half (54.5%, 95% CI 49.7-59.3%) of women experienced pelvic pain in the past 12 months at age 38. Dysmenorrhoea was reported by 46.2% (41.3-51.3%), dyspareunia by 11.6% (8.7-15.2%) and other pelvic pain (OPP) by 17.3% (13.8-21.2%). After adjusting for multiple factors, pregnancy (ARR 0.60, 95% CI 0.32-1.13) and childbirth (ARR 0.52, 95% CI 0.25-1.09) were borderline protective for dyspareunia and OPP, respectively. However, childbirth was not associated with dysmenorrhoea (ARR 0.97, 95% CI 0.74-1.28). Dysmenorrhoea and dyspareunia were strongly associated, and both were associated with endometriosis. ConclusionsOur data confirm that female pelvic pain is common, and suggest common gynaecological and obstetric causal pathways, but there was no strong evidence supporting a benefit of childbirth for dysmenorrhoea. Further research on obstetric events and pelvic pain is needed, with both being common experiences. Tweetable abstractPelvic pain was common at age 38, especially dysmenorrhoea (46.2%), and no improvement was detected following childbirth. Tweetable abstract Pelvic pain was common at age 38, especially dysmenorrhoea (46.2%), and no improvement was detected following childbirth.