Associations between premenstrual symptoms and (traumatic) stress: a systematic review and three multilevel meta-analyses

Abstract

Background Core premenstrual disorders (PMDs), including premenstrual syndrome (PMS) and premenstrual dysphoric disorder, can cause significant impairment. Despite evidence linking stress and premenstrual symptoms, a systematic synthesis is lacking. Aims To systematically review the literature and meta-analyse evidence on the relationship between premenstrual symptoms and stress. Method Four databases (Web of Science, PubMed, PsycINFO, Scopus) and Google Scholar were searched for studies indexed before 27 August 2024 (no language/year restrictions) assessing the relationship between self-reported stress and premenstrual symptoms in regularly menstruating individuals (PROSPERO: CRD42021244503). Three multilevel meta-analyses estimated (a) the correlation between stress and premenstrual symptom severity, (b) stress differences between individuals with and without core PMD across the menstrual cycle and (c) the impact of traumatic experiences on the occurrence of premenstrual symptoms. Study quality and publication bias were assessed. Results We synthesised 188 effect sizes from 66 studies (N = 38 344), indicating (a) a positive correlation (r = 0.29, 95% CI 0.23-0.36); (b) higher stress levels in participants with core PMD (d = 0.79, 95% CI 0.32-1.26), particularly during the luteal phase (d lut = 1.01, 95% CI 0.46-1.57); and (c) over twofold higher odds (odds ratio 2.45, 95% CI 1.87-3.23) of PMS in individuals with a history of trauma. Heterogeneity was high (I 2 = 84.64-91.38%); one meta-analysis (c) showed evidence of publication bias. Conclusions The results indicate an association between stress and premenstrual symptoms, an effect of cycle phase and trauma as a risk factor for PMS. Future research should explore underlying biopsychosocial mechanisms.