Characterizing endometriosis and adenomyosis symptom clusters and their impact on quality of life in the All of Us Research Program

STUDY QUESTION: Are there distinct symptom clusters in women with endometriosis and adenomyosis, and how do these clusters predict quality of life? SUMMARY ANSWER: Latent class analyses of 22 438 women with endometriosis identified four symptom phenotypes, including high pain-gastrointestinal and psychological-neurological clusters; adenomyosis cases were concentrated in high‑pain classes without a minimal‑symptom group, and high‑burden phenotypes had poorer quality of life. WHAT IS KNOWN ALREADY: Endometriosis and adenomyosis are heterogeneous conditions with diagnostic delays of 4-11 years. The classification based on surgical findings and imaging fails to capture the complex, systemic symptom combinations that affect patients' quality of life (QoL). STUDY DESIGN, SIZE, DURATION: Large-scale cross-sectional cohort study using data from the United States-based All of Us Research Program (Controlled Tier v8; May 2018-October 2023). PARTICIPANTS/MATERIALS, SETTING, METHODS: The dataset included 22 438 women with electronic health records (EHR) and/or self-reported endometriosis. Four Latent Class Analyses (LCA) were performed on different patient groups: (i) the full endometriosis cohort (n = 22 438); (ii) an age-corrected subset restricted to premenopausal women (aged 18-45; n = 5542) to minimize menopausal confounding; and two subgroups derived from the age-corrected cohort to test whether concomitant adenomyosis defines distinct symptomatic phenotypes: (iii) clinically confirmed endometriosis without adenomyosis (n = 1797) and (iv) adenomyosis with endometriosis (n = 643). LCA was performed using 19 self-reported and EHR-derived symptoms and comorbidities (e.g. chronic pelvic pain, migraine, depression, gastrointestinal symptoms) to identify patient subgroups. Multinomial logistic regression assessed sociodemographic (age, BMI, deprivation index) and clinical predictors (hormonal contraceptive use, surgical history) of class membership. The association between latent classes and QoL outcomes was evaluated using data from the 'Overall Health' and 'Health Care Access and Utilization' surveys. Data are available via the All of Us Researcher Workbench (https://workbench.allofus.org). MAIN RESULTS AND THE ROLE OF CHANCE: Latent class analysis of clinically confirmed endometriosis identified four distinct symptom phenotypes, most notably a severe 'High Pain & Gastrointestinal with Mood Symptoms' (High) cluster and a unique 'Predominantly Psychological/Neurological' cluster that challenges traditional gynecological-focused diagnostic frameworks. Patients with adenomyosis exhibited a distinct profile with two high-pain symptom classes and the absence of a fully asymptomatic group, indicating a higher overall disease burden. Membership in the High class was significantly associated with lower general health scores, reduced social satisfaction, and increased barriers to healthcare access. LIMITATIONS, REASONS FOR CAUTION: The All of Us is a United States-based research program, and the findings should be replicated in other independent cohorts to confirm generalizability across other geographical regions. The cross-sectional design limits causal inference regarding QoL outcomes. Also, the analyses rely on a mix of diagnostic approaches, including surgical and non-surgical clinical and self-reported diagnoses, which is both a strength and a limitation. The 'Minimal Symptom Burden' class may reflect under-documentation of symptoms in clinical records rather than a true lack of symptoms, and the High class may reflect to patients who have higher multimorbidity, hence have more hospital visits. WIDER IMPLICATIONS OF THE FINDINGS: These results support a shift toward personalized, interdisciplinary management of endometriosis that addresses mental health and gastrointestinal symptoms alongside pelvic pain. The identification of a 'Predominantly Psychological/Neurological' cluster suggests that young women presenting with non-classical symptoms (anxiety, migraine, depression) could be screened for endometriosis to reduce diagnostic delays and improve life-course potential. FUNDING: D.K. was supported by a 'Ramón y Cajal' fellowship from the Spanish Ministry of Science and Innovation (RYC2024-050099-I). O.G. was supported by a postdoctoral award from the Amy P Goldman Foundation. DISCLOSURES: I.F. is the co-founder and co-owner of Sur180 Therapeutics, which is developing a novel treatment for endometriosis, and the Chief Scientific Officer of Nura Health, which is developing a non-invasive diagnostic solution for the disease. I.F. has received ARPA-H grant no. ARPA-H-ICHUB-24-101-1035 as Principal Investigator, consulting fees from Nura Health and Sur180 Therapeutics, and stock and share options in both companies. I.F. is also the inventor on patents for 'Compositions and methods for treating endometriosis' (US10695341 and US10729693) and serves as an unpaid Board Member of the World Endometriosis Society and the Fundación Puertorriqueña de Pacientes con Endometriosis (ENDOPR). H.S.T. received speaker fees from Gedeon Richter, a grant from AbbVie through Yale University, consulting fees from Regeneron, and is a co-inventor on Yale's endometriosis biomarkers and treatments (including US11993816B2, US10982282B2, and US12286629B2). He also serves on the boards of the Environmental Health Trust and Environment and Human Health. TRIAL REGISTRATION NUMBER: N/A.