{"paper_id":"8e6e8d37-bd56-4703-9886-36d3ff6dcf9f","body_text":"Accepted for/Published in: JMIR Formative Research\nDate Submitted: Dec 12, 2025\nOpen Peer Review Period: Dec 15, 2025 - Feb 4, 2026\nDate Accepted: Apr 4, 2026\nDate Submitted to PubMed: Apr 16, 2026\n(closed for review but you can still tweet)\nReal-World Implementation of EndoConnect in Brazilian Primary Care: A Formative Study of Usability, Engagement, and Equity in Digital Endometriosis Care\nABSTRACT\nBackground:\nEndometriosis affects ≈10% of reproductive-age women worldwide (≈190 million) and is marked by debilitating chronic pain and infertility. In Brazil’s public health system (SUS), diagnostic delays average 7–10 years, disproportionately affecting low-income, rural, Black, and Indigenous women.\nObjective:\nThis formative study aimed to (1) develop and deploy EndoConnect Alpha—an offline-capable progressive web app integrating evidence-based education, symptom tracking, moderated community support, and privacy-by-design AI-assisted tele-ultrasound triage—in SUS primary care units in Ceará, Brazil; (2) evaluate usability, acceptability, engagement, and preliminary clinical-psychosocial impact; and (3) propose the NAM-Endora Framework for ethical AI governance in low- and middle-income countries (LMICs).\nMethods:\nApplied methodological study with quantitative cross-sectional formative design (January 2024–November 2025). After expert validation and software registration (INPI BR5120250005556-0), 60 participants (45 women aged 18–45 years with suspected/confirmed endometriosis and 15 primary care professionals) were recruited from 10 SUS units (60% rural). Usability (System Usability Scale), acceptability (Technology Acceptance Model), engagement (Firebase Analytics), and pre/post outcomes (pain VAS, EKES-15 knowledge, GAD-7 anxiety, adherence, referral rate) were assessed over 8 weeks. Ethics approval: CAAE 82094924.8.0000.5049.\nResults:\nMean SUS score 88.9 ± 9.8 (excellent); TAM 91.4%. Trail completion 79%; mean daily use 17.2 minutes. Significant improvements: pelvic pain −23% (P=.02), adherence +17% (P=.01), knowledge +21% (P<.001), anxiety −14% (P=.04), referrals +15% (P=.04). Largest benefits observed in rural, low-education, Black/Brown/Indigenous subgroups. The NAM-Endora Framework is proposed as the first LMIC-tailored ethical AI governance model.\nConclusions:\nEndoConnect Alpha is feasible and equity-enhancing in SUS primary care. The NAM-Endora Framework provides a novel, replicable model for responsible AI deployment in LMICs, with potential to reduce the global burden of endometriosis. Clinical Trial: Not applicable\nCitation\nRequest queued. Please wait while the file is being generated. It may take some time.\nCopyright\n© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.","source_license":"CC0","license_restricted":false}