From Symptom Diaries to Smart Diagnostics: A Systematic Review of Digital Technologies for the Early Detection of Premenstrual Dysphoric Disorder (PMDD)

Abstract

Premenstrual Dysphoric Disorder (PMDD) is a severe mood disorder affecting approximately 3–8% of menstruating individuals, characterised by recurrent affective, cognitive, and physical symptoms during the luteal phase of the menstrual cycle. Despite formal recognition in diagnostic manuals, PMDD remains substantially underdiagnosed due to symptom overlap with depressive and anxiety disorders, stigma, and limited clinical awareness. Prospective symptom tracking required for diagnosis is rarely implemented in routine practice, creating a persistent diagnostic gap.

This systematic review examines the role of digital technologies in supporting the diagnosis and early detection of PMDD, with a focus on diagnostic accuracy, usability, and ethical considerations.

A systematic review was conducted in accordance with PRISMA 2020 and PRISMA-DTA guidelines. Searches were performed across seven electronic databases (MEDLINE, EMBASE, PsycINFO, CINAHL, Scopus, Web of Science, and IEEE Xplore) and supplemented by grey literature sources. Studies published between 2015 and 2025 evaluating digital tools for PMDD symptom monitoring, screening, or diagnostic support were included. Quantitative, qualitative, and mixed-methods studies were synthesised using meta-analytic and narrative approaches as appropriate.

Nineteen studies met inclusion criteria, encompassing mobile health applications, algorithmic and artificial intelligence-based models, telehealth platforms, and wearable-enabled systems. Evidence indicates that digital tools can enhance prospective symptom tracking, patient engagement, and early recognition of PMDD patterns. Algorithmic approaches, including probabilistic and Bayesian models, demonstrated potential for improving diagnostic precision, with one validated tool (C-PASS) achieving high agreement with clinician diagnosis. However, most digital solutions lacked external validation, clinical integration, and transparency. Usability and adoption were strongly influenced by perceived usefulness, trust, and self-efficacy. Ethical concerns related to data privacy, equity, and inclusivity were consistently reported.

Digital technologies offer promising avenues to address long-standing barriers in PMDD diagnosis by enabling scalable, patient-centred, and longitudinal symptom assessment. Nevertheless, their clinical utility remains constrained by limited validation, governance challenges, and inequitable design. Future efforts must prioritise rigorous diagnostic evaluation, ethical data stewardship, and integration within healthcare systems to realise the transformative potential of digital diagnostics in PMDD and women’s mental health.