JMIR Publications

ABSTRACT

Background:

Mobile health (mHealth) applications for menstrual cycle and fertility tracking are widely used to support self-monitoring, reproductive planning, and health awareness among women. While these tools promise personalized predictions and convenient access to reproductive health information, concerns persist regarding their clinical accuracy, adaptability to irregular cycles, transparency of algorithms, and real-world user experience.

Objective:

This structured review aimed to evaluate the features, physiological integration, predictive performance, validation practices, and user-reported outcomes of mobile applications designed for menstrual and fertility tracking, and to contextualize current evidence using COSMIN and ISPOR evaluation frameworks.

Methods:

A structured narrative review with systematic elements was conducted following the PRISMA-like reporting framework. Literature published between January 2013 and October 2025 was identified through searches of PubMed, EMBASE, Scopus, and Web of Science, supplemented by semantic and citation-based searches in the Semantic Scholar, OpenAlex, and Google Scholar databases. AI-assisted relevance ranking supported the initial screening, followed by an independent human review. Forty studies meeting the predefined eligibility criteria were included in the qualitative synthesis. Owing to the heterogeneity in study designs, outcomes, and validation methods, a quantitative meta-analysis was not performed.

Results:

Of the 40 included studies, most were observational and relied on self-reported data from predominantly high-income, technology-literate population. Twenty-four applications incorporated physiological inputs, such as basal body temperature, luteinizing hormone measurements, or wearable-derived metrics, whereas others relied primarily on calendar-based predictions. Multiparameter and sensor-augmented approaches generally demonstrate higher agreement with biological or clinical reference standards than calendar-only methods, with reported fertile window prediction accuracies ranging from approximately 85% to 90% under optimal conditions. However, only a small subset of applications has reported formal clinical validation or regulatory clearance. User satisfaction was strongly associated with perceived accuracy, personalization, and usability, whereas inaccurate predictions, particularly among users with irregular cycles, were linked to frustration, anxiety, and high attrition.

Conclusions:

Menstrual and fertility tracking applications that integrate physiological signals outperform calendar-based approaches in terms of predictive performance; however, robust clinical validation, transparency, and inclusivity remain limited. Reported accuracy metrics should be interpreted cautiously because real-world adherence, irregular cycle patterns, and algorithmic bias substantially affect reliability. These tools are best positioned as decision-support and self-awareness technologies, rather than as autonomous diagnostic instruments. Future evaluations should apply standardized frameworks, such as COSMIN and ISPOR, explicitly communicate uncertainty, and prioritize diverse and irregular cycle populations to ensure equitable and clinically meaningful digital reproductive health solutions.