The extent and causes of natural variation in menstrual cycles: Integrating empirically-based models of ovarian cycling into research on women’s health

Purpose

Menstrual cycle variability has been extensively documented, yet this basic physiological fact has not been well integrated into studies of women’s health. We examine the extent, causes, and implications for clinical research of non-pathological variation in ovarian cycling, and propose guidelines for evaluating the potential contribution of cycle variability to study outcomes.

Major sources of information

This review relied on clinical data accessed through literature searches.

Data synthesis in the model context

Cycle length, occurrence and timing of ovulation, and hormone profiles vary considerably between cycles, women, and populations. The reproductive system is highly responsive to internal and external signals, a consequence of tradeoffs in resource allocation to reproduction versus other bodily functions. Temporary pauses in reproductive effort, which can yield greater lifetime reproductive success, are not necessarily pathological and should, instead, be recognized as a feature of normal reproductive functioning.

Incorporating the new understanding into clinical and/or research relevance

Research on women’s health should incorporate empirically verified biomarkers of cycle physiology and avoid narrow participant inclusion criteria. Cycle length is not an adequate biomarker of either ovulation or progesterone production. Potential cycle-related confounders (cycle phase, hormone concentrations, ovulation status, early pregnancy loss) should be included in research on women’s health.

Conclusions

We can improve our understanding of sex-related differences in the prevalence, severity, diagnosis, and outcomes of disease states, and thereby improve health outcomes for women, through more accurate characterization of menstrual cycle variability and inclusion of relevant empirically grounded cycle biomarkers in research and clinical studies.