Characterizing physiological and symptomatic variation in menstrual cycles using self-tracked mobile-health data

Abstract

The menstrual cycle is a key indicator of overall health for women of reproductive age. Previously, menstruation was primarily studied through survey results; however, as menstrual tracking mobile apps become more widely adopted, they provide an increasingly large, content-rich source of menstrual health experiences and behaviors over time. By exploring a database of usertracked observations from the Clue app by BioWink GmbH of over 378,000 users and 4.9 million natural cycles, we show that selfreported menstrual tracker data can reveal statistically significant relationships between per-person cycle length variability and selfreported qualitative symptoms. A concern for self-tracked data is that they reflect not only physiological behaviors, but also the engagement dynamics of app users. To mitigate such potential artifacts, we develop a procedure to exclude cycles lacking user engagement, thereby allowing us to better distinguish true menstrual patterns from tracking anomalies. We uncover that women located at different ends of the menstrual variability spectrum, based on the consistency of their cycle length statistics, exhibit statistically significant differences in their cycle characteristics and symptom tracking patterns. We also find that cycle and period length statistics are stationary over the app usage timeline across the variability spectrum. The symptoms that we identify as showing statistically significant association with timing data can be useful to clinicians and users for predicting cycle variability from symptoms, or as potential health indicators for conditions like endometriosis. Our findings showcase the potential of longitudinal, high-resolution self-tracked data to improve understanding of menstruation and women's health as a whole.

Dynamic Hormone Control of Stress and Fertility

“…This can be done in combination with a range of tools from network physiology and machine learning that consider the dynamic links between coupled body rhythms, such as body temperature and sleep (Bashan et al, 2012 ; Bartsch et al, 2015 ; Ivanov et al, 2016 ). We anticipate that healthcare technologies, such as wearable devices and smartphone apps collecting vast amounts of data on body rhythms, together with computer algorithms characterizing inter-individual variability, will help refine and personalize neuroendocrinological models (Kim et al, 2020 ; Li et al, 2020 ; Wang et al, 2020 ).…”

Dynamic hormone control of stress and fertility

“…We believe that the first generation mathematical model presented here could be used to inform further investigations into the timing of stress perturbations in reproductive health, including dysregulations induced by strenuous exercise [16], mood disorders [40], as well as clinical interventions such as in vitro fertilisation [41]. We anticipate that healthcare technologies such as wearable devices and smartphone apps collecting vast amounts of data on body rhythms, together with computer algorithms characterising inter-individual variability, will help refine and personalise neuroendocrinological models [42,43].…”

Modelling Menstrual Cycle Length in Athletes Using State-space Models

“…As few apps are accurate in terms of menstrual cycle length prediction 7 , the development of an appropriate accurate parametric model for one-step-ahead forecast cycle length is required. Such a model should take into account the between and within-woman variability to identify menstrual cycle patterns and how each symptoms could affect cycle length, alongside the implications of significant alterations in cycle length According to several studies [8][9][10][11] , the menstrual cycle length can be classified into two groups 'standard' and 'menstrual dysfunction', where a cycle length greater than 35 days is classified as 'menstrual dysfunction' and otherwise as standard. Many statistical models have been proposed in the literature in order to describe these different groups of menstrual cycles 2,[12][13][14][15] .…”

“…Many statistical models have been proposed in the literature in order to describe these different groups of menstrual cycles 2,[12][13][14][15] . Generally, cycle length related to the 'standard' group can be analysed using classical statistical approaches while the mixture of standard and non-standard cycles can be analysed using a mixture distribution to account for the major symmetric distribution and for the component corresponding to the heavy right tail 11,12 . To account for the within individual variability we focused on the dynamic aspect of menstrual cycles over time, as discussed by ref.…”

“…Clearly, self-track symptoms depend on both user engagement, app design, and avoid ambiguous language to describe each level of a symptom (such as flow amount light and spotting). Consequently, to turn it more consistent filtering the original database based on scientific literature is essential way to reduce the effect of bias in the covariates used to fit the model, as described by ref 11. .…”