JMIR Publications

ABSTRACT

Background:

By allowing users to track their menstrual periods and the associated signs and symptoms, Menstrual Cycle Tracking Apps (MCTAs) may have potential in epidemiological studies of women’s health. However, there is limited information on the characteristics of MCTA users compared to both those who track their cycles in other ways and those who do not track their cycles.

Objective:

We sought to examine differences between individuals tracking their cycles through MCTAs (app-users), individuals tracking through means other than MCTAs (other-trackers), and non-trackers in the Ovulation and Menstruation Health Pilot Study survey

Methods:

Our study collected survey responses from 263 participants recruited from a gynecology clinic (n=36), a community fair (n=61), and the internet (n=166). We then compared differences in demographic/socioeconomic/lifestyle factors, health conditions, and menstrual cycle characteristics between MCTA users and non-app users and between women who tracked their cycles (with any forms of tracking) and those who did not.

Results:

Overall, participants were mostly white, had attained 4-years of college education or higher, and had a median annual household income between $50,000-$74,999. Among all participants, 103 were MCTA users (app-users), 62 used other forms of cycle tracking (i.e., paper, digital calendar, software, birth control, and memory) (other-trackers), and 95 did not engage in any form of tracking (non-trackers). No meaningful differences in race/ethnicity, household income, and education level between app-users and other-trackers and between other-trackers and non-trackers were found. Menstrual cycle characteristics, such as cycle length, frequencies of normal and abnormal cycles, time to cycle regularity, and age at menarche were also comparable between app-users and other-trackers and between other-trackers and non-trackers. For self-reported factors, other trackers compared to app-users were more likely to smoke (18% vs 6%), use hormonal birth control (89% vs 72%), and had a higher prevalence of endometriosis (5% vs 1%) but a lower prevalence of heartburn (17% vs 26 %). Other trackers had a higher proportion of cholesterol (14%v 7 %), diabetes (5% v 0 %), and anxiety disorder (41% v 27%) compared to non-trackers in this study.

Conclusions:

Our results suggest that app-users and other-trackers and other-trackers and non-trackers are largely comparable in demographic factors, socioeconomic factors, and menstrual cycle characteristics. Non-trackers, however, may potentially have a slightly differing disease burden compared to MCTA users. Our findings strengthen the generalizability of epidemiological results drawn from studies utilizing MCTAs. Within studies using MCTA sourced data, the addition of survey questions or health data may further facilitate whether differences exist between MCTA sourced data and comparison populations