Dataset Information

A predictive model for next cycle start date that accounts for adherence in menstrual self-tracking.

ABSTRACT:

Objective

The study sought to build predictive models of next menstrual cycle start date based on mobile health self-tracked cycle data. Because app users may skip tracking, disentangling physiological patterns of menstruation from tracking behaviors is necessary for the development of predictive models.

Materials and methods

We use data from a popular menstrual tracker (186 000 menstruators with over 2 million tracked cycles) to learn a predictive model, which (1) accounts explicitly for self-tracking adherence; (2) updates predictions as a given cycle evolves, allowing for interpretable insight into how these predictions change over time; and (3) enables modeling of an individual's cycle length history while incorporating population-level information.

Results

Compared with 5 baselines (mean, median, convolutional neural network, recurrent neural network, and long short-term memory network), the model yields better predictions and consistently outperforms them as the cycle evolves. The model also provides predictions of skipped tracking probabilities.

Discussion

Mobile health apps such as menstrual trackers provide a rich source of self-tracked observations, but these data have questionable reliability, as they hinge on user adherence to the app. By taking a machine learning approach to modeling self-tracked cycle lengths, we can separate true cycle behavior from user adherence, allowing for more informed predictions and insights into the underlying observed data structure.

Conclusions

Disentangling physiological patterns of menstruation from adherence allows for accurate and informative predictions of menstrual cycle start date and is necessary for mobile tracking apps. The proposed predictive model can support app users in being more aware of their self-tracking behavior and in better understanding their cycle dynamics.

SUBMITTER: Li K

PROVIDER: S-EPMC8714275 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Background and Objectives: Digital health and personalized medicine are advancing at an unprecedented pace. Users can document their menstrual cycle data in a variety of ways, including smartphone applications (apps), temperature tracking devices, and at-home urine hormone tests. Understanding the needs and goals of women using menstrual cycle tracking technologies is the first step to making these technologies more evidence based. The purpose of this study was to examine the current use of these technologies and explore how they are being used within the context of common hormonal and reproductive disorders, like polycystic ovary syndrome (PCOS), endometriosis, and infertility. Materials and Methods: This was a cross-sectional study evaluating menstrual cycle tracking technology use. Participants were recruited in January-March 2023 using social media groups and a Marquette Method instructor email listserv. Data were collected using an electronic survey with Qualtrics. Data collected included participant demographics, menstrual cycle characteristics, reproductive health history, and menstrual cycle tracking behavior. Results: Three-hundred and sixty-eight participants were included in the analysis. Women had various motivations for tracking their menstrual cycles. Most participants (72.8%) selected "to avoid getting pregnant" as the primary motivation. Three hundred and fifty-six participants (96.7%) reported using a fertility awareness-based method to track and interpret their menstrual cycle data. The Marquette Method, which utilizes urine hormone tracking, was the most frequently used method (n = 274, 68.2%). The most frequently used cycle technology was a urine hormone test or monitor (n = 299, 81.3%), followed by a smartphone app (n = 253, 68.8%), and a temperature tracking device (n = 116, 31.5%). Women with PCOS (63.6%), endometriosis (61.8%), and infertility (75%) in our study reported that the use of tracking technologies aided in the diagnosis. Most participants (87.2%) reported a high degree of satisfaction with their use and that they contributed to their reproductive health knowledge (73.9%). Conclusions: Women in our study reported avoiding pregnancy as their primary motivation for using menstrual cycle tracking technologies, with the most frequently used being a urine hormone test or monitor. Our study results emphasize the need to validate these technologies to support their use for family planning. Given that most women in this study reported using a fertility awareness-based method, the results cannot be generalized to all users of menstrual cycle tracking technologies.

Project description:Objective: The menstrual cycle may influence vulvodynia through hormonal pathways or vulvar irritation due to menstruation or menstrual hygiene. We assessed menstrual cycle characteristics in those with and without clinically confirmed vulvodynia. Materials and Methods: Participants were recruited from the administrative database of a health care network serving ∼27% of Minneapolis-Saint Paul residents. For 220 clinically confirmed cases and 224 controls, menstrual cycle characteristics were retrospectively assessed at three time points: before age 18, the year before onset of vulvar pain, and 3 months before study participation. Logistic regression was used to calculate odds ratios (ORs) and 95% confidence intervals (CIs) for the associations between menstrual characteristics at all three time points and vulvodynia. Models adjusted for prespecified confounders were evaluated against crude effect estimates. Results: Women with heavier menstrual flows had higher odds of vulvodynia compared with women with lighter menstrual flows during their adolescent years (OR 1.62, 95% CI 0.91-2.86), the year before onset of vulvar pain (OR = 2.11, 95% CI 1.10-4.02), and during the 3 months before study participation (OR = 1.67, 95% CI 0.91-3.06). Women with more severe cramps also had higher odds of vulvodynia compared with women with no or mild cramps during their adolescent years (OR = 2.45, 95% CI 1.45-4.15), the year before onset of vulvar pain (OR = 3.30, 95% CI 1.67-6.51), and during the 3 months before study participation (OR = 4.96, 95% CI 1.99-12.36). Women with specific premenstrual symptoms also reported higher odds of vulvodynia. Among those with vulvodynia, half reported a change in vulvar pain across the menstrual cycle, with 60% of these reporting greater pain just before and during menstruation. Furthermore, we observe a trend of decreased tampon use and increased use of sanitary pads as women with vulvodynia moved closer to their date of diagnosis. Conclusions: Menstrual cycle characteristics were associated with vulvodynia, and associations were consistent across different phases of the reproductive life cycle.