Dot is a smartphone app that estimates the menstrual cycle fertile window based on the user’s menstrual period start dates. Dot uses machine learning to adapt to cycles over time and informs users of ‘low’ and ‘high’ fertility days. We investigated Dot’s effectiveness, calculating perfect- and typical-use failure rates.

This prospective, 13 cycle observational study followed 718 women who were using Dot to prevent pregnancy. Participants contributed 6616 cycles, providing data on menstrual period start dates, daily sexual activity, and prospective intent to prevent pregnancy. We determined pregnancy through participant-administered urine pregnancy tests and written or verbal confirmation. We calculated perfect- and typical-use failure rates using multi-censoring, single-decrement life-table analysis and conducted sensitivity, attrition, and survival analyses.

The perfect-use failure rate was calculated to be 1.0%, and the typical-use failure rate was 5.8% for women aged 18–39 (n = 718). Survival analyses identified no significant differences among age or racial/ethnic groups or women in different types of relationships. Attrition analyses revealed no significant sociodemographic differences, except in period, between women completing 13 cycles and those exiting the study earlier.

The study concluded that dot’s effectiveness is within the range of other user-initiated contraceptive methods.