Digital health app data reveals an effect of ovarian hormones on long COVID and myalgic encephalomyelitis symptoms

BackgroundLong COVID and myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) disproportionately affect females, suggesting modulation by sex hormones. We investigated whether symptom severity is influenced by changes in sex hormones over the menstrual cycle, or by hormonal contraception. MethodsWe carried out a retrospective analysis of menstrual and symptom data, prospectively collected via the Visible app from individuals with long COVID, ME/CFS, or both, who had regular menstrual cycles (7th September 2022-6th March 2024). We examined associations between symptom severity, menstrual cycle phase and contraception type using mixed-effects models. Prospectively collected menstrual and symptom data from users of the Hertility menstrual cycle tracking app (7th September 2022-28th September 2025) were used as a population comparison cohort. Findings948 users were included in the disease cohort; 100% of users were female, 92.6% identified as women, 6.4% as non-binary, 0.6% preferred not to say and 0.3% self-described. Crashes (sudden and severe worsening of symptoms following exertion) were significantly more frequent during menstruation than other phases, most notably compared with the late luteal phase (odds ratio (OR) = 0{middle dot}888, 95% confidence interval (CI): 0{middle dot}838-0{middle dot}941). Users of combined hormonal contraception (n=70) had reduced crash incidence (OR = 0{middle dot}548, 95% CI: 0{middle dot}350-0{middle dot}856) and overall symptom score (incidence rate ratio = 0{middle dot}827, 95% CI: 0{middle dot}690-0{middle dot}992) compared to those not using contraception (n=786). Fatigue, brain fog, and headaches showed phase-specific variation in both disease and population comparison cohorts. InterpretationMenstruation is associated with increased crash incidence and worsened symptoms in long COVID and ME/CFS. Users of combined hormonal contraception report reduced crash incidence and lower symptom burden, suggesting a modulatory role of ovarian hormones. Analysis of the population comparison cohort indicates that some cyclical symptom variation may reflect normal menstrual-related fluctuations, albeit that symptoms are less frequently reported in the general population. Our models adjusted for factors including individual, disease type, age and gravidity, however, residual confounding accounting for differences in disease severity between contraceptive users and non-users cannot be excluded. Nevertheless, these findings could empower menstruating people living with long COVID and ME/CFS to anticipate cyclical changes in symptoms and plan their activities accordingly and could also inform their use of contraception. FundingThis study was funded by a UK Medical Research Council (MRC) iCASE PhD studentship to Abigail Goodship (MR/W00710X/1).