Background
The role that sex hormones play in asthma remains unclear. The oral contraceptive pill (OCP), commonly used by younger women, acutely increases sex hormones providing an opportunity to observe their effect.

Objective
Evaluate the association between OCP and asthma attacks.

Methods
Using the UK's Clinical Practice Research Datalink, linked to hospital admission and mortality data, 2004 to 2020, we observed women with asthma (18–50 years), comparing OCP never-users to new-users; separated into a combined oral contraceptive (COC) cohort and progestogen-only (POP) cohort. We applied inverse-probability of treatment weighting and Cox proportional hazards, accounting for demographics, asthma severity/control and comorbidities. Additionally, we stratified by potential modifiers: age, BMI, blood eosinophils (x109 cells·L−1, normal <0.3, eosinophilia ≥0.3) and corticosteroid use (lower use: ≤3 inhaled corticosteroids prescriptions, higher use: ≥4 inhaled and/or oral corticosteroids).

Results
132 676 and 129 151 were eligible for the COC and POP cohorts, respectively. There was no association between COC, or POP, and asthma attacks (weighted-HR, 95% CI: COC=1.00, 0.89–1.13; POP=1.11, 0.97–1.28). However, POP association was modified by asthma phenotype and corticosteroid use, but not BMI, after accounting for asthma severity/control, demographics and comorbidities. In the POP users, women who were younger than 35 years old (1.39, 1.12–1.72), those with eosinophilia (1.24, 0.97–1.58), or those with lower corticosteroid use (1.20, 1.03–1.40) had an elevated risk of asthma attacks.

Conclusions
Commencing exogenous progesterone without an estrogen component (POP) was associated with increased asthma attacks in asthma phenotypes including in younger women, eosinophilic-asthma, and women with lower corticosteroid use.