Investigators followed a nationwide registry cohort of women with Multiple Sclerosis who were aged 40 to 60 at cohort entry. Entry occurred between 2006 and 2024, with outcomes captured through 2025. The study used a new-user design to define exposure; systemic HRT was modeled as a time-varying indicator of current use and in quarterly oral-equivalent dose categories. The report did not supply formulation-level details or specific regimens for hormone preparations. Data sources were Swedish national registries and the Swedish MS Registry; individual-level data cannot be shared publicly under Swedish law, and analysis code is available from the corresponding author on request.
The analytic approach relied on Cox proportional hazards models. Models adjusted for baseline covariates and included time-varying age and disease-modifying therapy (DMT) class. The cohort comprised 7,623 women, of whom 1,192 (15.6%) initiated HRT during follow-up. The authors reported hazard ratios with 95% confidence intervals for the prespecified outcomes.
The study team specified sustained EDSS 4.0 and sustained EDSS 6.0 as primary disability endpoints. Current HRT use was associated with a lower hazard of reaching sustained EDSS 4.0 (adjusted hazard ratio [aHR] 0.70, 95% confidence interval [CI] 0.51 to 0.96). The cohort showed no statistically significant association between current HRT use and sustained EDSS 6.0 (aHR 0.80, 95% CI 0.55 to 1.16).
Case counts and event timing were reported in aggregate by outcome but the source did not provide granular event counts by HRT dose category in the public abstract. The lower hazard for EDSS 4.0 did not translate into a consistent signal at the more severe threshold of EDSS 6.0, a discrepancy the authors reported without asserting a mechanistic explanation. The study did not measure menopausal status directly, which the authors identified as a key limitation in interpreting associations between systemic HRT and disability progression.
Recurring relapses were assessed as a secondary clinical outcome. The investigators observed no association between current HRT use and recurring relapses; the adjusted hazard ratio was 1.00 (95% CI 0.74 to 1.36). This indicates a neutral relapse risk signal in the cohort as modeled.
Investigators also examined quarterly oral-equivalent dose categories for HRT exposure. The dose-category analyses showed no consistent gradient across outcomes in the summary presented. The source did not provide dose-specific absolute event numbers or regimen-specific efficacy or safety data in the abstract, so interpretation of any potential dose–response is limited to the report's summary statement that no consistent gradient was observed.
The study team conducted at least one sensitivity analysis that raised the cohort-entry age floor to 50 years. This sensitivity produced a consistent estimate for the EDSS 4.0 outcome (aHR 0.62, 95% CI 0.40 to 0.95). The report indicates that results for EDSS 6.0 and for relapses remained statistically non-significant in the main analyses; detailed sensitivity outputs and model diagnostics were not included in the abstract.
Models adjusted for time-varying DMT class to account for the contemporary era of high-efficacy disease-modifying therapy; the authors therefore positioned their estimates within current treatment contexts. No trial registration was reported for this retrospective registry analysis, and the source explicitly states that no study protocol was registered.
The authors acknowledged several limitations that constrain clinical interpretation. Menopausal status was not measured in the registries and therefore could not be incorporated into exposure classification or adjustment—this absence limits inference about whether effects differ by menopausal timing or hormonal milieu. The study protocol was not registered prior to analysis, and the abstract did not report formulation-specific HRT details or dose-level event counts. Individual-level Swedish registry data cannot be shared publicly because of national law; researchers may apply for access through Swedish authorities, and access to the Swedish MS Registry requires a separate application.
Ethical approvals were obtained: the Regional Ethical Review Board in Stockholm approved the study and the Swedish Ethical Review Authority approved a subsequent amendment. Individual informed consent was waived because the analysis used pseudonymized registry data. Competing interest disclosures were provided: several authors reported industry relationships and research funding consistent with standard academic disclosure practice; other authors reported no competing interests.
Replication in cohorts with verified menopausal status and, where possible, access to regimen- and formulation-specific HRT data was the specific next research step recommended by the authors. The abstract did not outline regulatory milestones or immediate practice changes; rather, the authors concluded that the absence of increased relapse risk and the association with lower hazard for sustained EDSS 4.0—without a dose gradient and without corroboration at EDSS 6.0—warrants replication with measured menopausal status for clearer clinical interpretation.