Article Text

Download PDFPDF

Prospective cohort study
Use of oestrogen only hormone replacement therapy associated with increased risk of asthma onset in postmenopausal women
  1. Julia Dratva
  1. Department of Epidemiology and Public Health, Swiss Tropical and Public Health Institute, Basel, Switzerland
  1. Correspondence to Julia Dratva
    Socinstrasse 57, PO Box, 4002 Basel, Switzerland; julia.dratva{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Commentary on:


There is abundant evidence of the importance of hormonal factors for incidence and course of asthma. However, whether exogenous hormones, such as menopausal hormonal therapy (MHT), have an impact on asthma is still controversial. While an increase of asthma incidence has been observed in the Nurses' Health Study1 and in a population-based Danish study,2 clinical studies have shown favourable effects of MHT on the course of asthma.3 The study by Romieu and colleagues investigated the impact of MHT on asthma incidence in a prospective cohort and potential differences according to the preparations used.


E3N, a prospective French cohort of female teachers, was established in 1990. Participants filled in a biennial self-administered questionnaire on health status, menopausal status and life-style characteristics until 2002. The analytic sample is based on 57 664 postmenopausal women free of asthma at baseline with data on MHT use and age of first use, corresponding to 495 448 person-years of follow-up. Asthma status was defined by American Thoracic Society criteria.

The main statistical analysis was a Cox-Regression analysis (HR). MHT ever-users were divided into recent users and past users. Women were counted as recent users until 1.5 years after cessation of therapy, after which time they were classified as past users. Never users were used as a reference group. The authors further investigated the difference in preparations (oestrogen alone or oestrogen/progesterone), duration of therapy (<2/≥2 years) and method of application (oral/transdermal). Covariates included in the model were tobacco use, body mass index (BMI), total caloric intake, ever use of oral contraceptives, parity and type of menopause. Stratified models were run for atopic status, smoking status (ever/never smokers), type of menopause (surgical/natural) and MHT preparation (oestrogen only or oestrogen/progesterone).


Between 1990 and 2002, 569 incident cases of asthma were reported in the E3N cohort (incidence 1.15/1000 person-years). Ever-users accounted for 60.3% of the person-years (recent use 55.8%, past use 4.5%). Oestrogen only therapy was reported by 11% of the ever-users – more frequently among women with surgical menopause.

The risk of asthma onset was significantly increased among ever-users of MHT compared to never-users (adjusted HR 1.21, 95% CI 1.00 to 1.46). This risk estimate is mainly explained by the increased risk in recent users of oestrogen therapy (adjusted HR 1.67, 95% CI 1.20 to 2.27), compared to past-users and women on oestrogen/progesterone therapy, who showed no increased risk. The study also identified subgroups of higher risk as follows:

  1. Never-smoking MHT users yielded a higher risk for asthma (adjusted HR 1.45, 95% CI 1.10 to 1.90), compared to smokers, in whom no significant effect was found.

  2. High BMI was associated with an increased risk of asthma onset, without evidence of interaction between MHT and BMI.

  3. Atopic women yielded a significantly increased risk, both for oestrogen alone and oestrogen/progesterone (adjusted HR 1.86, 95% CI 1.18 to 2.93; adjusted HR 1.39, 95% CI 1.01 to 1.91), compared to non-atopic women.


In their study sample, the authors found a significantly increased risk of asthma onset in ever-users of oestrogen, whereas only a marginal increase in risk in oestrogen/progesterone users. In accordance with the NHS study, the risk was highest in recent (and current) users4. Allergic disease and lifestyle characteristics, high BMI and never-smoking, were associated with an increased risk. The associations with atopy and BMI are consistent with the literature, although in postmenopausal women, lean women are also at risk. The never-smoker effect may be explained by the high risk attributed to smoking in asthma development, masking the MHT association, or pre-existing respiratory reasons for not smoking.

The analysis is based on a large, prospective cohort, an apparent strength of the study. It is, however, highly homogeneous with respect to socio-economic status, limiting the generalisability and possibly introducing selection bias. It is unfortunate that both exposure and outcome variables were only reported data. Although bias in reporting is improbable, we may assume some misclassification of asthma status and time of onset. Future studies might want to use spirometric data to confirm the results. Sixty-five per cent of the cases occurred in the 1.5 years after cessation. Information on the reasons for cessation might hold a clue for the timing of asthma onset, possibly not related to MHT, or indicate that cessation was linked to respiratory symptoms before the diagnosis of asthma. The analysis unfortunately did not look at age of starting MHT, since time of commencement of therapy could also be related to the incidence of asthma.

The article adds to the evidence of adverse impact of MHT on respiratory health. The finding of a differential effect of preparations is very important for elucidating the mechanisms involved, and it may also help doctors to prescribe the appropriate preparation individually.


View Abstract


  • Competing interests None.