Cigarette smoking and age of menopause: A large prospective study

Abstract

Objectives

One of the possible consequences of tobacco consumption is that it contributes to an earlier age of menopause, though the causal relationship is yet to be confirmed. This study aimed to examine the prospective association between smoking and earlier age of menopause in a cohort of middle age Australian women after adjustment for a number of potential confounders.

Study design

21-Year follow-up of a cohort prospective study, Brisbane, Australia.

Main outcome measures

Age of menopause measured at the 21-year follow-up. Smoking and menopausal status were assessed by self-report. Other covariates were measured prospectively in the previous follow-ups.

Results

This study is based on 3545 women who provided data on their menopausal status at the 21-year follow-up of the study, and prospective as well as concurrent data on smoking. In univariate analysis tobacco smoking during the reproductive life course, socio-economic status and gravidity were significantly associated with earlier age of menopause. In multivariate analyses women who smoked cigarettes were more likely to experience earlier menopause than non-smokers. Compared to current smokers, risk of early menopause was significantly lower in those women who quit smoking in the past.

Conclusions

The data suggest that the impact of smoking is independent of other covariates associated with both smoking and age of menopause. The findings raise the possibility that effective quit smoking interventions may lead to a later age of menopause, and reduce the risk of adverse health consequences of early menopause.

Introduction

Early menopause refers to menopause that occurs at or before the age of 45 [1]. Women going through early menopause are at increased risk of morbidity and premature mortality [2], [3], [4]. Early menopause can be spontaneous or induced due to medical or surgical interventions. Over the last few decades, researchers have studied risk factors for early menopause [5], [6], [7], [8]. Cigarette smoking has been recognised as one of the risk factors associated with occurrence of early menopause [7], [8], [9], [10], [11], [12]. However, even though the relationship between smoking and early menopause has been observed in several studies, the association has received little public health attention. This could be due to the difference in the findings of available studies and/or a paucity of knowledge regarding the possible mechanisms [8]. Considering the importance of early menopause for individuals and public health more generally, there is a need to investigate the prospective relationship between smoking and age of menopause independent of other factors that can potentially confound the observed association.

In general, it is suggested that women who smoke tobacco experience menopause one year earlier than non-smokers [7], although a few studies did not support this effect [13], [14]. In addition, the reduction in the age of menopause has been found to be greater among current smokers than those who have previously quit smoking [8]. Using a meta-analysis of 11 observational studies (including about 50,000 participants), Sun et al. [12] found that women who smoked cigarettes were at least 43% more likely to experience menopause before age 50 years. However, they reiterate a need for further evidence to address the existing limitations in the studies that investigated the impact of smoking on age at natural menopause [12].

Some of the previous studies are subject to various limitations that might have resulted in inconsistent findings [8], [15]. First, most available studies are based on a cross-sectional design [5], [13], [14], [16], [17], [18], [19]; cross-sectional studies are not able to accurately assess the temporal relations between the exposure (smoking) and outcome (age at menopause) variables. Second, various studies have used different definitions to measure menopause and/or smoking status [12], [18], [20], [21], [22]. For instance, some of the studies have not reported any definition of menopausal status [23], [24], whereas other research has defined menopause as the cessation of menstruation for a period between six [22], [25] and 12 months [6], [16], [17], [18], [26]. Another limitation concerns the measurement of and controlling for potential confounding factors [16], [25]. For example, Fleming et al. [26], Amigoni et al. [16] and Parazzini et al. [23] adjusted the association between smoking and age of menopause for a few potential confounding factors, while other studies did not [17], [18], [25]. Further, some of the studies did not exclude those who had amenorrhea attributable to medical reasons (e.g. surgical removal of both ovaries) from their analyses [8].

It is hypothesised that the toxic impact of tobacco smoking on ovarian function is non-reversible. If this is true, the risk of early menopause in smokers who quit smoking during their reproductive years should be comparable to those who do not quit. To date there is a shortage of empirical research about the effect quitting tobacco consumption might have on age of menopause [5]. There is a need for a large prospective study to examine the impact of smoking and quit smoking during the reproductive life course on the age of menopause [8].

In general, review of the existing literature on the relationship between smoking and age at menopause points to a need for a large prospective study that accurately assesses both smoking status and age at menopause. The present study aimed to address some of the limitations in the existing literature. It was based on a large prospective cohort that has repeated measures of smoking status from pregnancy to 21 years after the birth. The study aims were two-fold: (1) to examine association of smoking during reproductive life course and age of natural menopause; and (2) to explore if quit smoking reduces the risk of earlier menopause.