“Smoking increases heart risk more in women than men,” BBC News has reported. The broadcaster says that a study covering 30 years of research and including 2.4 million people found female smokers have a 25% greater risk than male smokers.
The study was a well-conducted systematic review that pooled the results of 86 smaller studies looking at the risk of coronary heart disease (CHD) in men and women. Combining these results researchers were able to calculate that the risk of CHD in female smokers was 25% higher than in male smokers. The review has many strengths including its large size, its use of checks to ensure that the results of the studies could be accurately combined and the fact that the main findings were based on the data of 75 studies that had accounted for factors such as cholesterol and certain chronic illnesses.
Because of the data available the researchers were not able to say whether the relationship had a biological cause or was due to differing smoking habits in males and females. Further investigation into why this occurs is needed. Regardless of the reasons, it is clear that smoking adversely affects risk of many health outcomes in both men and women, and smoking cessation schemes are beneficial for both women and men.
Where did the story come from?
The study was carried out by two researchers from the University of Minnesota and Johns Hopkins University. It was published in the peer-reviewed medical journal The Lancet. The study received no funding.
The study was covered accurately by the newspapers, which generally gave a good explanation of the research and the balance of risk for men and women.
What kind of research was this?
This was a systematic review and meta-analysis that aimed to investigate whether smoking has the same effect upon risk of coronary heart disease (CHD) in women as it does in men. The researchers aimed to estimate the effect smoking had on coronary risk for each sex once other major risk factors were taken into account.
The review pooled the results of cohort studies that had examined the association between smoking and CHD. A systematic review across the medical literature is the best way of identifying all studies that have examined the association between an exposure and a health outcome. The inherent limitation of all systematic reviews is that the individual studies that they pool may differ in their study population, the way they measured exposure and outcomes and whether or not they adjusted for potential confounders that could affect the relationship between the exposure and outcome. These things can potentially affect the accuracy of any combined risk estimate.
What did the research involve?
The researchers looked across online databases for prospective cohort studies published between 1966 and 2010 that had examined the association between cigarette smoking and coronary heart disease in men and women. To be eligible, studies had to give quantitative estimates of risk and adjust for, at the least, age. The researchers excluded studies in single sex populations or in populations that predominantly included people with a specific disease (for example, diabetes, previous cardiovascular disease or cancer). Where possible they also considered the effect of former smoking.
Their primary outcome of interest was a comparison of the sex-specific relative risk (RR) of CHD (both fatal or non-fatal) in current smokers versus non-smokers. From these risk figures, they estimated relative risk ratios (RRR) between women and men, which means the increased risk that smoking conferred upon women compared with men.
What were the basic results?
In total, 26 articles incorporating data from 86 cohort studies were eligible for inclusion. Two of these articles were themselves reviews that covered data from 60 studies. Studies were international, and varied in follow-up duration from between 5 years in one study and 40 years in one UK study. Some studies only adjusted for the minimum requirement of age, while others variably adjusted for other confounders such as body mass index (BMI), diabetes, cholesterol, blood pressure, other lifestyle factors and sociodemographic variables. The prevalence of smoking (in 21 studies reporting this) varied between 2 and 71% in men and 1 and 44% in women.
The total population across the studies was 3,912,809, among which there were 67,075 CHD events. When only considering the 75 cohorts from studies that had adjusted for other cardiovascular risk factors (covering 2.4 million people, 62% of the entire review population), the pooled adjusted female-to-male RRR of CHD from smoking compared with not smoking was 1.25 (95% CI 1.12 to 1.39). This result suggests that women who smoke have an estimated 25% greater risk of CHD compared with men who smoke.
The researchers found that studies with longer duration of follow-up found higher female-to-male RRR, and that the RRR for women increased by 2% with every additional year of study follow-up (p=0.03). This means that the observed difference in risk between men and women became greater as length of study increased. When analysing 53 studies that had compared CHD risk in former smokers and people who had never smoked, the researchers found no significant difference in risk between men and women.
The researchers found no evidence for publication bias (for example studies being more likely to be published if they found a significant relationship between two factors) and reported that between-study heterogeneity (differences in the nature of study results) was not significant.
How did the researchers interpret the results?
The researchers conclude that smoking raises the risk of CHD to a greater extent in women than men. They consider it to be unclear whether the apparent risk difference between sexes is due to a biological cause or related to differences in smoking behaviour between men and women. They suggest that tobacco-control programmes should specifically consider women, particularly in those countries where smoking among young women is increasing in prevalence.
This was a well-conducted systematic review on gender-related CHD risk in smokers that has searched all relevant literature and analysed data on 3.9 million participants from 86 eligible cohort studies. Combining these results they were able to calculate the difference in risk of CHD from smoking in women compared with men, and found the risk to be 25% higher in women.
The authors’ conclusions from this review are appropriate. It is not possible to tell why there is this apparent difference in risk of CHD between men and women, and whether this could be due to biological differences or to differences in smoking behaviour. In particular, it was not possible to gain information on smoking behaviours from the individual studies, so it is not known how males and females differ in terms of age at onset of smoking, duration of smoking, number of cigarettes smoked or patterns of smoking (for example, smoking at certain times of the day or social smoking when out). This is particularly relevant given that studies were international and smoking practices between men and women can also be expected to differ significantly between cultures, such as in certain Asian countries where women are unlikely to smoke but there are a high proportion of male smokers.
The low level of heterogeneity (differences) between study results is a strength of the review, as is the fact that their main analysis included 75 cohorts that had adjusted for other potential confounders associated with CHD risk (for example cholesterol, diabetes, blood pressure and BMI). Differences between study methods is a limitation of some reviews, so this lack of heterogeneity is important and increases the confidence we can have in the study’s results. However, there are still potential limitations, in particular that it is not possible to tell from individual studies how (or whether) they adequately excluded the presence of CHD at study start, and how they measured CHD outcomes during follow-up. Another important limitation that the researchers highlight is that they were unable to adjust for the use of oral contraceptives, and this is associated with an increased risk of coronary heart disease in women who smoke.
While this study was not able to find the mechanism behind the apparent increased risk of CHD in female smokers, there is clear evidence that smoking adversely affects risk of many health outcomes in both men and women. The review’s finding that smoking may increase risk of CHD in women more than men is worthy of further study to try to investigate the underlying mechanism that may be behind the phenomenon. However, the benefits of smoking cessation should be promoted in all individuals – both women and men.