The EPI-SCAN study was conducted based on the following sample size calculation: a COPD prevalence of 12% was estimated, with precision ±1%. Assuming 20% losses, the theoretical number of individuals for inclusion in the study was 5071. Considering a posteriori that among the 11 areas in EPI-SCAN, the mean number of participants was 345 per area, with a maximum of 439 in Burgos and a minimum of 136 in Asturias, there was statistical power for most area comparisons with respect to the mean. As this was a descriptive population study, participation quotas by sex or other variables were not set. The estimator and its 95% confidence interval are presented in most of the analyses. The results for each variable are presented as mean and standard deviation in the case of continuous variables, and using the number of cases in each category and the relative frequency of the total number of responses in the case of categorical variables.

To quantify the number of COPD cases in the population, the rates obtained in EPI-SCAN were standardized according to the age and sex distribution of the Spanish population from official statistics.20 Statistical significance was analyzed by performing an initial ANOVA and then a two-sided test for continuous variables and a Chi2 test for categorical variables.

Finally multivariate analysis by logistic regression of the EPI-SCAN variable “being previously diagnosed with COPD” in the subsample of COPD individuals according to GOLD is presented, exploring, in addition to sex, the following factors by relative risk and 95% confidence interval (RR, 95% CI): age, smoking, education level, mMRC dyspnea, GOLD COPD severity, Charlson index and having performed previous spirometry. A statistical significance level of less than 0.05 was used in all statistical tests.

In Spain, the IBERPOC study, conducted with a similar methodology 10 years before EPI-SCAN, observed a prevalence of COPD in women of 3.9% using the lower limit of normal,10 which is almost equal to the 3.5% observed in EPI-SCAN with the same criteria. In IBERPOC, 22% of COPD cases detected were women, compared with 29% in EPI-SCAN. It was also observed in the IBERPOC study that being female was significantly and independently associated with a higher risk of not being previously diagnosed with COPD.10 Comparing these new results with the IBERPOC results, women continue to have a 1.3-fold increased risk (95% CI 0.3–6.2) of having underdiagnosed COPD. In the IBERPOC study, a large percentage of newly diagnosed COPD cases occurred in women with little exposure to tobacco smoke and respiratory symptoms. Therefore, it can be concluded that the situation has not improved in 10 years. In comparison with the PLATINO study, where underdiagnosis of COPD exceeded 88%, the factors associated independently with underdiagnosis were younger age, less severity of airway obstruction (lower GOLD severity), fewer respiratory symptoms (cough, sputum, wheezing and difficulty breathing), and no prior diagnosis of asthma; however, women in the PLATINO study did not show significant differences in underdiagnosis of COPD with respect to men, with 1.38 (95% CI 0.76–2.52, P=.29).22 The PLATINO study including other population studies (mostly in a recent review in these pages)7 indicates that overall the prevalence of COPD in women is lower than that in men. For example, in the multicentre BOLD study,14 the prevalence of COPD in women varied between 7.6% in Guangzhou, China and 25.7% in Salzburg, Austria, while the prevalence of COPD in men varied between 15.3% in Guangzhou, China (i.e. twice that in women there) and a maximum of 28.7% in Capetown, South Africa.14

The last Spanish National Health Survey in 2009 estimated the prevalence of smoking at 26.4% of the population aged over 16 years, with a clear difference between men (31.17%) and women (21.33%), regardless of the Autonomous Region.23 In the absence of more recent data, a slow progressive decline in the number of smokers has been observed in the general population in recent years,24 and the mean age for starting smoking is an average of 21.32 years (20.76 in men and 22.10 in women), which, compared with the 2006 National Health Survey, shows that an average delay of almost four years has been achieved in the age at which individuals start smoking; this is greater than four years in the case of women (in 2006, it was 17.04 in men and 17.78 in women).

It is estimated that more than 18 thousand people in Spain die each year due to COPD, of which one quarter (more than 4500) are women.25 In Spain, the latest data available are from 2009. According to these data, COPD is the fifth leading cause of death among men, with an annual rate of 44.0 deaths per 100000 persons, and the seventh leading cause for women, with an annual rate of 14.2 deaths per 100000 persons. In comparison, Spain is in the intermediate range within the European Union, headed by some Eastern European and Anglo-Saxon countries. Among the 19 European countries with available data, we are in eighth position in terms of mortality due to COPD in men and are among the lowest in women.26 The morbidity and use of healthcare resources associated with COPD in Spain are high.27

Irrespective of the disease, the estimated prevalence in the population depends on the definition used for its diagnosis. In the case of COPD, different approaches have historically been used, including the following: medical diagnosis, diagnosis based on the presence of respiratory symptoms, and diagnosis based on the presence of airflow limitation (with or without a bronchodilator test). Each of these methods will produce very different results. Up until 2000, international COPD guidelines such as those of the European Respiratory Society28 recommended different spirometric thresholds for defining COPD in women, such as FEV1/FVC ratio <88% in men and <89% in women. Nowadays, the use of new sex-specific reference equations29 has eliminated this distinction.

Currently in Spain, more women (42.7%) than men (40.8%) smoke between the ages of 16 and 24 years.23 This is a new and disturbing phenomenon in the history of tobacco use in Spain, which has also been observed in the young population in other developing countries in recent years. In the 25–44 years old age group, almost half are smokers (52.6% of men vs 43.5% of women). Proper implementation of recent legislative measures such as Act 42/2010,30 which came into effect on 2 January 2011, limiting the use of tobacco in public places and its advertising, should have an impact not only on greater protection for non-smokers, but also on a future reduction in smoking in the Spanish population, as has been observed in Ireland, Italy and California. Furthermore, Spanish population growth is expected to peak in 2050, with 53 million inhabitants, with peak aging around 2060,31 resulting in an increase in the population at risk of developing irreversible airflow obstruction. Our results agree with those obtained recently by Naberán et al.,32 who concluded in a patient population in Primary Care centers and hospitals in Spain that women with COPD were younger, smoked less and had better lung function than men; they also observed that they had poorer quality of life (measured by EuroQoL and AQ20) and more often associated anxiety and depression with COPD. Some studies established differences in the clinical presentation in women and men with diagnosed and underdiagnosed COPD.7 EPI-SCAN found that 55.3% of women with COPD had never smoked; this percentage is even lower than the 75% observed in IBERPOC.33 The IBERPOC results also showed that women with COPD, in addition to lower exposure to tobacco smoke than men, had less impairment of spirometric parameters, fewer symptoms and better quality of life.10

The use and interpretation of spirometry as a tool for defining COPD are different at population compared to individual level. For example, in the population strategy, of all the women in EPI-SCAN with spirometry consistent with COPD, it may be surprising that more than half reported that they had never smoked. At clinical level, a doctor would explore the cause of this airflow limitation (passive smoking, other occupational exposures, comorbidities, etc.) and would start classifying his/her symptoms to diagnose, and if need be, initiate treatment. Unfortunately, it seems that, to date, population-based epidemiological initiatives (or case-finding initiatives in sub-populations at risk) for COPD screening have only provided partial solutions to the still high underdiagnosis of COPD, which continues to be a mainly female problem.34

The general perception of physicians of women with respiratory symptoms and suspected COPD must improve, and informative and educational measures on the importance and different characteristics of COPD in women would help to reduce underdiagnosis and its undertreatment.3,31

Monitoring of risk factors and generation of new epidemiological data on COPD and its determinants by sex should be considered a research priority in Spain. In comparison to other chronic diseases, there are few data with respect to the distribution and characteristics of COPD in women. However, in the light of IBERPOC, EPI-SCAN and other studies, we can now say that women are already and will increasingly become the object of medical care for COPD. This implies early diagnostic strategies that are appropriate for women. As with smoking, it seems that women have a higher susceptibility to inhaled industrial pollutants for developing respiratory diseases in general, compared to men. The efficacy of treatments in COPD may be different, especially with respect to smoking cessation and pulmonary rehabilitation. As regards respiratory drugs, there were no differences in efficacy initially,35 although new drugs appear to be less effective in women.36 As has previously been reported, although the clinical presentation of COPD in women may have slight differences, there should be greater emphasis on comorbidities in women, such as osteoporosis, anxiety and depression as accompanying elements.37 The population shift toward chronic illnesses with the triad of aging, comorbidities and polypharmacy38 can already be applied to COPD, along with increased relative feminization.