Chronic obstructive pulmonary disease (COPD) is the leading cause of respiratory death and the third overall cause of death worldwide after ischemic heart disease and cerebrovascular disease.1 Genetic and environmental factors are involved in its pathogenesis, and of these, smoking is the greatest risk factor. However, in both the United States2 and Europe,3,4 about 25% of the adult population with COPD are never-smokers, so the cause of COPD in these individuals needs to be investigated.

Occupational exposure has a major impact on the respiratory health of the population, given the large number of potentially affected people. Most studies have been conducted in populations in the US,5–7 China,8 and in some countries in central9,10 and northern Europe.11–14 Aside from some discrepancies, an association between occupational exposure, COPD risk, and the development of respiratory symptoms has been demonstrated in these populations. Differences in effects may be due to individual factors, differences in work activity and working conditions in different countries. The population attributable fraction (PAF) of occupational exposure in COPD prevalence has been estimated at around 20% in smokers and 31% in non-smokers.15 Occupational exposure has also been linked to the presence of respiratory symptoms, such as cough and chronic bronchitis.16–20

The risk of occupational exposure to respiratory health in Mediterranean populations is little known, and only two studies have been published in the Spanish population. One used a job-exposure matrix to show that exposure to high levels of biological dust was associated with symptoms of chronic bronchitis and worse spirometry results in young people between 20 and 44 years of age from five autonomous communities.21 In a second study, occupational exposure to dust, fumes and gases was associated with symptoms of chronic bronchitis and airflow obstruction in an industrialized region of Catalonia.22

The aim of this study was to analyze the impact of occupational exposure on the risk of developing respiratory symptoms and COPD in a representative sample of the general Spanish population that included subjects of all ages from all regions of Spain enrolled in the third study of COPD prevalence in Spain (EPISCAN II).4

Of the 9092 patients with valid spirometry included in the EPISCAN II study,4 1590 cases were excluded because the occupational exposure questionnaire was not properly completed. Therefore, a total of 7502 individuals of both sexes were included (Fig. 1), with a mean age of 60±11 years, of which 3831 (51.1%) were women and 55.6% were educated to higher than primary school level. One third (34.7%) were former smokers, 19.9% were active smokers, and 17.2% had been exposed at some point in their life to biomass smoke. The characteristics of excluded subjects are shown in Table S1 of the Supplement.

Fig. 1.

Flow diagram of study participants distributed according to COPD diagnosis and occupational exposure.

Exposure was defined as an affirmative answer to the question Have you ever been exposed to vapors, gases, dust or fumes at work? or at least one positive response to the list of substances in the questionnaire.

(0.22MB).

Overall, 3056 (40.7%) subjects reported exposure to VGDF, 851 (11.3%) met COPD criteria on spirometry, and 53.2% reported experiencing respiratory symptoms, the most frequent being wheezing (33.6%), cough (15%), and expectoration (12.6%). A total of 7.9% of the subjects experienced respiratory symptoms during their work activity (Table 1).

The excluded subjects were younger, fewer were men, and they had a lower body mass index and a lower prevalence of COPD and respiratory symptoms compared to participants (Supplement Table S1). Individuals with occupational exposure to VGDF were younger, fewer were educated to university level, and a higher proportion were smokers. Overall, 58.4% of those exposed were or had been active smokers compared to 51% who were never-smokers (p<0.001) (Supplement Table S2).

The most common substances to which the study population had been exposed were solvents (13.1%), chemical irritants (11.5%), diesel or gasoline (9.6%), and detergents (9.2%); when considered individually, no significant differences in exposure were observed between subjects with and without COPD, bearing in mind that the number of subjects exposed to any particular substance was relatively low. The most common occupations were cleaning staff (8.7%), health workers (8.4%), and the armed forces (7.1%) (Table 2).

This study shows that occupational exposure is independently associated with a higher risk of COPD and respiratory symptoms among the Spanish population. Specific associations were also identified between COPD and the steel and petrochemical industries and between respiratory symptoms and a range of substances and occupations. An association between occupational exposure and symptoms manifested during work also emerged.

The impact of occupational exposure in southern Europe has not been fully characterized. This study is the largest conducted to date in the Mediterranean area. Two previous Spanish studies have been reported: one was limited to an industrialized region of Catalonia22 and the other included subjects under 44 years of age from 5 autonomous communities.21 Our study is the first nationwide study with a large sample of subjects of both sexes and a wide age range that is representative of the population as a whole.

The association of occupational exposure to VGDF with COPD, with an OR of 1.22, confirms the risk of certain occupations to the respiratory health of our population. This finding is comparable to that reported in a recent meta-analysis,31 which estimated a global OR of 1.43, a rate slightly lower than that of studies conducted in the United States and northern Europe that reported ORs ranging from 1.6129 to 2.0.5 In our study, the PAF% – or estimated reduction in the proportion of COPD in exposed subjects if the exposure is eliminated – was 8.2%. This estimate varies widely in different studies and its interpretation may be limited in transversal designs such as ours. Some studies based on self-administered questionnaires to determine VGDF exposure have reported higher PAF% values: for example, 17.4%, 20% and 37%,5,14,32 while others reported similar results to ours, such as 9% in Catalonia22 and 10% in China,33 and as low as 5.2% in the Philippines34 and 3.9% in another region of China.35 An ATS/ERS consensus document estimated that the PAF% for exposure to VGDF would be around 14.1%.36 These differences could be due to several factors, such as methodological differences or the characteristics of the population samples, for example, a case–control design or the inclusion of subjects in certain age ranges.14 Another important factor is the percentage of exposed individuals in each population sample studied. Exposure in our study was 40.7%, a percentage comparable to that of similar studies, such as 33%–58% in a European study,18 35% in a US series,29 and 52% in Catalonia.22 These values show that levels of occupational exposure are similar in industrialized countries. Our study offers interesting insights into a little-known factor, namely, the specific association of the steel/petrochemical industry with COPD, as previously observed in a French study.37

Our results indicate that occupational exposure to VGDF is directly associated with general respiratory symptoms (OR 1.45) and with specific symptoms such as cough, chronic bronchitis, expectoration, dyspnea and wheezing (OR 1.3–1.5). These results are consistent with a study in men exposed to VGDF29 that found an OR for cough, expectoration and wheezing of 1.83, 1.84 and 2.01, respectively, and a study conducted in a section of the Spanish population that showed an OR of 2.1 and 1.5 for cough and expectoration, respectively. Another finding worth noting is the association between respiratory symptoms and most of the substances and occupations included in the questionnaire. The agents associated with the highest risks were wood dust and cement, substances to which the working population is commonly exposed.

The association of occupational exposure to VGDF and specific substances and jobs with the appearance of respiratory symptoms in the workplace is a very interesting result of this study and one that has not been reported to date. High risks with ORs of between 6 and 7 have been detected for the steel, petrochemical, shipbuilding, and mining industries. This association underlines the relationship between occupational exposure and respiratory disease. The results mentioned above also suggest that the appearance of these symptoms can indirectly influence work absenteeism and reduce productivity. In fact, there is evidence that exposure to VGDF is associated with absenteeism due to respiratory disease,38 and that people with COPD and asthma have a higher frequency of absenteeism, which affects labor productivity.39

This study has several limitations. Firstly, it is a cross-sectional study, so we cannot determine the course of exposure or establish a causal relationship with COPD and respiratory symptoms. However, we do not believe that a hypothetical change in employment would alter our findings, as patients were questioned about exposures at any time in their lifetime, and a question was included about the need to change occupations due to respiratory symptoms. Secondly, the exclusion of 1590 subjects who had not properly completed their occupational exposure questionnaire may have led to an inclusion bias, and we cannot determine if this possible bias led to an over- or under-estimation of the associations detected. These subjects were older and had a higher prevalence of COPD, so their exclusion is unlikely to weaken the associations. Thirdly, the existence of a memory bias that would lead patients with COPD and/or respiratory symptoms to overestimate previous occupational exposures or symptoms cannot be ruled out. This bias is inevitable in any study based on self-reported exposure. On the other hand, our study has obvious strengths, primarily, in our opinion, its population design with the inclusion of a large number of randomly selected subjects that clearly represent the Spanish population. The rigor of the EPISCAN II study, within which this study is nested, in which post-bronchodilator spirometry was performed according to the highest quality control standards in hospitals throughout the 17 Spanish autonomous communities, ensures the validity of the information on the respiratory health of the participants. The use of a structured occupational exposure questionnaire is another strong point of the study. The questionnaire, in addition to the list of occupations, included a list of professions and risk substances that according to the previous literature, had been associated with COPD; this questionnaire, therefore, offers a detailed overview of the participants’ self-reported occupational exposure. Finally, the high number of adjustment variables, especially the adjustment for tobacco exposure, strengthens the multivariate analyses and increases the strength of the associations found.

This study shows that occupational exposure is associated with COPD and respiratory symptoms in the Spanish population. The results support the available evidence on the need to follow strict prevention measures in our work environment to avoid affecting respiratory health. In the future, this impact should be evaluated progressively in longitudinal studies.

All participants gave written informed consent to participate in the study.

The study was approved by the ethics committee (EC) of each of the participating centers, with the EC of Hospital Universitario La Princesa acting as the reference committee. The EPISCAN II protocol is registered in https://clinicaltrials.gov under no. NCT03028207 and in www.gsk-clinicalstudyregister.com/study/205932.

Both this study and EPISCAN II were sponsored by GSK.

Eduardo Loeb has received honoraria for speaking engagements, attendance at congresses and consultancy from AstraZeneca, GlaxoSmithKline, Boehringer Ingelheim and Pfizer. Jean-Paul Zock has no conflict of interest. Marc Miravitlles has received honoraria for speaking engagements and consultancy from AstraZeneca, Bial, Boehringer Ingelheim, Chiesi, Cipla, CSL Behring, Laboratorios Esteve, Gebro Pharma, Kamada, GlaxoSmithKline, Grifols, Menarini, Mereo Biopharma, Novartis, pH Pharma, Palobiofarma SL, Rovi, Teva, Spin Therapeutics, Verona Pharma and Zambon, and research grants from Grifols. Juan José Soler-Catalonia has received honoraria for speaking engagements from AstraZeneca, Bial, Boehringer Ingelheim, Chiesi, Esteve, Ferrer, GlaxoSmithKline, Menarini, Novartis, Teva, and consultancy fees from AstraZeneca, Bial, Boehringer Ingelheim, GlaxoSmithKline, Ferrer and Novartis. Joan B. Soriano has no conflict of interest. Francisco García-Rio has received honoraria for speaking engagements and consultancy from AstraZeneca, Boehringer Ingelheim, Chiesi, GlaxoSmithKline, Menarini, Novartis, Pfizer and Rovi, and research grants from Chiesi, Esteve, Gebro Pharma, GlaxoSmithKline, Menarini and Teva. Pilar de Lucas has no conflict of interest. Inmaculada Alfageme has no conflict of interest. Ciro Casanova has received honoraria for speaking engagements and consultancy from AstraZeneca, Bial, Boehringer Ingelheim, Chiesi, GlaxoSmithKline, Menarini, Novartis, and research grants from GlaxoSmithKline, Menarini and AstraZeneca. José Miguel Rodríguez González-Moro has no conflict of interest. Julio Ancochea has received honoraria for speaking engagements and consultancy from Actelion, Air Liquide, Almirall, AstraZeneca, Boehringer Ingelheim, Carburos Médica, Chiesi, Faes Farma, Ferrer, GlaxoSmithKline, InterMune, Linde Healthcare, Menarini, MSD, Mundipharma, Novartis, Pfizer, Roche, Rovi, Sandoz, Takeda and Teva. Borja G. Cosio has received honoraria for speaking engagements and consultancy from AstraZeneca, Boehringer Ingelheim, Chiesi, GlaxoSmithKline, Menarini, Novartis, Teva, and research grants from GlaxoSmithKline, Menarini and AstraZeneca. Jaume Ferrer has received honoraria for speaking engagements, attendance at congresses and consultancy from AstraZeneca, GlaxoSmithKline, Boehringer Ingelheim, Chiesi and Novartis. Esther Rodríguez has no conflict of interest.