Work-related Dermatoses in Greece.

Abstract

To the Editor The article “Occupational Dermatoses by Type of Work in Greece” by Zorba et al [1], which I read with great interest, adds to the few field studies on occupational dermatoses in Greece [2–4]. Skin disorders are among the most prevalent work-related health problems in Europe, close to infectious diseases, hearing disorders, and cardiovascular disorders, whereas musculoskeletal (42.2%), stress, depression and anxiety (22.5%), and pulmonary disorders (5.9%) ranked higher in the list [5]. The researchers implemented a huge project by the inclusion of 104 small and medium companies spread over all regions of Greece. All these firms were officially served (contracted) for occupational health services by one licensed occupational health physician [1]. The authors argued that these companies were considered to be a representative sample of Greek companies of the same type, regarding size of workforce, years of operation, working practices, and weather conditions, even though no supporting data are given [1]. Very high prevalence rates of occupational dermatoses, especially among hairdressers, cooks, bitumen workers, construction workers, industrial cleaning workers, and farmers were reported [1]. The researchers obtained a response rate of 100%, because—as they argued—all employees of the participating companies were obliged, in accordance with Greek laws, to be examined by the enterprise physician [1]. However, several of the authors' points raised questions and deserve further comments. The authors decided to include in the study only current employees, “each of whom had worked in each type of enterprise for 5 years (no more, no less),” and they randomly selected 4000 of them [1]. I wonder about the total employment force of these 104 small- to medium-sized companies and the employment duration of their employees during the study period; I assume that, in order to gather this sample of the randomly selected 4000 employees, about 15,000 employees or more might have been screened by the researchers in these 104 firms in the 6-year period. And although it seems possible to trace randomly 200 office workers or kitchen staff with “5 years of employment and never previously worked in a similar type of enterprise” in 10 companies, it is far less possible to randomly select 200 paint manufacturing industry workers, bitumen laying workers, or footwear artisan factory workers with the abovementioned employment characteristics in one, two, or four small to medium firms. Detailed data on employment characteristics (total population per company, the percentage of employees with from 1 year up to 5 years of employment, and the new entries during the first 2 years of the study) and on sampling procedure (how and how many employees were randomly selected in each company per year) should be presented to ensure study validity. As far as diagnosis is concerned, in most cases the members of the research team in the 6-year period, recognized the symptoms and signs of the skin diseases and recorded the diagnosis; the authors did not mention if there was any training or measure to account for inter- or intrarater variability in the accuracy and the consistency of the diagnoses during the long study period. In addition, the researchers administered “specially designed occupational skin disease questionnaires,” but no data on the reliability and validity of the questionnaires are given [1]. Interestingly, the authors gathered data on the use of personal protective equipment (PPEs), on occupational stress, etc., but the tools used and the results are not given in the text. Similarly, the authors chose not to describe in detail the less “subjective” results of the patch or skin prick testing, and the same holds for the results of blood examinations that were recommended for all workers with dermatitis, although it is not clear why every worker with dermatitis has to be tested, e.g., for the costly antinuclear antibodies. Last but not least is the paradox that the authors selected identical exposure times for all the participants while they failed to adjust their results for age, the major confounding factor in prevalence studies. Sufficiently sound studies on the prevalence of common work-related disorders in various employment types are always of value in the occupational health domain, because large differences are monitored among countries and various occupational groups [5–7].