Adherence to reporting guidelines in observational studies concerning exposure to persistent organic pollutants and effects on semen parameters.

STUDY QUESTION Are studies on semen quality in men exposed to persistent pesticides reported according to the 'strengthening the reporting of observational studies in epidemiology' (STROBE) recommendations and the guidelines for the appraisal of semen quality studies (SEMQUA)? SUMMARY ANSWER Most studies of the impact of pesticides on semen quality do not follow the STROBE and SEMQUA guidelines, thus adherence is low, especially in methodological aspects. WHAT IS KNOWN ALREADY Much of the controversy about reduced semen quality in recent decades arises from a lack of standardization in the methodology applied, despite the existence of several validated instruments for evaluating the quality of reporting. Indeed, SEMQUA was purpose-designed for the particular characteristics of semen quality studies. STUDY DESIGN, SIZE, DURATION A structured literature search identified eligible articles reporting on persistent pesticides and human semen quality, published in English before 1 September 2012. Opinion articles and reviews were excluded. We assessed the adherence to reporting guidelines of the articles, using and comparing the STROBE statement and the SEMQUA guidelines, in both cases with indicators relevant to observational studies of semen quality. PARTICIPANTS/MATERIALS, SETTING, METHODS A comprehensive bibliographic search in various electronic literature databases using the key words 'sperm' and 'pesticide' obtained 1179 papers, of which 46 were valid for our purposes. The papers examined occupational (26) and environmental exposure (20). Two of the present authors independently piloted the data extraction form for this review. The articles were then evaluated by two researchers using the STROBE and SEMQUA checklists. MAIN RESULTS AND THE ROLE OF CHANGE Although no significant differences were found between the overall degree of compliance with STROBE and SEMQUA (47.0 ± 18.5% versus 43.1 ± 11.6%), there were significant differences when only methodological aspects were considered (48.4 ± 21.0% versus 39.5 ± 17.4%; P < 0.001). We observed an increase over time in the degree of compliance, for SEMQUA (r = 0.61 and P < 0.001) and STROBE (r = 0.45 and P < 0.01). The papers that reported a negative effect of exposure to persistent pesticides on sperm concentration presented a lower level of compliance to SEMQUA (42.1 ± 18.3% versus 57.6 ± 14.2%; P < 0.01) and STROBE (40.2 ± 10.3% versus 49.5 ± 11.6%; P < 0.05) than those which recorded no such influence. The year of publication and the observed effect on sperm concentration were the only candidate variables included in the model of stepwise multiple regression model for the 'degree of compliance' variables of SEMQUA and STROBE. LIMITATIONS, REASONS FOR CAUTION Other characteristics of reporting quality, such as legibility, were not evaluated. WIDER IMPLICATIONS OF THE FINDINGS The low degree of compliance observed is consistent with that observed in other studies of reproductive medicine and highlights the need to improve the design of studies of semen quality. SEMQUA proved to be a more specific tool than STROBE for the field of semen quality. Editors, reviewers and authors should be aware of SEMQUA and apply it when assessing papers on semen quality. STUDY FUNDING/COMPETING INTEREST(S) No research funding was received and none of the authors have any conflict of interests.