Suggestions for better data presentation in papers: an experience from a comprehensive study on national and sub-national trends of overweight and obesity.

BACKGROUND The importance of data quality whether in collection, analysis or presenting stage is a tangible and undeniable scientific fact and the main objects of researches implementation. OBJECTIVE This paper aims at explaining the main problems of the Iranian scientific papers for providing better data in the field of national and sub-national prevalence, incidence estimates and trends of obesity and overweight. METHODS To assess and evaluate papers, we systematically followed an approved standard protocol. Retrieval of studies was performed through Thomson Reuters Web of Science, PubMed, and Scopus, as well as Iranian databases including Irandoc, Scientific Information Database (SID), and IranMedex. Using GBD (Global Burden of Diseases) validated quality assessment forms to assess the quality and availability of data in papers, we considered the following four main domains: a) Quality of studies, b) Quality report of the results, c) Responsiveness of corresponding authors, and d) Diversity in study settings. RESULTS We retrieved 3,253 records; of these 1,875 were from international and 1378 from national databases. After refining steps, 129 (3.97%) papers remained related to our study domain. More than 51% of relevant papers were excluded because of poor quality of studies. The number of reported total population and points of data were 22,972 and 29 for boys, and 38,985 and 47 for girls, respectively. For all measures, missing values and diversities in studies' setting limited our ability to compare and analyze the results. Moreover, we had some serious problems in contacting the corresponding authors for complementary information necessary (Receptiveness: 17.9%). CONCLUSION As the present paper focused on the main problems of Iranian scientific papers and proposed suggestions, the results will have implications for better policy making.