159 Avoidance of under-reporting and selection bias in occupational injury and illness surveillance system

Introduction Under-reporting of occupational injury and illness (OII) and accompanied selection bias have always been a difficult challenge for policy making and epidemiological studies in many countries, and each country has come up with various reporting mechanisms to obtain reliable data from different sources. The object of this study is reviewing reporting mechanisms in nine countries, including Taiwan, Japan, Korea, Singapore, Australia, the United States, the United Kingdom, Germany, and France, and concluding a generalised model for constructing an ideal OII surveillance system. Methods We carry out institutional comparison of various reporting mechanisms in different countries, both quantitatively and qualitatively, based on yearly statistics on OII from each reporting mechanism, and searchable online information, including statutory laws, official webpages, government documents and journal articles, respectively. Results We categorise various reporting mechanisms into four types, including OSH-Act-based, compensation-based, medical-practice-based, and health-examination-based, and all countries have more than two types of reporting mechanism. Each type has different reporting incentives, coming from fulfilling statutory reporting obligation by employer, claiming insurance benefit by employee, or seeking information feedback for clinical decision-making and clinical studies by physicians, and thus has its own unique strengths and weaknesses in reporting certain kinds of OII. Discussion Due to different coverage of OII, each type of reporting mechanism has an irreplaceable role in OII surveillance. Thus, in addition to the operational effectiveness of individual reporting mechanism, emphasis should be put on inclusion of all four types of reporting mechanism with close coordination and integration in constructing an ideal OII surveillance system. To avoid under-reporting and accompanied selection bias, researchers should also get familiar with characteristics and OII coverage of certain reporting mechanism before the utility of reported statistics from that mechanism, and pooling data from more than one reporting mechanisms may be necessary for a more comprehensive picture of OII.