Clinical question

Scand J Work Environ Health 2002;28(3):197-204    pdf


Evidence-based medicine for occupational health

by Verbeek JH, van Dijk FJ, Malmivaara A, Hulshof CT, Räsänen K, Kankaanpää EE, Mukala K

Objectives This study attempted to determine the feasibility and utility of methods used in evidence-based medicine for some common questions in the practice of occupational medicine.

Methods The following clinical questions were generated that were representative of the type of problems encountered by occupational health physicians: is work a cause of health problems and is impaired health a cause of diminished work capacity for a specific job? Answers were generated according to the method used in evidence-based medicine by formulating an answerable question, searching the literature, critically appraising the results, and applying the results to the clinical question.

Results Answers were found to all the questions in a reasonable amount of time. The searches revealed a need for more systematic reviews and studies that use work-related health outcomes like return to work. However, there is more evidence available in Medline than is generally assumed by occupational health physicians. Using this evidence led to better clinical decisions. Pitfalls during the literature search were typing mistakes, problems in using medical subject headings, and unreliable search strategies. With the use of the abstracts only, most clinical questions could be answered satisfactorily, but concrete risk estimates were often lacking. The lack of availability of full text journals decreased the reliability of the critical appraisal and risk estimation.

Conclusions Evidence-based medicine is a feasible and useful method for occupational medicine. Instruction and training is needed for most occupational health physicians to increase their searching and critical appraisal skills. More research is needed to determine the information needs of occupational health physicians and to develop tools that facilitate literature searches.