[Evaluation of pathologist blinding in medical research proposals of Kerman university of medical sciences]

ABSTRACT

Human behavior is influenced by what we know or believe. In research projects, when the researcher expected a particular finding then explaining the results would be biased. Thus, blinding is a fundamental aspect of all research projects. A retrospective descriptive study was performed. During a 4-year period [2001- 2004] all approved research projects in deputy of research affaires of Kerman University of Medical Sciences that had been referred to pathology laboratories were evaluated. The mechanism used for blinding is questioned by interview with researchers or reviewing the proposals. Only 31% of research proposals had observed blinding policies correctly. Mistakes occurred most often in blinding process were as follow single pathologist who was aware of research design, inappropriate group labeling by alphabetic abbreviations rather than case labeling by randomized numbers, remaining of interventions such as orthopedic devices in the test group, step by step remittance of specimens to pathology laboratory in multiphase studies, etc. Our results revealed that the pathologist should be blinded to group assignment. The blinding process and evaluating techniques should be supervised by statisticians and the monitoring committee [research center]