This paper reviews sources of data typically used in guideline development, available grading systems, their pros and cons, and the methods for evaluating risks of bias in publications, and proposes a revised method for grading evidence and recommendations for use in development of clinical treatment guidelines.
The new World Federation of Societies of Biological Psychiatry (WFSBP) grading system allows guideline developers to follow a multi-step approach of defining levels of evidence, applying criteria for grading (define the acceptability) and the grading of recommendations.
Further, these updated WFSBP recommendations for rating evidence and treatment recommendations provide a grading system that takes into account potential biases in sources of evidence in arriving at final ratings that are likely more clinically meaningful and pragmatic and thus should be used for the development of future treatment guidelines.