WHO-Recommendations not Based on Best Evidence
21 May, 2007 12:41 pm
Guideline development groups in the World Health Organization rarely base their recommendations on systematic reviews of research findings, but rather rely heavily on the opinions of topic-experts. This is one of the main findings in a study conducted by researcher Andy Oxman and colleagues, published in The Lancet May 9th.
The interviews took place in September 2003 and February 2004.
The topics for the guidelines and recommendations varied widely. In addition to clinical subjects the recommendations covered matters such as promotion of mental health, human resources for policy development, model list of essential medicines, tobacco legislation and bioterrorism. The most common reasons for developing recommendations were a perceived need for guidance, perceived need to update existing guidelines, and demand from member states.
Convening expert committees was the standard procedure when developing recommendations in most departments, and external consultation was a common component of the development process. Few directors had developed dissemination or implementation plans.
Many directors reported that they had used background documents to inform the work of the expert committees, but there were no consistent methods for preparing such documents. For instance, the background documents could be selected by the participating experts according to their own standards. A small number of the directors mentioned using systematic reviews and only one reported grading the quality of the evidence.
Costs were often taken into consideration, but other value judgements, such as weighing potential harms against potential benefits were rarely addressed explicitly. Possible harmful consequences were only mentioned for clinical interventions, not for public health or policy interventions. In fact, one director was quoted saying that "No harms are likely, since the recommendations were made by the top experts".
The descriptions the directors provided of the group processes in play suggested that the participants were implicitly weighing evidence of benefits and harms along with values and ethical considerations, as one interviewee is quoted saying: "There is a tendency to get people around the table and get consensus—everything they do has a scientific part and a political part. This usually means you go to the lowest common denominator or the views of a "strong" person at the table."
When asked whether the recommendation-development process could have been improved, several directors pointed at the use of evidence, for instance: "I would have liked to have had more evidence to base recommendations on. We should have conducted a literature search." Also the lack of dissemination and implementation plans was commonly mentioned: "We published it, but just left it there.... The recommendations were never transformed into a programmatic approach. It is a common in-house failure to transform recommendations into action."
In 2003 WHO developed a set of in-house standards for guideline-development that were consistent with similar documents from other organisations. The findings from this study clearly indicate that these standards are not adhered to, and only two of the directors had plans to use the guidelines for WHO guidelines. However, several interviewees recognised the need for more systematic and transparent guideline development processes and believed that the organisation was moving in that direction. One director was quoted saying that: "We are in the middle of a process, which needs time. There is increasing understanding of the need for evidence-based guidance and it is becoming part of the WHO culture."
In response to the study findings the WHO leadership announced the establishment of a Guidelines Review Committee. The move was applauded in a Lancet-editorial on May 19th: "The creation of a Guidelines Review Committee is good news for WHO, and provides an example of transparency to the UN, and of accountability to its stakeholders. The World Health Assembly, convening in Geneva this week, can take pride in these actions, but must recognise that the synthesis, implementation, and evaluation of science-based guidelines will require considerable resources if the commitment is to be sustained as a core activity."
Oxman AD, Lavis J, Fretheim A. Use of evidence in WHO recommendations. Lancet 2007, published online May 9, DOI: 10.1016/S0140-6736(07)60675-8.