Some important points to consider when planning a review and developing a protocol are as follows.
Review questions should address the choices (practical options) people face when deciding about health care.
Reviews should address outcomes that are meaningful to people making decisions about health care.
Review authors should describe how they will address adverse effects as well as beneficial effects.
The methods used in a review should be selected to optimize the likelihood that the results will provide the best current evidence upon which to base decisions, and should be described in sufficient detail in the protocol for the readers to fully understand the planned steps.
It is important to let people know when there is no reliable evidence, or no evidence about particular outcomes that are likely to be important to decision makers. No evidence of effect should not be confused with evidence of no effect.
It is not helpful to include evidence for which there is a high risk of bias in a review, even if there is no better evidence. See Chapter 8 for a more detailed discussion of bias.
Similarly, it is not helpful to focus on trivial outcomes simply because those are what researchers have chosen to measure in the individual studies (see Chapter 5).
So far as is possible, it is important to take an international perspective. The evidence collected should not be restricted by nationality or language without good reason, background information such as prevalence and morbidity should where possible take a global view, and some attempt should be made to put the results of the review in a broad context.