What is new? Describes the rationale for developing well-formulated systematic review questions. Describes how complexity relates to framing a systematic review question. Offers guidance for formulating systematic review questions of complex interventions. Offers suggestions for future methodological research related to framing systematic review questions of complex interventions.
The first and most important decision in preparing a systematic review is to determine its focus. This is best done by clearly framing the questions the reviewer seeks to answer. Well-formulated questions will guide many aspects of the review process, including determining eligibility criteria, searching for studies, collecting data from the included studies, and presenting the findings [1], [2], [3]. In Cochrane reviews, questions are traditionally stated broadly as review “Objectives.” As well as focusing the conduct of the review, the contents of the objectives/questions are used by stakeholders in their initial assessments of whether the review is likely to be directly relevant to the issues they face. In this article, we discuss how to formulate good systematic review questions of complex interventions. Specifically, we (1) describe the rationale for developing well-formulated review questions, (2) review the existing guidance on formulating review questions, (3) discuss complexity in terms of how it relates to formulating review questions, and (4) offer recommendations to assist review authors in formulating a review question for a systematic review of complex interventions.
Current guidance states that a clear and concise statement of a review's objectives (or questions) is critical and should begin with a precise statement of the primary objective, including the interventions reviewed and the targeted problem; ideally, this would be presented in a single sentence [4], [5]. Where possible the style should be of the form “To assess the effects of [intervention or comparison] for [health problem] in [types of people, disease or problem, and setting if specified].”
PICO, although a very useful tool, is not always straightforward to apply, particularly if synthesis teams are interested in the effects of complex interventions. No guidance currently exists on how to formulate review questions of complex interventions. If an intervention is complex, often the synthesis team will need to undertake considerable background work to specify the intervention of interest in detail. For example, typically in reviews of drug interventions, the definition of the
Complexity can come in many forms: it may be related to characteristics of the intervention, the study population, the outcomes measured, or other methodological issues relating to the conduct of primary studies. In this series on systematic reviews of complex interventions, we define complexity using the Medical Research Council guidance, which focuses on the characteristics of the intervention and identifies five key sources of complexity: (1) the number of interacting components within the
In this section, we discuss how three specific complexity issues might impact framing the review question and offer recommendations for overcoming these issues. We discuss complexity issues relating to (1) the type of question, (2) the scope of the review (i.e., lumping vs. splitting), and (3) specification of the intervention.
A few final notes of caution to review authors are presented next. First, developing a good definition of a complex intervention (core to formulating the review question) that can be operationalized within a systematic review will frequently require several iterations. This should preferably include involvement of content experts outside the review team to ensure that the resulting definitions are likely to be robust and meaningful. Second, some of our suggestions, such as conducting scoping
There are many areas where future methodological research aimed at improving question formulation, especially as it relates to complex interventions, would be valuable. Among these are the following:
Empirical research that identifies prototypical and discretionary intervention components of complex interventions, and which components must be fixed vs. adapted locally.
Empirical ways to arrive at dichotomous inclusion/exclusion decisions for complex interventions when boundaries are fuzzy.
