Background: the consideration of the patient perspective is crucial when developing recommendations. Colorectal cancer (CRC) is one of the most common cancers, has a significant impact on the health of the population and its management poses many situations in which the incorporation of patients' preferences is warranted.
Objective: to describe how colorectal practice guidelines (PGs) incorporate the patient perspective.
Design: we conducted a systematic survey of colorectal cancer guidelines. We searched the GIN library, MEDLINE, the National Guideline Clearinghouse, the National Health Service (NHS) evidence database and Trip-database up to November 2016, and tracked for guidelines updates in January 2018). Two review authors independently assessed the eligibility of retrieved PGs. We tracked included PGs for updates in January 2018. One review author extracted the data and another checked for accuracy. We obtained general information of the guidelines and the method they used to assess the quality of the evidence and to formulate recommendations. Additionally, we analysed the inclusion of patients or their representatives in the guideline development process and methods to incorporate the patient perspective into recommendations. We considered recommendations rated or worded as weak or conditional, or suggesting multiple options, as potentially preference sensitive, and then two review authors independently evaluated if the patient perspective was incorporated. The review protocol was registered in PROSPERO (CRD42018095606).
Results: we included 28 guidelines that contained 588 recommendations, of which 256 were preference-sensitive. One third of the guidelines included patients in the development process. The most common steps with patient involvement were question formulation (40%) and guideline revision (70%). Twelve (43%) guidelines provided information about patients' preferences, obtained mostly through panel’s judgement (58%) or a systematic review of the evidence (50%). One-third of the guidelines included recommendations in which the patient perspective was explicitly considered in their development, and 13 (46.4%) recommended a discussion with the patient. Only one out of 10 potentially preference-sensitive recommendations considered the patient perspective (25/256). The guidelines that included patients through the development process incorporated their perspective more frequently in those recommendations sensitive to their preferences.
Conclusions: guideline users, both clinicians and patients, should be aware that the incorporation of the patient perspective in guidelines is suboptimal. Guideline developers should make greater efforts to incorporate the patient perspective, especially in preference-sensitive recommendations.
Patient involvement: there was no patient involvement for conducting this research.