Quantity and quality of complementary and alternative medicine recommendations in clinical practice guidelines for type 2 diabetes mellitus: A systematic review

The overall objectives were well-defined in all CPGs. Scores for the CAM sections of the CPGs for the objectives were the same as the overall CPGs. The health questions were also well-presented typically including the intervention or action, the outcome, and the population. The health questions were easy to identify in most CPGs [,,,], while a few were difficult to find [,,]. In four CPGs, scores for the CAM sections for the health questions were similar to the rest of the CPG [,,,]. The others did not clearly mention CAM interventions in the questions [,,]. The population to whom the CPG was meant to apply was also well-stated. All CPGs referred to the target population as being diabetic and sometimes also those at risk of developing diabetes, and in most instances, the target population was only referring to adults. Although severity and comorbidities were not explicitly stated, they were identifiable throughout the CPGs. Three CPGs mentioned that they were designed for populations in certain countries [,,]. None of the CPGs gave specific ages and there was no mention of gender. Considering that the population is the same for the whole CPG, scores of the CAM sections were the same as the rest of the CPG.

All CPGs included the names of the development group. Most CPGs also included the discipline, institution, and location of the development group members, while one did not []. Three CPGs clearly identified the role of the members [,,]. Two CPGs mentioned that their CPG development groups were interdisciplinary or had representation from different subspecialties [,]. One CPG did not provide details of the development group in English []. Four CPGs did not have any CAM members as a part of the development group [,,,]. One CPG included two CAM providers on their development panel, including a chiropractor and a dietician []. One CPG identified those who worked on the CAM section of the CPG but only stated their names []. Four CPGs did not seek the views and preferences of the target population [,,,]. Two CPGs included patient representatives in the CPG development group, but did not include the methods for including them or how their perspectives were taken into consideration [,], while one CPG provided their methods and strategy to gain target population perspectives []. For those CPGs that did seek target population preferences, the CAM sections of the CPGs scored similar or slightly lower since it was assumed the preferences were sought for the whole CPG, however, there was no discussion of specific views or preferences of CAM users. Target users were clearly defined by three CPGs as “all healthcare workers involved” [,,], while one CPG also more specifically stated “primary care physicians across India” []. The three other CPGs did not explicitly state the users [,,]. How the CPG is to be used by the user was clearly stated in four CPGs [,,,]. Only one CPG specifically mentioned dietitians as CAM users of the CPG []. Overall users were the same for the overall CPG; however, since CAM users were not clearly defined, the CAM sections scored slightly lower.

Two CPGs provided detailed search methods [,], however one did not include a full search strategy []. Two CPGs only provided the database used and/or the period that was searched [,]. The methods to search for evidence were not provided in 2 CPGs [,]. The criteria for selecting the evidence was not provided for three CPGs [,,], while others provided little information [,,]. One CPG provided search methods and criteria for selecting evidence through a link, which was inaccessible at the time of assessment []. The scores for the CAM section were also the same for this category. The strengths and limitations of the body of evidence that were collected were described in most CPGs [,,,,], but one CPG did not describe how the evidence was evaluated []. The scores for the CAM section were similar for this section. Most CPGs had little to no information on the methods of formulating the recommendations [,,,,,], while one CPG explained how evidence was used in the recommendation developing process []. The scores for the CAM section were the same for this category. All CPGs considered some health benefits, side effects, and/or risks to some extent when formulating their recommendations. Sometimes they were not very easily identifiable [,,]. The CAM sections of the CPGs received a lower score since they generally had less detailed benefits and harms and sometimes only covered one or the other. One CPG, however, did not include this information for its CAM section []. All CPGs provided an explicit link between their recommendations and the supporting evidence, with links to the references and/or with evidence in the accompanying paragraphs. How the evidence was used to form the recommendations was not always clear. The CAM sections received similar scores. One CPG only had a link to references but no accompanying paragraph for the CAM sections []. Most CPGs stated that they were externally reviewed by experts prior to publication, but did not include some of the following information: outcomes of the review, purpose of the review, or methods employed for the external review [,,,,]. Two CPGs did not state whether they were externally reviewed [,]. None of the CPGs that were externally reviewed mentioned that there were CAM experts as a part of the reviewers. Most CPGs stated that the CPG will be updated. One provided a timeframe and methods [], while others provided only a timeframe [,,]. One CPG only stated that it will be “updated regularly” []. Two CPGs had no information about a protocol for updating [,]. These scores were the same for the overall CPGs and CAM sections.

Generally, the CPGs offered specific and unambiguous recommendations, however, in all CPGs, some recommendations lacked one or more of the following details: identification of the purpose, relevant population, or caveats. The scores for the CAM sections were slightly lower than the rest of the CPG for most CPGs. All CPGs scored highly in presenting different options for the management of the condition or health issue, thus contributing to this high scaled domain percentage. With respect to the CAM section scores, three CPGs had provided information about more CAM therapies [,,], while the others had fewer options [,,,]; one primarily only provided warnings against CAM use []. Key recommendations were also easily identifiable, but only two had a key recommendation section [,]. The other CPGs had sections with grouped recommendations but no specific section for key recommendations [,,,,]. The CAM sections of the CPGs scored similarly to the rest of the CPG, except for four CPGs in which the CAM recommendations were not separated or easily identifiable [,,,].

There was limited discussion about facilitators and barriers in the CPGs, some of which had specific implementation sections, otherwise spread out through the CPG. One CPG had more discussion on dissemination and implementation strategies []. All CPGs included advice and/or tools to support the implementation of the recommendations. Four CPGs contained little discussion on resource implications [,,,], while three had more detailed information on the topic [,,]. Two CPGs provided detailed monitoring and auditing criteria [,]; one of these CPGs only provided this for a subset of the recommendations made []. Other CPGs either did not explicitly state auditing criteria [,,], or were missing some information such as the frequency or specific measurements to monitor [,]. The CAM sections scored low as there was less or no discussion of facilitators/barriers to implementation, tools, resource implications, or auditing criteria specific to CAM recommendations [,,,,,,].