Supported self-management for people with type 2 diabetes: a meta-review of quantitative systematic reviews

Objectives Self-management support aims to give people with chronic disease confidence to actively manage their disease, in partnership with their healthcare provider. A meta-review can inform policy-makers and healthcare managers about the effectiveness of self-management support strategies for people with type 2 diabetes, and which interventions work best and for whom. Design A meta-review of systematic reviews of randomised controlled trials (RCTs) was performed adapting Cochrane methodology. Setting and participants Eight databases were searched for systematic reviews of RCTs from January 1993 to October 2016, with a pre-publication update in April 2017. Forward citation was performed on included reviews in Institute for Scientific Information (ISI) Proceedings. We extracted data and assessed quality with the Revised-Assessment of Multiple Systematic Reviews (R-AMSTAR). Primary and secondary outcome measures Glycaemic control as measured by glycated haemoglobin (HbA1c) was the primary outcome. Body mass Index, lipid profiles, blood pressure and quality of life scoring were secondary outcomes. Meta-analyses reporting HbA1c were summarised in meta-forest plots; other outcomes were synthesised narratively. Results 41 systematic reviews incorporating data from 459 unique RCTs in diverse socio-economic and ethnic communities across 33 countries were included. R-AMSTAR quality score ranged from 20 to 42 (maximum 44). Apart from one outlier, the majority of reviews found an HbA1c improvement between 0.2% and 0.6% (2.2–6.5 mmol/mol) at 6 months post-intervention, but attenuated at 12 and 24 months. Impact on secondary outcomes was inconsistent and generally non-significant. Diverse self-management support strategies were employed; no single approach appeared optimally effective (or ineffective). Effective programmes tended to be multi-component and provide adequate contact time (>10 hours). Technology-facilitated self-management support showed a similar impact as traditional approaches (HbA1c MD −0.21% to −0.6%). Conclusions Self-management interventions using a range of approaches improve short-term glycaemic control in people with type 2 diabetes including culturally diverse populations. These findings can inform researchers, policy-makers and healthcare professionals re-evaluating the provision of self-management support in routine care. Further research should consider implementation and sustainability.

1. In the page 8, line 21, result part. What is this "The PRISMA diagram (Error! Reference source not found.)" Please check the citation and reference in EndNote or RefWorks. In the page 9, line 27. The same errors happened:" Error! Reference source not found. " Page 10 again. Please checked carefully. 2. In the results part. Page 9. The text indicated "The quality of the reviews ranged from 23 [19] to 42[30] from a R-AMSTAR total of 44(supplemental table 3" But in the appendixes, it is showed that "Supplemental Table 3: Summary table of characteristics of included studies, and main findings". Please check and add a table to show the assessment of the quality of systematic reviews included in the present study. The information of heterogeneity of outcomes and score of R-AMSTAR for each systematic review included in the study can be added in proper tables. 3. In page 5, line 21-25, "Supported self-management aims to give people with chronic disease confidence in taking an active role in all aspects of their disease management, and support in choosing healthy behaviors". Therefore, the outcome should include selfefficacy for self-care and self-management behavior. But, in the outcomes, page 12, line 34-56, there are only minority reviews 6/39 (n=8) included in the study showed the outcome self-efficacy and 16 reviews in the study showed the outcome of health behavior. Please add some explanation. 4. Please give a detailed definition of self-management support, which can guide you searching for the targeted studies. Definition of self-management was shown in table 3, what about the selfmanagement support and/or strategy? 5. Discussion part is very simple. It should not only repeat the results and should focus on discussion of these the questions aimed to address: the review questions were: "Do self-management support interventions improve glycaemic, and other physiological outcomes for people with type 2 diabetes in comparison to usual care? What works, for whom, and in what contexts?" page 5, line 40-46.
6. The table number 1,2 in page 27 was different from tables in page 28 and 29 with same title. 7. Could you please add the note under the figure 2-3 to make it read easier? For example what each line represents, the summary effect size for a meta-analysis? What the abbreviations in the figures represent for? Thank you,

REVIEWER
Jennifer Pillay University of Alberta, Canada REVIEW RETURNED 21-Jun-2018

GENERAL COMMENTS
Thank you for the opportunity to review this manuscript. This is welldescribed meta-review and follows standard methodology. There are two main points that I suggest could be addressed to improve the review reporting and findings, especially with respect to the interpretation of what works, for whom and when. 1. The review is missing a description of excluded reviews which would help the reader understand reasons for some exclusions and interpret whether or not the search and selection was adequate. There may be some need to better describe the inclusion/exclusion criteria if this will help. Several reviews I have in my files were not included, although appear relevant, are listed below.
2. While the authors' understandably conclude that there is much heterogeneity in the available RCTs and reviews on their populations and interventions, a few major aspects could be better integrated especially if the author's objective is to describe what works and for whom. Despite many differences in personnel, delivery mode, setting, and populations, some key components across all reviews that are applicable include duration of intervention, complexity of the intervention (single strategy vs mutlicomponent SM program), and co-interventions and/or active comparators. For duration of the intervention, several reviews have found this to be a key moderating factor for effectiveness of these interventions. Interventions range from a couple weeks to many months and this has not been addressed in the review at all, despite several within-review analyses about this. Focusing on duration of followup without this additional piece seems inadequate. This factor is very important for policy makers as well as clinicians and other program implementors. Moreover, some reviews have focused on single strategies (e.g. peer support) while others have focused on complex programs (e.g. culturally competent SM programs including peer support plus interactive sessions and various other behavior change techniques). Further, the relative difference between the intervention groups may heavily rely on the control in the studies. Our experience with these RCTs (Pillay et al. as cited below) is that many of the control groups are actually SM interventions on their own and should be distinguished as such rather than grouped with much more minimal interventions. If there is no reasonable way to separate the effects by comparator, one approach could be to better account for (with critique where suitable) some of the subgroup analyses within the reviews. (e.g,. Palti finding removal of effects of peer support in studies where education was provided to both groups). Grouping the reviews into those studying more vs less intensive interventions may work? The range cited for improvement in HbA1C (-0.06% to -0.53%) is so broad (and likely crosses many people's thresholds for decision making) that I'm not sure this is that useful to the audience without a more comprehensive look at the strongest moderating factors applicable to all studies.
List of possibly relevant reviews: This is a well-written manuscript and the authors have done a lot of work. But there some issues were needed to be addressed.
Thank you 1. In the page 8, line 21, result part. What is this "The PRISMA diagram (Error! Reference source not found.)" Please check the citation and reference in EndNote or RefWorks. In the page 9, line 27. The same errors happened:" Error! Reference source not found. " Page 10 again. Please checked carefully Thank you for bringing this to our attention, the cross-references have been checked and adjusted where needed.
2. In the results part. Page 9. The text indicated "The quality of the reviews ranged from 23[19] to 42[30] from a R-AMSTAR total of 44(supplemental table 3" But in the appendixes, it is showed that "Supplemental Table 3: Summary table of characteristics of included studies, and main findings". Please check and add a table to show the assessment of the quality of systematic reviews included in the present study. The information of heterogeneity of outcomes and score of R-AMSTAR for each systematic review included in the study can be added in proper tables.
Thank you for this, we have now added a Supplemental Table 5 detailing the quality assessment of the systematic reviews.
3. In page 5, line 21-25, "Supported self-management aims to give people with chronic disease confidence in taking an active role in all aspects of their disease management, and support in choosing healthy behaviors". Therefore, the outcome should include self-efficacy for self-care and self-management behavior. But, in the outcomes, page 12, line 34-56, there are only minority reviews 6/39 (n=8) included in the study showed the outcome self-efficacy and 16 reviews in the study showed the outcome of health behavior. Please add some explanation.
We agree with the reviewer that, compared with measures of glycaemic control, self-efficacy and self-management were infrequently measured and rarely the focus of the systematic reviews (only two reviews undertook meta-analysis of self-efficacy). Where systematic reviews reported self-efficacy and self-management behaviours we have described this in the text (page 13) with details in Supplemental Tables 4 and 6. We have now highlighted the research gap identified by this limited evidence on self-efficacy in our discussion of the implications for research. (page 21).
'Included reviews rarely used outcomes such as patient activation or self-efficacy that might have informed the process of behaviour change, suggesting a fruitful research agenda in exploring how people relate to their type 2 diabetes diagnosis and how that influences the optimal timing, delivery, components and overall direction of their self-management.' 4. Please give a detailed definition of self-management support, which can guide you searching for the targeted studies. Definition of self-management was shown in table 3, what about the selfmanagement support and/or strategy?
Thank you for raising this issue. The definition of self-management, self-management support and the components of the PRISMS taxonomy 1 were used together to identify all systematic reviews that focussed on self-management support interventions even if they did not use the term explicitly.
We have now added a definition for self-management support in Table 1 and direct the reader to these definitions early in the methods. The text (page 6) reads: ' Table 1 gives the definitions that we used to identify relevant reviews: in summary, we included reviews of interventions that supported individuals to actively manage the medical, role or emotional components of their type 2 diabetes. 2,3 ' Table 1 gives the following definitions: 'We defined self-management as: "The tasks that individuals must undertake to live with one or more chronic conditions. These tasks include having the confidence to deal with medical management, role management and emotional management of their conditions". 2 This definition implies action on the part of the individual. Therefore, we defined self-management support interventions as any interventions that facilitated individuals to actively manage the medical, role or emotional components of their type 2 diabetes. 3 Interventions that solely provided one-way instructions or education to participants were not classified as selfmanagement support.' We have also made our use of the PRISMS taxonomy of self-management support 1 more explicit in the methods section. The text on pages 7-8 now reads: 'In addition to the definition of self-management and self-management support that were used to select relevant studies (Table 1), we also used the PRISMS Taxonomy of Self-Management Support 1 to identify self-management components within systematic reviews, even if the term "self-management" was not used explicitly' 5. Discussion part is very simple. It should not only repeat the results and should focus on discussion of these the questions aimed to address: the review questions were: "Do self-management support interventions improve glycaemic, and other physiological outcomes for people with type 2 diabetes in comparison to usual care? What works, for whom, and in what contexts?" page 5, line 40-46.
Many thanks for this guidance. We agree about the importance of making the findings of this meta-review as clear as possible for readers. We have addressed this comment in the discussion by using clearer headings and extended our discussion about implementation.
We have restructured the discussion to respond specifically to the review questions using the headings (pages 18-21):  Impact of self-management on glycaemic control  Impact of self-management on secondary outcomes  Implementation: what works, for whom and in what contexts  Implications for research 6. The table number 1,2 in page 27 was different from tables in page 28 and 29 with same title.
Many thanks for pointing this out, we have gone through our manuscript and checked all table numbers and titles 7. Could you please add the note under the figure 2-3 to make it read easier? For example what each line represents, the summary effect size for a meta-analysis? What the abbreviations in the figures represent for?
We have revised figures 2 and 3 to make them clearer for the reader. As the software used for the meta-forest plots was restrictive in terms of formatting, we have re-created the plots in Microsoft Excel so we are able to make these changes. We have also added a footnote to these figures clarifying what the lines represent. This reads:

'Each line represents the summary mean difference and 95% confidence intervals reported by each systematic review'
In addition, this is explained in the text in the overview of results (page 10) where we direct the reader to the figures. This reads: 'Meta-Forest plots (Figures 2 and 3

a-c) illustrate the summary statistics of the included metaanalyses for the primary outcome of HbA1c'
Furthermore, this is reiterated in the Limitations (in the context of explaining why we could not undertake 'meta-analysis of the results of meta-analyses) (page 18) 'Data from commonly cited RCTs were included in several different systematic reviews so that their findings will be presented in several meta-analyses; we recognised this by cataloguing the overlap in RCTs included in the systematic reviews (see Supplemental Figure 1). For example, one RCT was captured in seven meta-analyses. The Forest plots thus illustrate the findings from each meta-analysis rather than summarising them' Reviewer: 2 (Jennifer Pillay) Thank you for the opportunity to review this manuscript. This is well-described meta-review and follows standard methodology. There are two main points that I suggest could be addressed to improve the review reporting and findings, especially with respect to the interpretation of what works, for whom and when.
Thank you 1. The review is missing a description of excluded reviews which would help the reader understand reasons for some exclusions and interpret whether or not the search and selection was adequate. There may be some need to better describe the inclusion/exclusion criteria if this will help.
The 'search and selection' process was undertaken in two phases (the original PRISMS review and the subsequent update). As it is now more than five years since the original PRISMS review, the database of searches and papers rejected at title/abstract and full text screening is no longer available. We have, of course, a full audit trail of the search selection process for the Update.
We have now provided a table of studies excluded at full text screening in the update process (Supplemental table 3) Several reviews I have in my files were not included, although appear relevant, are listed below.
Thank you for providing a list of papers from your files. We have separated the list into those papers that would have been considered in the Update (2012 onwards) where we have a full audit trial and those considered in the original PRISMS screening.
 Seven of the papers listed were considered in the update, six were clearly excluded (See  table below with reasons for exclusion). However, Pillay 2015 should have been included and was indeed in the database for title and abstract screening. We can only conclude that it was screened out by human error. The title is almost identical to a paper reporting findings in type 1 diabetes (which would have been rejected) and we wondered if the reviewer had (incorrectly) rejected it as a duplicate during the screening of 8,404 abstracts. We are grateful for the opportunity to include this title as it was highly relevant. We have modified the main text throughout and modified the PRISMA diagram to integrate this study.
We have also discussed the limitation imposed by only having resources for a single title and abstract reviewer (page 19) 'Title and abstract screening was carried out by one reviewer, increasing the risk of missing relevant papers. Structured training, and random duplicate checking (?95% agreement) was undertaken to maintain quality.'  In the absence of the original PRISMS database of titles and abstracts, we cannot check the process. However, we have checked the 14 'missed' papers listed by the reviewer and provided reasons (see table below) why all but one would have been rejected. We are unsure why the one remaining paper (Fan 2009) was not included, as it appears relevant, though it proved to have the highest risk of bias of any of the included reviews and we wondered if it was rejected because of some methodological concerns. Specifically, the search/selection process is very poorly described and it is thus impossible to be sure how 'systematic' the process was raising the possibility that it was excluded on 4: not a systematic review. Of note, it is described as a 'meta-analysis' rather than a 'systematic review'. After some discussion we decided that we should include it because of the relevance of the topic, and because we could not be sure why it was excluded. We have now integrated it into the manuscript. . Unable to data extract information on RCTs for T2D separately from the rest of the findings * Note: We specified that self-management support interventions would be multi-component, so that a mono-component intervention (e.g. blood glucose monitoring) would be excluded unless it also offered (say) education on self-management actions to take in response to high/low readings.
2. While the authors understandably conclude that there is much heterogeneity in the available RCTs and reviews on their populations and interventions, a few major aspects could be better integrated especially if the author's objective is to describe what works and for whom. Despite many differences in personnel, delivery mode, setting, and populations, some key components across all reviews that are applicable include duration of intervention, complexity of the intervention (single strategy vs multicomponent SM program), and co-interventions and/or active comparators.
Thank you for suggesting this approach to our interpretation. Although, much of the information is inconsistent or conflicting we have now tried to highlight key messages and conclusions that would be helpful for policy makers. To do this, we have revised the results and discussion section to address the characteristics of what works for whom in a clearer and more accessible way. These revisions may be found throughout the result s (on pages 14-17) and in the section in the Discussion headed 'Implementation: what works, for whom and in what contexts' (page 20-21).
See below for detailed response to specific points.
For duration of the intervention, several reviews have found this to be a key moderating factor for effectiveness of these interventions. Interventions range from a couple weeks to many months and this has not been addressed in the review at all, despite several within-review analyses about this. Focusing on duration of followup without this additional piece seems inadequate. This factor is very important for policy makers as well as clinicians and other program implementors. Moreover, some reviews have focused on single strategies (e.g. peer support) while others have focused on complex programs (e.g. culturally competent SM programs including peer support plus interactive sessions and various other behavior change techniques).
We have now specifically addressed the question about intensity of the intervention (frequency and length of contact, duration of the programme, any reinforcement). The findings related to 'dose' of the intervention are summarised on page 14 where the text now reads: 'Generally, review authors concluded that intensity of the intervention influenced effectiveness. Five reviews identified that effective interventions provided moderate or high frequency of contacts, though only two gave specific guidance of 'over 11 hours' or '23.6 hours' to achieve in a 1% (10.9mmol/mol) HbA1c reduction'. Eight reviews recommended longer duration of interventions, however, guidance for the optimal duration varied from three months, over six months to two years with regular reinforcement identified as important in seven studies. Two studies found intense short duration interventions to be more effective if reinforcement was provided'