Patient Education and Counseling 104 (2021) 760–769

Contents lists available at ScienceDirect

Patient Education and Counseling

journal homepage: www.elsevier.com/locate/pateducou

Review Article

The effectiveness of peer support on self-efficacy and self-management

in people with type 2 diabetes: A meta-analysis
Dandan Lianga , Ruiying Jiaa,1, Xiang Zhoua , Guangli Lub,* , Zhen Wua , Jingfen Yua ,
Zihui Wanga , Haitao Huanga , Jieyu Guoa , Chaoran Chena,*
a
Institute of Nursing and Health, College of Nursing and Health, Henan University, Jinming Avenue, Kaifeng, Henan, 475004, China
b
Institute of Business, School of Business, Henan University, Jinming Avenue, Kaifeng, Henan, 475004, China

A R T I C L E I N F O A B S T R A C T

Article history: Objectives: This study aims to investigate the effectiveness of peer support on self-efficacy and self-
Received 13 July 2020 management in people with type 2 diabetes.
Received in revised form 3 November 2020 Methods: Eight databases were utilized for selecting eligible studies that were published from inception
Accepted 7 November 2020
to Jan., 2020. The eligible studies were screened, extracted and then the methodological quality was
evaluated independently by two researchers. RevMan version 5.3 software and Stata version 14.0
Keywords: software were utilized for the meta-analysis.
Peer support
Results: Seventeen studies were included in the meta-analysis. Compared with the control group, peer
Type 2 diabetes
Self-efficacy
support significantly improved self-efficacy [SMD = 0.41, 95 % CI = (0.20, 0.62), p = 0.0001] and self-
Self-management management [SMD = 1.21, 95 % CI = (0.58, 1.84), p = 0.0002] in people with type 2 diabetes, but had no
Meta-analysis significant effect on distress (p = 0.34).
Conclusions: Peer support significantly improved self-efficacy and self-management, but there was no
clear evidence that peer support improved distress in people with type 2 diabetes. More studies are
needed to further verify the validity of the results.
Practice implications: This meta-analysis suggested that peer support should be considered as a
complementary treatment for patients with type 2 diabetes. Medical staff can encourage the use of peer
support in the teaching content of patients with type 2 diabetes to improve their self-efficacy and self-
management.
© 2020 Elsevier B.V. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ............... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 761

2. Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ............... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 761
2.1. Search strategy . . . . . . . . . . . . . . . . . . . . . ............... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 761
2.2. Inclusion and exclusion criteria . . . . . . . . ............... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 762
2.3. Study selection and data extraction . . . . . ............... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 762
2.4. Assessment of risk of bias . . . . . . . . . . . . ............... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 762
2.5. Data synthesis and analysis . . . . . . . . . . . ............... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 764
3. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ............... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 764
3.1. Study selection . . . . . . . . . . . . . . . . . . . . . ............... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 764
3.2. Study characteristics . . . . . . . . . . . . . . . . . ............... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 764
3.3. Risk of bias . . . . . . . . . . . . . . . . . . . . . . . . ............... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 764
3.4. The results of meta-analysis . . . . . . . . . . . ............... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 765
3.4.1. Meta-analysis of peer support on self-efficacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 765
3.4.2. Meta-analysis of peer support on self-management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 766

* Corresponding authors.
E-mail addresses: [email protected] (G. Lu), [email protected] (C. Chen).
1
Ruiying Jia was the co-first author.

https://doi.org/10.1016/j.pec.2020.11.011
0738-3991/© 2020 Elsevier B.V. All rights reserved.
D. Liang et al. Patient Education and Counseling 104 (2021) 760–769

3.4.3. Meta-analysis of peer support on distress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 766

3.5. Sensitivity analysis and publication bias assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 767
4. Discussion and conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 767
4.1. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 767
4.1.1. Summary of main findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 767
4.1.2. Agreements and disagreements with other meta-analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 767
4.1.3. Strengths and limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 768
4.1.4. Implications for future research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 768
4.2. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 768
4.3. Practice implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 768
Authorship . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 768
Funding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 768
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 768
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 768

1. Introduction emotional support, (3) linkage to clinical care and community

resources, and (4) ongoing availability of support [23]. Since peer
Diabetes is a group of metabolic diseases characterized by support leverages the ability of people with diabetes to support
hyperglycemia due to defective insulin secretion and/or impaired each other in promoting healthy lives [24], it not only releases the
biological effects [1]. The World Health Organization (WHO) pressure on health systems, but also reduces financial burden
reported that 171 million people in the world had diabetes in 2000 [25,26].
[2]. In 2019, the International Diabetes Federation (IDF) estimated Previously, there were five similar meta-analyses that focused
that approximately 463 million adults worldwide had diabetes and on the effects of peer support in people with type 2 diabetes [27–
this number is expected to increase to 700 million by 2045 [3]. 31] and yielded significantly different/controversial results. Kong
Diabetes, if not well treated, could lead to serious complications et al. [30] found that peer support did not significantly improve the
that affect the circulatory and nervous systems, kidneys, eyes, and self-efficacy of people with type 2 diabetes; Zhao et al. [28]
feet [4], thus bringing serious physical and psychological harm. indicated that peer support did not improve self-efficacy and self-
Diabetes is classified into type 1 diabetes, type 2 diabetes, management in people with type 2 diabetes; and Kong et al. [31]
gestational diabetes mellitus, and other specific types of diabetes found that peer support intervention did not significantly improve
[5]. The cases of type 2 diabetes account for 90 %–95 % of the total the distress of people with type 2 diabetes, compared with the
cases [1]. Current evidence indicates that type 2 diabetes imposes a control group. On the contrary, the results of Song et al. [27] and
significant physical, mental, and financial burden on individuals Chen et al. [29] showed that peer support significantly improved
[6,7]. Moreover, the management of type 2 diabetes has become the self-management ability of people with type 2 diabetes.
one of the major challenges for the world's health systems and Furthermore, Kong et al. [30] analyzed the effect of peer support on
economies [8,9]. In order to alleviate the distress for people with self-efficacy of people with type 2 diabetes with intervention time
type 2 diabetes and reduce the financial burden, effective self- (greater than or less than 6 months) as a subgroup. To explore the
efficacy and self-management become critical for people with type specific role of peer support on diabetes distress, Kong et al. [31]
2 diabetes [10–12]. conducted subgroup analyses according to the intervention model
Self-management is a process of active participation in self-care and duration. However, none of the other meta-analyses included a
activities that requires the acquisition of knowledge and the subgroup analysis [27–29]. Many newer randomized controlled
necessary skills to help manage the negative effects of disease trials (RCTs) about the effect of peer support on people with type 2
[4,13]. Self-efficacy refers to one’s beliefs in personal capabilities of diabetes have been published since these previous meta-analyses
performing a specific behavior. It is the most predictive factor in were conducted. A new meta-analysis including these newer
developing and maintaining a new behavior [14]. Effective self- studies may resolve the discrepant results of these five previous
management of diabetes is essential to achieve optimal glycemic meta-analyses. Therefore, in this review and meta-analysis, we
levels and reduce morbidity and mortality [4]. In addition, self- have included all RCTs from the previous studies that met our
efficacy can develop confidence in people's ability to overcome inclusion criteria and have added all available more-recently
barriers and take actions to achieve the goal, improving people’s published studies.
behaviors towards healthy lifestyles [10]. However, most research
has found that a significant proportion of people with diabetes do 2. Methods
not manage their diabetes adequately [13,15], and poor self-
efficacy is also found as an extreme disadvantage in diabetes Our study followed the Preferred Reporting Items for System-
management [16]. atic Reviews and Meta-Analyses (PRISMA) guidelines [32] (See
Currently, there is growing evidence that suggests peer support Supplementary File 1), however, it wasn't registered before the
interventions can be a promising solution for self-efficacy and self- study began.
management in people with type 2 diabetes [17–20]. Peer support
refers to a variety of forms that enable people with similar diseases, 2.1. Search strategy
physical conditions, or experiences to provide substantive help and
emotional support to each other [21]. Several more successful The following electronic databases were utilized for selecting
models of peer support have been accepted, for example, health eligible studies published from inception to Jan. 11, 2020: CNKI,
care provider (HCP)-led peer exchange, peer-led face-to-face (F2F) PubMed, Embase, Cochrane Library, EBSCO(MEDLINE, Psychology
self-management programs, peer coaches, community health & Behavioral Sciences Collection), Nursing Reference Center, Web
workers, support groups, telephone-based support and web-based of Science and PsycINFO. Search terms included the synonyms of
peer support [22]. In addition, the core functions of peer support “Peer support” and “Diabetes Mellitus, Type 200 . Supplementary File
include: (1) assistance in daily management, (2) social and 2 includes the full electronic search strategy for three databases.

761
D. Liang et al. Patient Education and Counseling 104 (2021) 760–769

2.2. Inclusion and exclusion criteria reviewers were divided, differences were resolved by consultation
with a third reviewer. If articles met the selection criteria, they
The inclusion criteria were: (1) types of studies: randomized were selected for full-text review.
controlled trials; (2) study participants: adults who have been Data extraction was performed using Cochrane's guidelines for
diagnosed with type 2 diabetes; (3) types of interventions: the systematic reviews [33]. Two reviewers extracted data separately,
intervention group adopted peer support or the combination of and any disagreement was resolved by consulting with a third
peer support and usual care. Peer supporters were people with reviewer. The following study information was recorded: first
diabetes or people caring for someone with diabetes, but not author, year of publication, country, sample size, recruitment site,
health professionals; (4) types of control: the control group intervention duration, outcome indicators, outcome measures,
included people who accepted usual diabetes care or usual specific treatment for the intervention and control groups, follow-
diabetes education (e.g., the standard diabetes education, diabetes up time and intervention content. The authors were contacted to
knowledge lecture), (5) outcome indicators: the studies assessed obtain missing or unclear data for further analyses.
measurable self-efficacy, self-management, or distress; (6) pub-
lished in either Chinese or English. 2.4. Assessment of risk of bias
The exclusion criteria were: (1) duplicate reports of a study; (2)
studies with insufficient data (e.g., protocols, conference proceed- We used the tool recommended by the Cochrane Handbook
ings or abstracts, and among others) without the author’s Version 5.1.0 [34] to analyze the risk of bias in the trials from the
response; (3) people with severe diabetes-related complications following seven aspects: random sequence generation, allocation
(e.g., ketoacidosis, diabetes-related kidney disease), or people with concealment, blinding of participants and personnel, blinding of
other serious physical diseases (such as stroke, myocardial outcome assessment, incomplete outcome data, selective report-
infarction, and malignant tumors). ing, and other bias. Every item was classified as yes (“low risk of
bias”), no (“high risk of bias”), or unclear (“moderate risk of bias”).
2.3. Study selection and data extraction When the risk of bias of all seven components was defined as “low
risk of bias,” the trial was defined as the overall “low risk of bias.” At
The titles and abstracts of the selected studies were screened the same time, when one or more of the seven bias components
independently by two reviewers. When the opinions of the two were classified as high risk, the trial was graded as “High risk of

Fig. 1. Flow diagram of the literature screening process and results.

762
 D. Liang et al.
Table 1
Characteristics of included studies.

Author (year) Country Sample Mean age Recruitment Intervention Intervention Control Outcome indicators Outcome measures Follow- Intervention content
size(I/C) (y) (I/C) site group duration group up time
Loring et al. (2008) USA † 219/198 52.90/ Community Peer support 6M Usual Self-efficacy, Distress DSES ", The health 12M Diabetes self-management programme
[36] 52.80 care distress scale # (e.g., problem solving, decision making)
Loring et al. (2009) USA † 156/159 67.70/ Community Peer support 6M Usual Self-efficacy DSES " 12M Peer-Led Diabetes Self-management
[38] 65.40 care (highly interactive with emphasis on
action planning and problem solve)
Dale et al. (2009) [37] UK † 90/97 NR Hospital Peer support 6M Usual Self-efficacy, Distress DMSES ", PAID # 6M Motivating adherence and behaviour
care change by a series of telephone support
Smith et al. (2010) Ireland † 192/203 61.10/ NR Peer support 24M Usual Self-efficacy DMSES " NR Nine peer support sessions over two
[40] 63.20 care years, each meeting had a suggested
theme and a small structured
component
Wu SF et al. (2011) China z 72/73 64.80/ NR Peer support + 6M Usual Self-efficacy, Self- C-DMSES ", SDSCA " NR Participants received a booklet entitled
[41] 64.05 Usual care care management “Diabetes Self-Care”, viewed a 10 min
DVD, attended efficacy-enhancing
counseling sessions, and a telephone
follow-up.
van der Wulp et al. Netherlands 59/60 60.00/ NR Peer support 3M Usual Self-efficacy, Distress DMSES ", PAID # NR Discussion topics and exercises on
(2012) [43] † 62.50 care lifestyle changes by home visits and
telephone support
Baghianimoghadam Iran z 40/40 47.70/ DRC Peer support 3M Usual Self-efficacy Self-efficacy scale " 3M Educate the audience through lectures,
et al. (2012) [42] 50.30 education films and group conversations, and
conduct telephone follow-up.
Dang et al. (2013) [44] USA † 47/50 NR Hospital Peer support 6M Usual Self-efficacy, Self- DMSES ", SDSCA " NR Attend at least three of the four sessions
care management (basic knowledge of diabetes and self-
monitoring blood glucose, medications
763

and healthy eating, exercise and

complications, and stress management
and relaxation techniques of how to
seek help from others with roles of
peers and peer leaders)
Han Y et al. (2014) China z 150/150 67.51/ Community Peer support + 6M Usual Self-efficacy, Self- DMSES ", SDSCA " NR Regular diabetes education once a
[47] 63.63 Usual education education management month, discussion and experience
sharing through theme activities
(diabetes diet, exercise, etc.), random
contact (grocery shopping, walking,
etc.)
Chan et al. (2014) [46] China z 312/316 54.50/ DRC Peer support + 12M Usual Self-efficacy, Distress, CDES ", SDSCA ", NR Communication on self-care (e.g., diet

Patient Education and Counseling 104 (2021) 760–769

54.80 Usual care care Self-management DDS-15 # and self-monitoring of blood glucose)
and emotional support by telephone
Liu Y et al. (2015) [49] China z 63/64 62.60/ Hospital Peer support 6M Usual Self-management, DSCS ", DDS-17# NR Self-care skills, emotional support and
64.10 education Distress encouragement provided by group
discussions (sharing experience), and
indefinite media support
Wei H et al. (2017) China z 56/56 57.30/ Hospital Peer support + 2M Usual Self-management SDSCA" NR The activity is mainly in the form of
[51] 55.90 Usual education education peer educators to introduce self-
management experience, diabetes
related knowledge, and group
discussion
Ju C et al. (2018) [52] China z 200/200 67.80/ Community Peer support + 12M Usual Distress DDS-17 # NR Themed and non-themed activities on
68.60 Usual education education diabetes knowledge and skills by
discussion, communication, and
telephone or visit
Gagliardino et al. Argentina z 93/105 NR Primary care Peer support + 12M Usual Distress DDS-17 # NR Diabetes knowledge and skills,
(2013) [45] institution Usual education education psychological and behavioural support
D. Liang et al. Patient Education and Counseling 104 (2021) 760–769

Notes: NR, No report; M, Month; y, years; †, Developed country; z, Developing country; #, Lower value is desirable; ", Higher value is desirable; DSES, Diabetes Self-Efficacy Scale; CDES, the Chinese version of the Diabetes
Empowerment Scale; SDSCA, The Summary of Diabetes Self-care Activities; DSCS, Diabetes Self-care Scale; DMSES, Diabetes Management Self-Efficacy Scale; C-DMSES, the Chinese version of the Diabetes Management Self-Efficacy
Scale; DDS-14, 14-item Diabetes Distress Scale; DDS-15, 15-item Diabetes Distress Scale; DDS-17, 17-item Diabetes Distress Scale; PAID, The Problem Areas In Diabetes questionnaire; DRC, Diabetes Research Center; Follow-up time,
bias.” In other cases, the trial was graded as “Unclear risk.”

The sharing of diabetes knowledge and

educators and group discussion are the

related knowledge and discussing with

by group sessions, visits, and telephone

management issues and three optional

Peer support education in the form of

WeChat groups, introducing diabetes
self-management experience of peer
Peer partners called weekly on self-
Disagreements in bias classification were resolved by discussions

main activities, supplemented by

among the two reviewers and, if necessary, through discussions

1.5-h group sessions to share

with the authors.

games and other activities

2.5. Data synthesis and analysis
Intervention content

The meta-analysis was performed by using RevMan version 5.3

software (Copenhagen: The Nordic Cochrane Center, The Cochrane

each other.
experience

Collaboration, 2014) and Stata version 14.0. We entered group

support

means, standard deviations (SD), and the number of participants in

RevMan 5.3 and conducted a random-effects model meta-analysis,
and Stata14.0 was used for sensitivity analysis and Egger's test.
up time
Follow-

Self-efficacy and self-management were the primary outcomes

3M
NR

NR

and distress was the secondary outcome. All of the outcome

measures in these studies related to self-efficacy and self-
Outcome measures

management were designed so that a higher score is better, while

DSES ", SDSCA "

all of the outcome measures related to distress were designed so

that a lower score is better. If the evaluated trials used different
DDS-14 #

SDSCA "

scales to measure the same outcomes, data were synthesized by

using Hedge's g of standardized mean difference (SMD) with 95 %
confidence interval (CI). We used sensitivity analysis to examine
the stability of the results by removing individual trials to
Outcome indicators

determine whether the removed study had a particular effect.

Self-efficacy, Self-

Self-management

Furthermore, after a sensitivity analysis was performed while

education management

excluding studies with a high risk of bias, we performed funnel

plots and visually examined the signs of asymmetry to investigate
Distress

publication bias, then used Egger’s test [35] as a formal test of

publication bias when the number of the included studies was
more than 10 (n  10).
education
Control
group

Usual

Usual

Usual
care

3. Results
Intervention

3.1. Study selection

The time between the end of the intervention and the outcome measure of follow-up in the included study.
duration

A total of 3072 articles were initially included; and 1086 of them

6M

3M

3M

were duplicates, thus they were deleted. By filtering titles and

abstracts, 1884 articles were excluded. The remaining 102 articles
Usual education
Peer support +

qualified for full-text review, as a result, 85 articles were excluded

Peer support

Peer support
Intervention

according to the selection criteria. Finally, 17 RCTs met our

inclusion criteria. Detailed information can be found in Fig. 1.
group

3.2. Study characteristics

Recruitment

The characteristics of the included studies are shown in Table 1.

Hospital

Hospital

A total of 17 studies with 3884 participants were included in this

site

NR

meta-analysis [36–52]. They were published between 2008 and

2018 either in English (N = 13) [36–46,49,52] or Chinese (N = 4)
Mean age

[47,48,50,51]. These studies were conducted in USA (N = 4)

(y) (I/C)

49.08/
61.80/

49.10/
62.30

50.34

48.50

[36,38,39,44], China (N = 8) [41,46,37–52], UK (N = 1) [37], Iran (N =

1) [42], Ireland (N = 1) [40], Netherlands (N = 1) [43] and Argentina
(N = 1) [45].
size(I/C)

125/119
Sample

30/30

30/30

3.3. Risk of bias

In the analysis of random sequence generation, eleven studies

Country

[36–38,40,42–44,46,49–51] were found to have a “low risk of bias”,

China z

China z
USA †

and these studies were randomized by using random number

tables or computerized randomization modules. Four studies
Si L et al. (2016) [50]

[36,39,44,46] described details of allocation concealment, such as

Heisler et al. (2010)
Table 1 (Continued)

Luo Q et al. (2013)

opaque sealed envelopes. Fifteen studies [38–52] did not report

information on the blinding of participants and personnel, thus
Author (year)

had an unclear risk of bias. Because of the nature of the

intervention behavior changes, none of the selected studies
[48]
[39]

applied a double blinding intervention. Eight studies [37,39–

42,44,46,49] had a “low risk of bias” in the blind method of

764
D. Liang et al. Patient Education and Counseling 104 (2021) 760–769

a “high risk of bias”. The risk of bias evaluations for the included
studies is presented in Fig. 2.

3.4. The results of meta-analysis

3.4.1. Meta-analysis of peer support on self-efficacy

Of the 17 eligible studies, eleven (experimental group: n = 1248,
control group: n = 1207) [36–38,40–44,46–48] included outcome
measures related to peer support on self-efficacy. Meta-analysis
results showed that peer support significantly improved the self-
efficacy of people with type 2 diabetes compared with the control
group [SMD = 0.41, 95 % CI= (0.20, 0.62), p = 0.0001]. Detailed
information can be found in Fig. 3.

3.4.1.1. Subgroup analysis of peer support on self-efficacy. To explore

the specific role of peer support on self-efficacy in people with type
2 diabetes, subgroup analyses were conducted according to
outcome measure, intervention time, country, and age. The
specific analysis was conducted as follows (detailed information
can be found in Table 2):
Outcome measure: Subgroup analysis of the outcome measure
divided results in two categories: DMSES (n = 5) and the others (n =
6). Whether DMSES [SMD = 0.48, 95 % CI= (0.02, 0.95), p = 0.04] or
the others [SMD= 0.35, 95 % CI= (0.17, 0.54), p = 0.0002] was used,
compared with the control group, peer support significantly
improved the self-efficacy of people with type 2 diabetes. The test
for subgroup differences indicated that there was no significant
difference between the two subgroups (p = 0.62).
Intervention time: The intervention time of the peer support
intervention was divided into two groups (<6 months and 6
months) due to the variation across studies (3 months to 24
months). Peer support significantly improved self-efficacy in
people with type 2 diabetes compared with the control group
when the intervention time is <6 months [SMD = 0.27, 95 % CI=
Fig. 2. Risk of bias assessment of included studies. (For interpretation of the
(0.02, 0.51), p = 0.03] and 6 months [SMD = 0.45, 95 % CI= (0.19,
references to colour in this figure legend, the reader is referred to the web version of 0.71), p = 0.0006]. The subgroup difference test indicated no
this article.) significant difference between the two subgroups (p = 0.31).
Notes: Red, high risk of bias; Yellow, moderate risk of bias; Green, low risk of bias Country: Country was divided into two groups for subgroup
analysis: developing countries (n = 5) and developed countries (n =
outcome assessment. Sixteen studies [36–41,43,39–52] found a 6). Compared with the control group, peer support can signifi-
“low risk of bias” in incomplete outcome data, as some of them cantly improve the self-efficacy of people with type 2 diabetes in
used flowcharts to report the number of randomly assigned both developing countries [SMD = 0.44, 95 % CI= (0.14, 0.74), p =
participants, intended treatment, loss and exclusion after ran- 0.005] and developed countries [SMD= 0.39, 95 % CI= (0.07, 0.72), p
domization, and reasons for losses in the follow-up intervention = 0.02]. The test for subgroup differences indicated that there was
period. Seventeen studies [36,39–52] reported all primary or no significant difference between the two subgroups (p = 0.85).
secondary outcome measures, including assessment tools for each Age: Eleven RCTs were used to calculate the effect of peer
outcome measure, and how and when participants were assessed. support on self-efficacy in people of different ages. However, two
Overall, fifteen studies [38–52] were classified as having a studies had incomplete data, so nine studies were finally analyzed.
“moderate risk of bias” and the remaining two [36,37] as having Age was divided into two groups (< 60 years old and  60 years old)

Fig. 3. Forest plot of the effect of peer support on self-efficacy compared with the control group in people with type 2 diabetes.
Notes: IV, Inverse variance.

765
D. Liang et al. Patient Education and Counseling 104 (2021) 760–769

Table 2
Effect of peer support on self-efficacy in people with type 2 diabetes (subgroups).

Group Literature number Heterogeneity test SMD (95 %CI) P Subgroup effect (p-value)
2
P I
outcome measure DMSES 5 <0.001 91.00 0.48(0.02,0.95) 0.04 0.62
the others 6 0.02 62.00 0.35(0.17,0.54) 0.0002
intervention time <6 months 3 0.48 0.00 0.27(0.02,0.51) 0.03 0.31
6 months 8 <0.001 88.00 0.45(0.19,0.71) 0.0006
country developing countries 5 0.0004 81.00 0.44(0.14,0.74) 0.005 0.85
developed countries 6 <0.001 87.00 0.39(0.07,0.72) 0.02
age <60 years old 4 0.14 45.00 0.26(0.07,0.45) 0.007 0.57
60 years old 5 0.0001 83.00 0.36(0.07,0.65) 0.01

Fig. 4. Forest plot of the effect of peer support on self-management in people with type 2 diabetes compared with the control group.
Notes: IV, Inverse variance.

Table 3
Effect of peer support on self-management in people with type 2 diabetes (subgroups).

Group Literature number Heterogeneity test SMD (95 %CI) P Subgroup effect (p-value)
2
P I
intervention time <6 months 3 <0.0001 93.00 1.74(0.51,2.97) 0.006 0.24
6 months 5 <0.0001 96.00 0.91(0.24,1.58) 0.008
age <60 years old 4 <0.0001 98.00 1.29(-0.06,2.63) 0.06 0.56
60 years old 3 0.005 81.00 0.86(0.45,1.28) <0.0001

according to age differences between studies. No matter if the age intervention time < 6 months [SMD = 1.74, 95 % CI= (0.51, 2.97), p =
was < 60 years old [SMD = 0.26, 95 % CI= (0.07, 0.45), p = 0.007] or  0.006] and studies with intervention time 6 months [SMD = 0.91,
60 years old [SMD= 0.36, 95 % CI= (0.07, 0.65), p = 0.01], peer 95 % CI= (0.24, 1.58), p = 0.008] showed peer support significantly
support significantly improved self-efficacy in people with type 2 improved self-management in people with type 2 diabetes
diabetes compared with the control group. The test for subgroup compared with the control group. The subgroup difference test
differences indicated no significant difference between the two indicated that there was no significant difference between the two
subgroups (p = 0.57). subgroups (p = 0.24).
Age: Eight RCTs were utilized to calculate the effects of peer
3.4.2. Meta-analysis of peer support on self-management support on self-management in people of different ages, however,
Of the 17 eligible studies, eight (experimental group: n = 739, one study had incomplete data, so seven studies were finally
control group: n = 737) [41,44,46–51] included outcome measures analyzed. The age subgroup differentiated between < 60 years old
related to peer support on self-management. Meta-analysis (n = 4) and  60 years old (n = 3). When the age was 60 years old,
showed that peer support significantly improved self-manage- peer support significantly improved self-management compared
ment of people with type 2 diabetes compared with the control with the control group [SMD = 0.86, 95 % CI= (0.45, 1.28), p<
group [SMD = 1.21, 95 % CI = (0.58, 1.84), p = 0.0002]. Detailed 0.0001], but when the age was < 60 years old, there was no
information can be found in Fig. 4. statistically significant difference [SMD = 1.29, 95 % CI= (-0.06,
2.63), p = 0.06]. The test for subgroup differences indicated that
3.4.2.1. Subgroup analysis of peer support on self-management. To there was no significant difference between the two subgroups (p =
explore the specific role of peer support on self-management in 0.56).
people with type 2 diabetes, we conducted subgroup analyses
based on intervention time and age. The specific analysis was 3.4.3. Meta-analysis of peer support on distress
conducted as follows (detailed information can be found in Of the 17 eligible studies, eight (experimental group: n = 1030,
Table 3): control group: n = 1051) [36,37,39,43,45,46,49,52] included
Intervention time: The peer support intervention time was outcome measures related to peer support on distress. There
divided into two groups: < 6 months and 6 months. Studies with was no statistically significant difference in the improvement in

766
D. Liang et al. Patient Education and Counseling 104 (2021) 760–769

distress in people with type 2 diabetes compared to the control

group [SMD= -0.08, 95 % CI= (-0.25, 0.08), p = 0.34]. Detailed
information can be found in Fig. 5.

3.5. Sensitivity analysis and publication bias assessment

We assessed the publication bias of self-efficacy by visual

examination of funnel plots and quantitative analysis by Egger's
test. The funnel plot of proximal SE between the peer support
intervention group and the control group was slightly asymmetric;
whereas the Egger’s test revealed no significant publishing bias (t =
1.74, p = 0.116). The funnel plot of self-efficacy is presented in Fig. 6.
Sensitivity analysis was performed to test the reliability of the
meta-analysis results by eliminating studies one by one, the result
demonstrated that there was no significant difference between the
two groups. Thus, the findings that peer support significantly
improved self-efficacy in people with type 2 diabetes compared to
the control group would be considered relatively reliable. Fig. 6. Funnel plot of self-efficacy.

4. Discussion and conclusion ages. It also significantly improved the self-management of people
in different intervention times and at the age of 60 or older.
4.1. Discussion However, more studies are needed to evaluate the effect of peer
support on self-management of people with type 2 diabetes under
4.1.1. Summary of main findings 60 years old.
The purpose of our meta-analysis was to assess the effect of
peer support on self-efficacy and self-management in people with 4.1.2. Agreements and disagreements with other meta-analyses
type 2 diabetes. The meta-analysis included 17 RCTs, of which 15 There were five previous meta-analyses [27–31] similar to our
studies [38–52] were classified as “moderate risk of bias” and the study. Our study included all seven of the RCTs included in Kong
remaining two studies [36,37] as “high risk of bias”. The results et al.'s [30] investigation of the effect of peer support on self-
showed that peer support significantly improved self-efficacy and efficacy of people with type 2 diabetes. In addition, our study
self-management in people with type 2 diabetes. The reason may includes two RCTs missed by Kong et al. [41,44] and two studies
be the majority of the peer supporters were professionally trained published afterwards [47,48]. The standardized mean difference of
people who share similar diabetes care experience, which could these four additional studies in favor of peer support ranges from
help people with type 2 diabetes better understand the focus of 0.30 to 1.65. Consequently, our results showed a significant
diabetes care, thus increasing their confidence of self-management positive effect for peer support whereas Kong et al. did not. Zhao
and self-efficacy [53]. However, there was no clear evidence et al. [28] used two RCTs to evaluate the effect of peer support on
supported that peer support was effective in the distress of people self-efficacy in people with type 2 diabetes, we included nine
with type 2 diabetes. This may be because most current peer [36,38,41–44,46–48] additional RCTs in our study than Zhao et al. .
support interventions focus on educational and behavioral The standardized mean differences in favor of peer support ranged
changes, such as medication, diet and exercise, rather than directly from 0.09 to 1.65; three RCTs used descriptive analysis to evaluate
reducing the distress of people with diabetes, [31]. The other the effect of peer support on self-management because they were
explanation could be the mix of several types of interventions (e.g., unable to extract data from the different assessment tools. Overall,
telephone support, support groups and peer-led face-to-face (F2F) more RCTs were used in our study, and the results of the positive
self-management programs), and which were mostly planned. role of peer support obtained through data extraction and data
Therefore, future studies can use separate types of interventions to synthesis may be more credible than the results of ineffective peer
evaluate the effect of peer support on the distress of people with support by Zhao et al.'s study. The results of Song et al. [27] and
type 2 diabetes [54]. Chen et al. [29] were consistent with our study that peer support
The results of the subgroup analysis showed that peer support could significantly improve self-management in people with type
significantly improved self-efficacy in people with type 2 diabetes 2 diabetes, but they had some flaws as follows. Song et al. [27] only
in different countries, different intervention times, and at different included two RCTs, and the data collection was not comprehensive

Fig. 5. Forest plot of the effect of peer support on distress compared with the control group in people with type 2 diabetes.
Notes: IV, Inverse variance.

767
D. Liang et al. Patient Education and Counseling 104 (2021) 760–769

enough. Although Chen et al. [29] included 11 studies to evaluate management in people with type 2 diabetes, providing evidence
the effect of peer support, the included studies did not meet the that support the clinical application of peer support in people with
inclusion criteria, such as non randomized controlled trials, and type 2 diabetes. However, the results of this study showed no clear
participants had diabetes related complications or other serious evidence that peer support improves the distress of people with
diseases. In addition, all their included studies were conducted in type 2 diabetes. With the increase of mental health problems in
China, which may lead to regional bias. Finally, Kong et al. [31] people with type 2 diabetes today, more updated relevant studies
evaluated the effect of peer support on people with type 2 diabetes need to be included to assess the impact of peer support.
only by taking distress as the outcome indicator, while our study
investigated the effect of peer support from the three aspects of 4.3. Practice implications
self-efficacy, self-management and distress, which was more
comprehensive. This meta-analysis suggests that peer support should be
considered as a complementary treatment for people with type
4.1.3. Strengths and limitations 2 diabetes. Medical staff can recommend peer support as part of
The strengths of this study are as follows. First, in previous the daily care of patients with type 2 diabetes to improve their self-
studies on the effect of peer support on people with type 2 efficacy and self-management.
diabetes, self-management was only analyzed as a secondary
outcome indicator. In this study, self-management was taken as Authorship
the primary outcome indicator for the first time, and it was further
studied from the intervention time and age. Second, as for self- We confirm that all listed authors meet the authorship criteria,
efficacy in people with type 2 diabetes, in the past studies, only and all authors are in agreement with the content of the
Kong et al. [30] conducted a subgroup analysis based on the manuscript. CRC and XZ designed this research; DDL contributed
intervention time when evaluating the effect of peer support on to the later stages of the design. DDL, ZW, JYG identified and
self-efficacy of people with type 2 diabetes, while other relevant screened the included randomized controlled trials. DDL, ZW, RYJ,
meta-analyses did not carry out a subgroup analysis [27–29], thus HTH analyzed and evaluated the data. All authors give suggestions
could not determine the impact of peer support on people with to the data analysis and helped to interpret the results. DDL
type 2 diabetes under different specific circumstances. Our study completed the writing of this paper, CRC, GLL, RYJ, JFY, ZW, ZHW
assessed the effect of peer support on self-efficacy under specific and XZ revised the manuscript. All authors read and approved the
conditions (e.g., outcome measure, intervention time, country, final manuscript.
age). Finally, we utilized more databases and free words to expand
the search range and then acquired more relevant literature to Funding
improve the reliability and the internal validity of the study.
The current study has limitations. First, the attributes of the The study was supported by Chinese Henan Provincial
heterogeneity among included studies (e.g. population character- Education Department Project (18B310001), Chinese Henan
istics, diagnostic results, sample size, frequency and intensity of province Social Science Planning and Decision-making Consulting
intervention, measurement tools, follow-up, recruitment site, etc.,) Project (2018BJC38), Subsidy for the Project of Innovation and
were not fully explored because of the limited information access. Quality Improvement of Postgraduate Education in Henan
Second, some subgroups in our subgroup analyses have small University (SYL19060141) and Henan Province Teacher Education
number of studies, some have high heterogeneity, and some have Curriculum and Reform Project (2016-JSJYZD-003).
both. Third, because the authors were only fluent in English and
Chinese, they were only able to retrieve published research in either Declaration of Competing Interest
English or Chinese, but not for research in other languages in the field.
Because there are no journals for manual retrieval, literature retrieval The authors report no declarations of interest.
is also limited which may lead to research selection bias.
Acknowledgments
4.1.4. Implications for future research
There were no precise criteria to examine how peer support The study was supported by Chinese Henan Provincial
interventions could be more effective, which may be influenced by Education Department Project (18B310001), Chinese Henan
different countries and health care systems, as well as many province Social Science Planning and Decision-making Consulting
situational factors such as the psychological, cultural, and social Project (2018BJC38), Subsidy for the Project of Innovation and
environment. Therefore, in future research, a rational intervention Quality Improvement of Postgraduate Education in Henan
peer support program should flexibly adapt to various environ- University (SYL19060141) and Henan Province Teacher Education
mental, demographic, social, cultural, organizational, and eco- Curriculum and Reform Project (2016-JSJYZD-003). The authors
nomic conditions. In addition, researchers should pay more gratefully acknowledge the experts and members of our group for
attention to the quality of the selected research, such as the their help and advice.
generation of explicit random sequences and the concealment of
allocation. Finally, the follow-up analysis was not conducted in this Appendix A. Supplementary data
meta-analysis due to the lack of relevant data, thus the duration of
the effect of peer support on people with type 2 diabetes could not Supplementary material related to this article can be found, in
be determined. With the increase of related research in the future, the online version, at doi:https://doi.org/10.1016/j.pec.2020.11.011.
the follow-up effect of peer support for people with type 2 diabetes
could be further investigated. References

4.2. Conclusion [1] Diagnosis and classification of diabetes mellitus, Diabetes Care 37
(Supplement 1) (2014) S81–S90.
[2] S. Wild, G. Roglic, A. Green, R. Sicree, H. King, Global prevalence of diabetes:
As a promising approach, peer support plays a significant estimates for the year 2000 and projections for 2030, Diabetes Care 27 (2004)
positive role in the improvement of self-efficacy and self- 1047–1053.

768
D. Liang et al. Patient Education and Counseling 104 (2021) 760–769

[3] International Diabetes Federation, IDF Diabetes Atlas, 9th ed., International [31] L.N. Kong, P. Hu, Q.H. Zhao, H.Y. Yao, S.Z. Chen, Effect of peer support
Diabetes Federation, Brussels, Belgium, 2019. intervention on diabetes distress in people with type 2 diabetes: a systematic
[4] J. Reyes, T. Tripp-Reimer, E. Parker, B. Muller, H. Laroche, Factors influencing review and meta-analysis, Int. J. Nurs. Pract. (2020)e12830.
diabetes self-management among medically underserved patients with type II [32] A. Liberati, D.G. Altman, J. Tetzlaff, C. Mulrow, P.C. Gotzsche, J.P. Ioannidis, et al.,
diabetes, Glob. Qual. Nurs. Res. 4 (2017) 2333393617713097. The PRISMA statement for reporting systematic reviews and meta-analyses of
[5] Classification and diagnosis of diabetes: standards of medical care in diabetes- studies that evaluate health care interventions: explanation and elaboration, J.
2020, Diabetes Care 43 (2020) S14–s31. Clin. Epidemiol. 62 (2009) e1–34.
[6] H. King, R.E. Aubert, W.H. Herman, Global burden of diabetes, 1995–2025: [33] J.P. Higgins, D.G. Altman, P.C. Gotzsche, P. Juni, D. Moher, A.D. Oxman, et al., The
prevalence, numerical estimates, and projections, Diabetes Care 21 (1998) Cochrane Collaboration’s tool for assessing risk of bias in randomised trials, Br.
1414–1431. Med. J. 343 (2011) d5928.
[7] M.M. Engelgau, L.S. Geiss, J.B. Saaddine, J.P. Boyle, S.M. Benjamin, E.W. Gregg, [34] J.J. Deeks, J. Higgins, D.G. Altman, S. Green, Cochrane Handbook for Systematic
et al., The evolving diabetes burden in the United States, Ann. Intern. Med. 140 Reviews of Interventions Version 5.1. 0 (updated March 2011), The Cochrane
(2004) 945–950. Collaboration, 2011.
[8] F. Boudreau, M. Moreau, J. Côté, Effectiveness of computer tailoring versus peer [35] M. Egger, G. Davey Smith, M. Schneider, C. Minder, Bias in meta-analysis
support web-based interventions in promoting physical activity among detected by a simple, graphical test, Br. Med. J. 315 (1997) 629–634.
insufficiently active Canadian adults with type 2 diabetes: protocol for a [36] K. Lorig, P.L. Ritter, F. Villa, J.D. Piette, Spanish diabetes self-management with
randomized controlled trial, JMIR Res. Protoc. 5 (2016) e20-e20. and without automated telephone reinforcement: two randomized trials,
[9] S. Besancon, C. Fleury, Peer network for the therapeutic education of patients Diabetes Care 31 (2008) 408–414.
with type 2 diabetes in Mali, Soins 63 (2018) 55–58. [37] J. Dale, I. Caramlau, J. Sturt, T. Friede, R. Walker, Telephone peer-delivered
[10] S.K. Ahmad Sharoni, M.N. Mohd Razi, N.F. Abdul Rashid, Y.E. Mahmood, Self- intervention for diabetes motivation and support: the telecare exploratory
efficacy of foot care behaviour of elderly patients with diabetes, Malays. Fam. RCT, Patient Educ. Couns. 75 (2009) 91–98.
Phys. 12 (2017) 2–8. [38] K. Lorig, P.L. Ritter, F.J. Villa, J. Armas, Community-based peer-led diabetes self-
[11] B.M. Kwan, B. Jortberg, M.K. Warman, I. Kane, R. Wearner, R. Koren, et al., management: a randomized trial, Diabetes Educ. 35 (2009) 641–651.
Stakeholder engagement in diabetes self-management: patient preference for [39] M. Heisler, S. Vijan, F. Makki, J.D. Piette, Diabetes control with reciprocal peer
peer support and other insights, Fam. Pract. 34 (2017) 358–363. support versus nurse care management a randomized trial, Ann. Intern. Med.
[12] M.L. Litchman, H.R. Walker, A.H. Ng, S.E. Wawrzynski, S.M. Oser, D.A. 153 (2010) 507–515.
Greenwood, et al., State of the science: a scoping review and gap analysis [40] S.M. Smith, G. Paul, A. Kelly, D.L. Whitford, E. O’Shea, T. O’Dowd, Peer support
of diabetes online communities, J. Diabetes Sci. Technol. 13 (2019) 466–492. for patients with type 2 diabetes: cluster randomised controlled trial, Br. Med.
[13] R. Carpenter, T. DiChiacchio, K. Barker, Interventions for self-management of type J. 342 (2011) d715.
2 diabetes: an integrative review, Int. J. Nurs. Midwifery Sci. 6 (2019) 70–91. [41] S.F. Wu, M.C. Lee, S.Y. Liang, Y.Y. Lu, T.J. Wang, H.H. Tung, Effectiveness of a self-
[14] A. Bandura, Self-efficacy: toward a unifying theory of behavioral change, efficacy program for persons with diabetes: a randomized controlled trial,
Psychol. Rev. 84 (1977) 191–215. Nurs. Health Sci. 13 (2011) 335–343.
[15] A.D. Kurnia, A. Amatayakul, S. Karuncharernpanit, Predictors of diabetes self- [42] M.H. Baghianimoghadam, M. Hadavandkhani, M. Mohammadi, H. Fallahzade,
management among type 2 diabetics in Indonesia: application theory of the B. Baghianimoghadam, Current education versus peer-education on walking in
health promotion model, Int. J. Nurs. Sci. 4 (2017) 260–265. type 2 diabetic patients based on Health Belief Model: a randomized control
[16] R.E. Glasgow, D.J. Toobert, C.D. Gillette, Psychosocial barriers to diabetes self- trial study, Rom. J. Intern. Med. 50 (2012) 165–172.
management and quality of life, Diabetes Spectr. 14 (2001) 33–41. [43] I. van der Wulp, J.R. de Leeuw, K.J. Gorter, G.E. Rutten, Effectiveness of peer-led
[17] L. Qi, Q. Liu, X. Qi, N. Wu, W. Tang, H. Xiong, Effectiveness of peer support for self-management coaching for patients recently diagnosed with Type 2
improving glycaemic control in patients with type 2 diabetes: a meta-analysis diabetes mellitus in primary care: a randomized controlled trial, Diabet. Med.
of randomized controlled trials, BMC Publ. Health 15 (2015) 471. 29 (2012) e390–e397.
[18] C.J. Chiou, H.L. Huang, Y.C. Lin, R.H. Wang, Developing and psychometrically [44] T.T.N. Dang, W. Deoisres, P. Keeratiyutawong, L. Baumann, Effectiveness of a
testing the functions of diabetes peer support group scale, J. Nurs. Res. 24 diabetes self-management support intervention in Vietnamese adults with
(2016) 357–364. type 2 diabetes, J. Sci. Technol. Hum. 11 (2013) 13–23.
[19] H. Warshaw, L. Hodgson, M. Heyman, T.K. Oser, H.R. Walker, P. Deroze, et al., [45] J. Jose Gagliardino, V. Arrechea, D. Assad, G.G. Gagliardino, L. Gonzalez, S.
The role and value of ongoing and peer support in diabetes care and education, Lucero, et al., Type 2 diabetes patients educated by other patients perform at
Diabetes Educ. 45 (2019) 569–579. least as well as patients trained by professionals, Diabetes Metab. Res. Rev. 29
[20] H. Warshaw, D. Edelman, Building bridges through collaboration and (2013) 152–160.
consensus: expanding awareness and use of peer support and peer support [46] J.C. Chan, Y. Sui, B. Oldenburg, Y. Zhang, H.H. Chung, W. Goggins, et al., Effects of
communities among people with diabetes, caregivers, and health care telephone-based peer support in patients with type 2 diabetes mellitus
providers, J. Diabetes Sci. Technol. 13 (2019) 206–212. receiving integrated care: a randomized clinical trial, J. Am. Med. Assoc. Intern.
[21] C.L. Dennis, Peer support within a health care context: a concept analysis, Int. J. Med. 174 (2014) 972–981.
Nurs. Stud. 40 (2003) 321–332. [47] Y. Han, J. Shi, Y. Liu, S. Li, W. Yang, Y. Wang, et al., Influence of Peer education on
[22] M. Heisler, Overview of peer support models to improve diabetes self- quality of life of type 2 diabetes patients in Dagang Oilfield, Med. Soc. 27 (2014)
management and clinical outcomes, Diabetes Spectr. 20 (2007) 214–221. 37–39.
[23] E.B. Fisher, R.I. Boothroyd, M.M. Coufal, L.C. Baumann, J.C. Mbanya, M.J. [48] Q. Luo, G. Xing, C. Yang, Effect of peer health education on anxiety and
Rotheram-Borus, et al., Peer support for self-management of diabetes depression in patients with type 2 diabetes mellitus, J. Psychiatry 27 (2014)
improved outcomes in international settings, Health Aff. (Millwood) 31 (2012) 53–55.
130–139. [49] Y. Liu, Y. Han, J. Shi, R. Li, S. Li, N. Jin, et al., Effect of peer education on self-
[24] J.F. Caro, E.B. Fisher, A solution might be within people with diabetes management and psychological status in type 2 diabetes patients with
themselves, Fam. Pract. 27 (2010) I1–I2. emotional disorders, J. Diabetes Invest. 6 (2015) 479–486.
[25] T. Seuring, Marthoenis, S. Rhode, L. Rogge, S. Vollmer, Using peer education to [50] L. Si, P. Duan, J. Liu, Effects of WeChat group-style peer support education on
improve diabetes management and outcomes in a low-income setting: a blood glucose and self-management in diabetic patients, Nurs. Res. 30 (2016)
randomized controlled trial, Trials 20 (2019) 548. 3382–3384.
[26] D.A. Greenwood, M.L. Litchman, A.H. Ng, P.M. Gee, H.M. Young, M. Ferrer, et al., [51] H. Wei, Y. Han, Study on the influence of peer education on the biochemical
Development of the intercultural diabetes online community research index, self-management ability and psychological status of type 2 diabetes
council: codesign and social media processes, J. Diabetes Sci. Technol. 13 patients, China Med. Herald 14 (2017) 62–65.
(2019) 176–186. [52] C. Ju, R. Shi, L. Yao, X. Ye, M. Jia, J. Han, et al., Effect of peer support on diabetes
[27] D. Song, S. Xu, Q. Sun, A meta-analysis of the impact of peer education on self- distress: a cluster randomized controlled trial, Diabet. Med. 35 (2018) 770–
management in community type 2 diabetes patients, Plast. Surg. Nurs. 32 775.
(2015) 1–5. [53] W. Yan, X. Wu, J. Li, Effects of peer support education on self-efficacy
[28] Z. Zhao, W. Xie, Q. Yang, Effectiveness of peer support for improving self- management and glycemic control in diabetic patients after discharge, Pract. J.
management in patients with type 2 diabetes: a meta-analysis of randomized Clin. Med. 11 (2014) 205–206.
controlled trials, Chin. J. Nurs. 51 (2016) 356–364. [54] M.L. Litchman, T.K. Oser, L. Hodgson, M. Heyman, H.R. Walker, P. Deroze, et al.,
[29] X. Chen, Y. Liu, P. Liu, M. Duan, X. Li, A meta-analysis of the effect of peer In-person and technology-mediated peer support in diabetes care: a
education on hba1c and self-management in type 2 diabetes, Mod. Prev. Med. systematic review of reviews and gap analysis, Diabetes Educ. 46 (2020) 230–
45 (2018) 964-968+984. 241.
[30] L.-N. Kong, P. Hu, L. Yang, D. Cui, The effectiveness of peer support on self-
efficacy and quality of life in adults with type 2 diabetes: a systematic review
and meta-analysis, J. Adv. Nurs. 75 (2019) 711–722.

769