Key findings from the review are discussed in this section and are summarized in Table 2 titled, Characteristics and Key Outcomes of Reviewed Studies.

Table 2 Characteristics and key outcomes of reviewed studiesIn-Person Peer Support

There is significant evidence for the effectiveness of in-person (IP) support in diabetes education programs [17•, 18, 22]. IP support allows direct verbal and non-verbal communication, enables social interactions and facilitates the exchange of peer-to-peer information, support, and goal setting.

Shared Medical Appointments

Shared medical appointments (SMAs) are typically IP doctor-patient visits in which groups of patients are seen by one or more healthcare providers. Patients interact with both providers and with peers during SMAs [23]. Heisler and colleagues compared the effectiveness of diabetes SMAs to usual care in improving HbA1c [24•]. Patients with T2DM and elevated HbA1c who had been prescribed medication were recruited from 5 U.S. Veterans Affairs (VA) health systems. Participants engaged in IP SMAs of 60–120 min totaling 6–8 h over 12 months where they received education, engaged in goal setting, and shared their experiences with peers. A SMA plus reciprocal peer support intervention arm was available in which participants had the option to be connected to another participant for mutual support, but uptake was low resulting in data being combined with the SMA-only group data. The control group received usual care.

At 6 months, the intervention group’s mean HbA1c was reduced by 0.35% compared to usual care (p = 0.001) and was lower than usual care by 0.53% in participants attending at least half of the sessions (p = 0.001), however, these differences were not significant at 12 months. Insulin starts were significantly greater in the intervention group at both 6 and 12 months. Significant mean differences were not observed for systolic blood pressure, statin starts, or HTN med class changes.

Health Clinic Staff-facilitated Support Groups

Mexico is among countries with the highest prevalence of diabetes, with more than 24% of adults over age 50 being diagnosed with T2DM [25]. The Mexican Ministry of Health’s primary care services include diabetes support groups known as Grupos de Ayuda Mutua (GAM), which provide a supportive group environment, glucose self-monitoring and health information via monthly meetings. Rosales and colleagues sought to complement the GAM program with Meta Salud Diabetes (MSD), a secondary prevention curriculum designed to operate within the framework of the GAM and evaluated its effects on Framingham CVD risk in a cluster-randomized trial [26]. The objective was for the MSD to improve the existing GAM by providing educational information, group discussions, and interactive workshops to promote a healthier diet, increased physical activity and reduced disease complications.

The GAM + MSD intervention consisted of 2-h IP participatory educational sessions and group-based interactive activities facilitated by trained health center staff that engaged GAM support groups at 22 government health clinics over 13 consecutive weeks. The following items were on the agenda at each session: blood pressure and glucose monitoring; readings, discussions and games related to the day’s topic of focus; implementation of a physical activity, and a follow-up goal setting exercise. The usual care group participated in standard GAM support groups.

Mean Framingham CVD risk score was significantly reduced by 3.17% points in the MSD arm versus GAM usual care at 3 months (p = 0.013), however, the between groups difference was not significant at 12 months. Greater reduction in mean Framingham CVD risk was observed in men than women and in those whose HbA1c was < 8% at baseline. No significant effects were observed for HbA1c, blood pressure, or blood lipids. Compared to baseline, diabetes distress in the GAM + MSD group was significantly lower at the 3- and 12-month follow-up assessments.

IP Diet & Lifestyle Intervention with Telephone Support

Sampson and colleagues implemented an IP diet and lifestyle intervention with and without additional telephone support provided by trained peers in the UK for individuals with newly diagnosed “screen detected” type 2 diabetes [27]. Participants were randomized into three arms. Those in the intervention arm (INT) received six 2-h group educational sessions over 12 weeks and up to fifteen 2.5-h group maintenance sessions 8 weeks apart; maximum possible contact time was 49.5 h over 30 months. Individuals randomized to the second arm received the INT intervention plus monthly telephone support from peer volunteers who themselves had T2DM and who had received training as diabetes peer mentors (INT-DPM). Telephone contacts were monthly for 3 months and then every 2 months thereafter for up to 46 months. Those in the control arm (CON) received no intervention.

At 12 months, mean HbA1c significantly decreased in the telephone peer support INT-DPM group compared to the control group (CON) (-0.3%; p = 0.007), however, differences were not significant between the INT-DPM and INT groups, or between the INT and control groups. Days of resistance exercise per week significantly increased compared to the control group in both the INT and INT-DPM groups (INT: 4.22 days, p = 0.01; INT-DPM: 3.32 days, p = 0.02). Glycemic control was significantly lower in INT-DPM group vs. the CON group for participants < 65 years of age compared to those > 65 years (INT-DPM -2.7% vs. CON -2.2%; interaction p = 0.007). No significant differences were observed between any of the groups for body weight, body fat %, or waist circumference.

Family- and Friend-focused Peer Intervention

Positive support by family and friends of individuals with T2DM is strongly associated with more consistent self-care and lower glycemic control [28]. However, it is possible for both helpful and harmful interactions to co-occur within families and friendships. As cultural peers, family members and friends can support or undermine an individual’s diabetes self-care behaviors, which affects the likelihood of them continuing to implement the behaviors [29].

Nelson and colleagues implemented an intervention entitled, Family/friend Activation to Motivate Self-care (FAMS) to assist patients with T2DM in pursuing self-care goals and in effectively responding to both helpful and harmful input from family and friends [30]. In this RCT, FAMS provided monthly health coaching phone calls by behavioral health providers that were subsequently reinforced by text messaging over the 9-month intervention period. Each FAMS participant had the option to designate a cultural peer support person (family member or friend) and invite them to receive the text messages sent by their health coach, potentially prompting peer interactions about the participant’s health behaviors and diabetes self-management goals. Monthly coaching calls to FAMS participants focused on building the following skills: improving the ability to recognize and manage both helpful and harmful messages from friends or family members, SMART goal setting, assertive communication, obtaining social support, collaborative problem-solving, and cognitive behavioral coping skills. Participants in the control group received usual care. Participants and support persons in both study arms received print materials on managing diabetes and providing support to individuals with diabetes, and participants received text messages advising how to access their HbA1c results.

At 6 months, mean HbA1c decreased by 0.25% compared to controls but did not reach statistical significance (p = 0.09). Effects on HbA1c vs. controls were also non-significant at 9 months (-0.05%; p = 0.9) and at 12 months (-0.02%; p = 0.9). Interestingly, participants who were not cohabitating with their support person had significantly lower HbA1c at 6 months (-0.64%; p = 0.033) but not at other time points. The investigators suggested that out-of-home support is likely to strengthen one’s support network, especially for those living alone, and may result in fewer instances of harmful social interactions than can occur with a cohabitating support person. At both 6 and 9 months, significant improvements were observed in intermediate targets of the FAMS intervention compared to controls including diabetes self-efficacy, problem eating behaviors, summative physical activity, helpful family/friend involvement, and autonomy support, although the study did not attempt to measure the relative contributions made by health coaches and cultural peer support persons to these outcomes. Another recent trial of a family and friend cultural peer support intervention for patients with diabetes yielded similar findings, with improvements in intermediate targets but non-significant effects on HbA1c [31].

Peer Coaches and Peer MentorsPeer Leader Supported Self-Care Intervention

Although large numbers of patients with T2DM receive diabetes care in primary care settings, patients at the highest risk are often managed in subspecialty settings. Tang and colleagues conducted a peer-led intervention for adult T2DM patients presenting to endocrinology and diabetes education centers in British Columbia, Canada [32] and compared it to a usual care control group. Peer leaders completed 30 h of knowledge and skills training in T2DM self-management, providing emotional support, and linking patients to clinical resources. It is important to note that the peer leaders themselves had T2DM and a mean self-reported HbA1c ≤ 8.0% [33]. Individuals in the usual care control group received all services normally available to them at the diabetes education centers, including access to a nurse, dietician and social worker.

Intervention group participants received 30 h of T2DM self-management education and were subsequently matched with a peer leader based on gender and geographic proximity. After an initial IP educational meeting between peer leader and participant, peer leaders made 12 weekly telephone contacts with their partners over the first 3 months of the 48-month intervention to discuss self-management challenges, solve problems, and set self-management goals. During the final 36 weeks, peer leaders made 18 biweekly telephone support calls to intervention group participants.

At 3 months and 12 months post-intervention, HbA1c was not significantly different either from baseline within the intervention group or between the intervention and control groups. Significant between group improvements in systolic blood pressure were observed in the intervention group compared to controls at 12 months (-4.1 points; p = 0.023). Overall distress and emotional distress scores in the intervention group were significantly lower at 3 months and 12 months than at baseline but neither were significantly different than controls at any time point.

Community Health Workers

Community health workers (CHWs) are cultural peers in that they typically share cultural, racial, ethnic, community, and socioeconomic backgrounds with the individuals they serve and they may have direct experience with target health conditions as a patient or in support of a family member [34]. With education and training in diabetes self-management and health behavior change skills, CHWs have the potential to positively affect diabetes outcomes [35].

Home-based Diabetes Education and Counseling by Trained Volunteers

In Nepal, the healthcare system has traditionally focused on communicable diseases rather than chronic conditions and has struggled to adapt to rapidly increasing rates of T2DM [36]. A national initiative to address this need involves the mobilization of more than 50,000 Female Community Health Volunteers (FCHVs) in communities across the country. In this study, these locally-based cultural peers delivered a 12-month IP community-based diabetes intervention to adults in semi-urban settings in Western Nepal with the goal of reducing mean blood glucose [37]. This cluster randomized trial enrolled community residents with a current diagnosis of T2DM or mean blood glucose levels of 126 mg/dL or higher. FCHVs provided home-based diabetes education and counseling, measured blood glucose, blood pressure, height, and weight, and referred patients for clinic visits as needed. Home visits by FCHVs took place once every 4 months (3 visits) for one year. Participants randomized to the control group received usual care in which they managed their T2DM as usual and were placed on a wait list with the option to receive the intervention after 12 months.

Mean fasting blood glucose in the intervention group significantly decreased by 2.7% in the FCHV intervention group while it increased by 2.2% in the usual care group (p < 0.001). Systolic blood pressure significantly decreased (-7.86 points; p = 0.002) in the intervention group vs. controls. Significant improvements were not observed in diastolic blood pressure, physical activity, BMI, alcohol consumption, smoking, fruit and vegetable intake, or medication adherence.

CHW-led Diabetes Education and Support

In a Houston, TX-based study (USA), Spanish-fluent CHWs served as cultural peers to Spanish-speaking low-income Latino(a) adults with T2DM [38]. In the intervention group, CHWs led monthly IP diabetes education group visits for 6 months and engaged in weekly m-Health communication (phone call or text) with study participants. To increase the CHWs’ knowledge of T2DM, they received quarterly diabetes training from physicians via telehealth. Participants randomized to the control group received usual care and placed on a wait list to receive the intervention after 6 months.

At 6 months, mean HbA1c in the intervention group was significantly lower (-0.98%) than controls (p = 0.002) and decreased to a significantly greater extent among intervention group participants with uncontrolled diabetes (-1.31%) compared to control group participants with uncontrolled diabetes (p = 0.007). Mean systolic blood pressure decreased significantly more in the intervention group than control group (-6.89 vs. + 0.03 points; p = 0.023) and the intervention group was also had a significantly greater reduction in mean diastolic blood pressure vs. the control group (-3.36 vs. + 0.2 points; p = 0.046). No significant between group differences were observed for body weight. Significantly more patients in the intervention group started insulin and statin medications compared to controls. Preventive care measures (B12 screening, foot and eye exams, immunizations, urine microalbumin) occurred significantly more often in the intervention group than the control group (p < 0.001).

Peer Coaching Supported DVD Education

The personal stories of peers with T2DM were leveraged in an intervention based on social cognitive theory that aimed to improve medication adherence among low-income African American adults with T2DM [39]. The study took place in a rural Alabama region predominantly populated by African Americans that includes some of the poorest counties in the US, with a prevalence of diabetes of roughly twice the national average [40]. Medical resources are scarce and age-adjusted mortality rates are 39% higher for African Americans compared to the US average. Study participants were taking diabetes oral medications and who reported medication non-adherence or wanted help taking their medications. Peer coaches were community residents who had diabetes or took care of family members with diabetes.

Intervention participants were provided with DVDs that featured diabetes education and personal stories about how community members accepted their diagnosis of T2DM and overcame barriers to medication adherence during a six-month, 11-session behavioral diabetes self-care program delivered by peer coaches over the phone. During a 6-week intensive phase, intervention participants watched a 15–30 min DVD each week about the challenges and successes experienced by patients with diabetes and afterwards discussed the video’s content with a peer coach in a 30–60 min telephone call. After 6 weeks, peer coaches made biweekly or monthly telephone calls over the remaining 4.5 months to monitor participants’ progress, review information, provide education, and develop self-management strategies. Control participants received a self-paced general health program on topics that were independent of study outcomes.

After 6 months, the study’s primary outcome, medication adherence, significantly improved in the intervention group vs. controls (-0.25; p < 0.0001), however the study arms did not differ in mean HbA1c, blood pressure, body weight, or LDL cholesterol. Intervention participants’ beliefs in the necessity of medications, their medication use self-efficacy, and their diabetes self-efficacy significantly increased, although participants’ self-rated quality of life was not significantly improved. The authors speculated that improvements in medication adherence may not have been substantial enough to affect the physiological measures after 6 months. A limitation that potentially attenuated outcomes was that follow-up blood draws for some participants were delayed for a significant period of time after completion of the intervention.

SMAs Utilizing Telehealth and Virtual Environments

Telehealth became heavily-utilized during the COVID-19 pandemic and has proven comparable to IP consultations in clinical effectiveness [41]. In Boston, MA (USA) a two-arm trial was implemented to compare conventional IP diabetes SMAs to virtual reality-augmented diabetes SMAs conducted via telehealth for African American or Black and Hispanic or Latina adult women with uncontrolled HbA1c (≥ 8.0%) [42].

A 3D computer-based simulated environment was used to engage study participants in SMAs and DSME using a telehealth platform with “immersive, experiential learning with animated educational content” [42]. Virtual world (VW) participants used personal avatars to represent themselves and engage with peers to carry out behavioral changes. They used the avatars to engage in SMAs and to practice the use of positive health behaviors such as dance and social support. Those in the VW group were provided with laptops and wireless internet access.

Both IP and virtual SMAs lasted 120 min including taking vital signs, documenting acute and chronic illness symptoms, health system visits, and self-management activities. All participants received the same 8-module DSME curriculum on topics including diabetes self-monitoring, preventive care, healthy eating, exercise, and stress management.

English- and Spanish-language cohorts of 6–12 individuals met in IP or VW settings for 8 weekly diabetes SMAs of 2-h duration. A healthcare provider met individually with each participant in a separate physical space or via a private telehealth session or telephone call for a one-on-one consult during every SMA. After 8 weeks, participants entered a 16-week maintenance phase when no formal SMAs occurred.

An intention-to-treat (ITT) analysis showed significant mean improvements in HbA1c from baseline to 6 months in the IP group (-0.8%; 10.2% to 9.4%) and in the VW group (-0.5%; 9.7% to 9.2%); between group differences were not statistically significant suggesting that the VW-based SMAs were no less effective than the traditional IP SMAs. Changes in HbA1c in a subset of participants completing at least 6 SMAs per protocol (PP) were similar to the results of the ITT analysis in that between group means were not significantly different.

Significant improvements in levels of physical activity were not observed in either the ITT or PP analyses. The investigators reported that participants had difficulties using the equipment to measure physical activity. Changes in depression scores within the IP and VM groups were non-significant, although substantial within group improvements were observed for diabetes distress in both groups.

Mechanisms of Peer Support

Peer support can play an important role in improving patients’ glycemic outcomes in T2DM [43]. Conceptually, peer support can be grounded in social cognitive theory [44] and self-determination theory [45]. Social cognitive theory proposes that learning can occur by observing the behavior of others and its consequences. Without the necessity of direct personal experience for learning, the volume of knowledge and skills that can be quickly learned by patients with T2DM is greatly expanded compared to experiential learning. Further, by observing peers as “social models,” self-efficacy and mastery of diabetes self-care behaviors may increase [44].

Self-determination theory posits that individuals are influenced by three factors: their capacity to exert control over their life and their future, their sense of competence (which increases with positive peer feedback which in turn increases motivation), and their connectedness to others [46]. Peer support can reinforce all 3 of these factors, particularly by providing positive feedback on behaviors, increasing self-efficacy, and leading to satisfying interpersonal connections [45].

The mechanisms of peer support relationships include having a shared lived experience, the identification of individual strengths, the provision of social and practical support, and the unique position of the peer as fellow patient, peer support specialist, or CHW cultural peer rather than a medical provider [47]. The mutual sharing of personal difficulties and experiences of suffering have been shown to increase feelings of normalization, connectedness and shared humanity [48, 49]. Because peers have faced many of the same problems, peer support is a practical way to learn ways to overcome barriers to diabetes self-management [6].