Review ArticleSystematic ReviewA

Peer Support Interventions for Adults With Diabetes: A Meta-Analysis of Hemoglobin A1c Outcomes

Sonal J. Patil, Todd Ruppar, Richelle J. Koopman, Erik J. Lindbloom, Susan G. Elliott, David R. Mehr and Vicki S. Conn
The Annals of Family Medicine November 2016, 14 (6) 540-551; DOI: https://doi.org/10.1370/afm.1982

Article Figures & Data

Figures

  • Figure 1
    Figure 1

    Results of the literature search.

    Note: Performed according to methods outlined by Moher et al.20

  • Figure 2
    Figure 2

    Effect of peer support interventions on hemoglobin A1c levels.

    HbA1c = hemoglobin A1c; SMD = standardized mean difference.

    Notes: Random effects model. I2 = 60.66%; P for heterogeneity = .001.

  • Figure 3
    Figure 3

    Subgroup analysis of the effect of peer support interventions on hemoglobin A1c levels in studies by predominant race/ethnicity of the participants.

    HbA1c = hemoglobin A1c; SMD = standardized mean difference.

    Notes: African American subgroup: I2 = 58.60%, P for heterogeneity = .08. Asian subgroup: I2 = 0.00%. Hispanic subgroup: I2 = 17.12%, P for heterogeneity = .30. White non-Hispanic subgroup: I2 = 59.41%, P for heterogeneity = .02.

  • Figure 4
    Figure 4

    Subgroup analysis of the effect of peer support interventions on hemoglobin A1c levels in studies by predominant minority status of the participants.

    HbA1c = hemoglobin A1c; SMD = standardized mean difference.

    Notes: Minority subgroup: I2 = 9.24%, P for heterogeneity = .35. Nonminority subgroup: I2 = 45.90%, P for heterogeneity = .06.

Tables

  • Table 1

    Descriptive Summary of the Characteristics of Included Studies

    Study, YearPatients, No.Setting, PopulationInterventionControlDuration of Peer TrainingPeer SupervisionAttendance Rates, Intensity
    Keyserling et al,23 2002133United States: African American women
    Mean HbA1c 11.1%    		Same as control; also 3 group sessions and monthly telephone calls from a peer counselor for 12 mo
    7 peer counselors for 67 patients    		Individual counseling visits with nutritionist at mo 1, 2, 3, and 416 hNone reported81% of participants attended at least 1 peer counselor session, 30% attended 2 sessions, and 19% attended 3 sessions
    Average phone calls per participant: 9.7
    Lorig et al,26 2008417United States: Spanish-speaking patients
    Mean HbA1c 7.41%    		6-wk program:
    2.5 h weekly led by 2 peer leaders
    2 peers per 10–15 patients    		Usual care24 hRandom observationsNot reported
    Lorig et al,25 2009345United States: white, non-Hispanic patients 67%
    Mean HbA1C 6.72%    		6-wk program:
    2.5 h weekly led by 2 peer leaders
    2 peers per 10–15 patients    		Usual care24 hRandom observationsMean attendance of 4.9 of 6 sessions
    Dale et al,32 2009187United Kingdom: white, non-Hispanic patients 96.65%
    Mean HbA1c 8.55%    		Peer supporter called participants after change in their diabetes care at days 7–10, 14–18, 28–35, 56–70, 120–150
    1 peer called median of 10 patients    		Usual care2-d training program developed for the studyNone reportedMean number of calls each patient received: 4.5 (range = 1–6 calls; SD, 2.2)
    Cade et al,31 2009207United Kingdom: white, European-origin patients 95%
    Mean HbA1C 7.4%    		Peer educators delivered 2-h education sessions per week for 7 wk
    5 peer educators for 86 participants    		Usual careResidential training course provided by the Expert Patients Programme of the National Health ServiceNone reportedOf 110 participants in intervention group, 18 attended only 1 session and 22 attended all 7 sessions
    63 participants attended final diabetes-specific session
    Heisler et al,22 2010244United States: Male veterans, white non-Hispanic 82%
    Mean HbA1c 7.97%    		Peer partners encouraged to call weekly
    Three optional 1.5-h face-to-face sessions at 1, 3, and 6 mo    		Usual careInitial 1.5-h training in peer communication skillsInteractive voice response–facilitated telephone platform recorded call initiation, frequency, and duration90% of peer pairs had at least 1 conversation; average number of calls per pair per month was 2.4, 1.7, 1.4, 1.0, 0.9, 0.8 for 6 consecutive mo
    Philis-Tsimikas et al,27 t2011207United States: Mexican Americans
    Mean HbA1c 10.4%    		Weekly 2.5-h education sessions by peer educators for 8 wkUsual care40 h of training, plus trainees cotaught 2 series of classes with their trainer and taught 2 series on their own, under observation by the trainerClasses were audio-recorded and reviewed using checklistsNot reported
    Smith et al,34 2011388Republic of Ireland: Cluster randomized
    50% population low income
    Ireland general demographics 94.3% white non-Hispanic
    Mean HbA1c 7.2%    		9 peer supporter–facilitated sessions over 2 y; at mo 1, at mo 2, and every 3 mo thereafterUsual careTwo 3-h evening training sessions conducted by research teamsMeetings were recordedParticipants attended mean of 5 peer support meetings; 18% never attended a meeting
    Long et al,24 201277United States: African American veterans
    Mean HbA1c 9.85%    		Weekly telephone calls by peer mentors; 34 mentors and 39 participantsUsual careSingle 1-h one-on-one training sessionNo supervisionFirst month, average of 4 calls/mo; by 6 mo, average of 2 calls/mo
    Gagliardino et al,37 2013198Argentina: Hispanic, nonminority in country of residence
    Mean HbA1c 7.2%    		4 weekly peer educator sessions of 90–120 min initially; 1 at 6 mo followed by weekly calls for 6 mo then biweekly calls for 3 mo
    Additional face-to-face visits among peers and their supportees were scheduled every second month if specific issues warranted    		4 weekly sessions of 90–120 min initially; 1 at 6 mo by educators3-d intensive, structured, small group interactive courseDiabetes knowledge of participants tested with multiple-choice questionnaireNot reported
    Siminerio et al,29 201368United States: white non-Hispanic ethnicity >80%
    Mean HbA1c 8.65%    		Same as control followed by monthly peer calls for 6 mo for diabetes self-management support6 wk of CDE diabetes self-management education intervention1-d peer training workshop for CDE with companion workbook-CDE then trained peers in 2-3–h small sessionsContact logs for communications; goal selections were trackedAverage of 5.03 calls per participant by peer supporter lasting approximately 25–30 min
    Thom et al,30 2013299United States: Hispanic 46.65%, African American 31.25%; coded as predominantly Hispanic
    Mean HbA1c 9.95%    		Telephone contacts with peers at least twice a month and 2 or more in-person contacts in 6 mo
    Coaches worked with median of 7 patients    		Usual care36 h of training over 8 wk in either English or SpanishNot reportedMedian of 5 (range = 0–29) interactions with the peer health coach
    123 patients (83%) had at least 1 interaction; most interactions (76.6%) were by telephone, and the remainder were in person
    Chan et al,36 2014628China: Chinese speaking 100%
    Mean HbA1c 8.2%    		Same as control and peer supporter telephone calls: biweekly for 3 mo, then monthly for 3 mo, and then 1 call every other month for 6 mo; anticipated 15 min per callComprehensive assessment, personalized report, 2-h nurse-led empowerment class, follow-up primary care visit with repeated laboratory assessment and mailing of follow-up reportsFour 8-h training sessionsPeer supporter completed and mailed checklists to document discussion items, duration of each call, and relevant remarks every 3 moMedian of 20 calls per patient
    Simmons et al,33 2015644England: Cluster randomized factorial design; white, non-Hispanic >90%
    Mean HbA1c 7.3%    		Peer-led group education sessions once a month for at least 5 mo and telephone/e-mail for 1:1 counselingUsual careMain training 14 h plus 3.5 h diabetes education sessionNot reported61.5% participants attended at least 1 education session; most participants had telephone or e-mail contacts with peers
    Mean number of group attendances: 3.7
    Safford et al,28 2015424United States: cluster randomized trial; African American >90%
    Mean HbA1c 7.9%    		Same as control and initial 45–60–min in-person or telephone get-to-know session with peer supporter followed by weekly calls for 2 mo followed by monthly calls for 8 mo1 h of group diabetes education class, 5-min counseling session, and diabetes report card12 h over 2 dContacts documented on forms and random contacts with intervention participantsMean number of contacts: 13.3 (SD, 8.1) 8.3% of participants had no contacts
    Ayala et al,21 2015336United States: predominantly Hispanic
    Mean HbA1c 8.7%    		8 telephone or in-person contacts with peer supporter in first 6 mo, then as needed contacts in the last 6 mo; 92% of participants had telephone contacts
    5–8 patients per peer leader    		Usual care40–50 hContact logs maintained and tracked by peer leader coordinatorMedian number of contacts per participant: 4 (range = 1–24) 7% received no intervention
    McGowan,35 2015361Canada: race/ethnicity not given
    Mean HbA1c 7.19%    		Two participant groups received 2 varied types of peer-led self-management programs with varying components: weekly meetings for 6 wkUsual care24 hSession attendance was loggedMean attendance for intervention group: 5 sessions

    • CDE = certified diabetes educator; HbA1c = hemoglobin A1c.

  • Table 2

    Quality of the Included Studies Assessed With the Cochrane Risk of Bias Tool12

    Study, YearAdequate Sequence GenerationAllocation ConcealmentBlinding of Investigators and/or Outcome AssessorsComplete Outcome Data
    Keyserling et al,23 2002YesYesNoYes
    Lorig et al,26 2008YesYesaNoYes
    Lorig et al,25 2009YesYesaNoYes
    Dale et al,32 2009YesYesNoYes
    Cade et al,31 2009YesbNocNocYes
    Heisler et al,22 2010YesYesYesYes
    Philis-Tsimikas et al,27 2011YesYesaNoaYes
    Smith et al,34 2011YesYesNoYes
    Long et al,24 2012YesYesYesYes
    Gagliardino et al,37 2013YesaNocNocNoc
    Siminerio et al,29 2013YesNoaNoYes
    Thom et al,30 2013YesYesNoaYes
    Chan et al,36 2014YesYesNocYes
    Simmons et al,33 2015YesYesYesYes
    Safford et al,28 2015YesYesNoYes
    Ayala et al,21 2015YesNocNocYes
    McGowan,35 2015YesYesNoYes

    • Notes: Yes indicates low risk of bias; No indicates high risk of bias. Selective outcome reporting risk was considered minimal as we selected only studies that reported hemoglobin A1c level as the outcome. No other major biases were noted in the included studies. None of the trials blinded participants.

    • ↵a Author response to information requests.

    • ↵b Likely low risk of bias but no details on sequence generation available from article and author.

    • ↵c Not reported or no response from author.

Additional Files

  • The Article in Brief

    Peer Support Interventions for Adults With Diabetes: A Meta-Analysis of Hemoglobin A1C Outcomes

    Sonal J. Patil , and colleagues

    Background Peer support is a promising approach to improving and sustaining diabetes self-management. This study analyzes existing research to assess the effectiveness of peer-support interventions on improving glycemic control in adult patients with diabetes as measured by HbA1C, compared to patients who received similar care except for peer delivered interventions.

    What This Study Found Peer support interventions delivered by people affected by diabetes are associated with a small but statistically significant reduction in glycosylated hemoglobin, with larger effects among minority (particularly Hispanic) participants. There was an overall 0.24 percent improvement in HbA1c. Studies with predominantly Hispanic participants showed an HbA1c improvement of 0.48 percent in the peer support intervention group compared with the control group. In contrast, the pooled effect size from studies with predominantly white, non-Hispanic participants showed no improvement in HbA1c level with peer support interventions.

    Implications

    • Peer health coaches might be providing more culturally appropriate health education in ethnic minority populations, particularly in the Latino population. The authors call for future research to assess the effect of peer interventions on long-term patient-centered outcomes.

  • Annals Journal Club

    Nov/Dec 2016: A Meta-Analysis of Peer-Support Interventions for Adults With Diabetes


    The Annals of Family Medicine encourages readers to develop a learning community of those seeking to improve health care and health through enhanced primary care. You can participate by conducting a RADICAL journal club. RADICAL is an acronym for Read, Ask, Discuss, Inquire, Collaborate, Act, and Learn. It also indicates the need to engage diverse participants in thinking critically about, and then acting on, important issues affecting primary care.1

    HOW IT WORKS

    In each issue, the Annals selects an article and provides discussion tips and questions. We encourage you to post a summary of your journal club's conversation. (Open the article and click on "TRACK Discussion: Submit a comment.") Details are available at http://www.AnnFamMed.org/site/AJC/.

    CURRENT SELECTION

    Article for Discussion

    Patil S, Ruppar T, Koopman RJ, et al. Peer support interventions for adults with diabetes: a meta-analysis of hemoglobin A1C outcomes. Ann Fam Med 2016;14(6):540-551.

    Discussion Tips

    Systematic reviews and meta-analyses are observational studies that provide an overview of a literature. However, analysis of these types of studies requires additional skills over those needed to analyze randomized controlled trials. This article offers an opportunity to investigate positive and negative aspects of this type of study in assessing the effect of peer support interventions on the narrow but summative outcome of glycosylated hemoglobin.

    Discussion Questions

    • What is a systematic review? What is a meta-analysis? How are they related? What are their primary weaknesses?
    • What was the research question for this review?
      • Was the scope of the question appropriate (too broad or narrow)? How adequate was the search?
      • What kind of important studies might have been missed due to scope or search criteria?
    • What is PRISMA?2 What are the benefits and downsides of following this reporting structure?
    • What is heterogeneity and why is it an important concept in systematic reviews and meta-analysis?
      • How is heterogeneity measured in this study?
    • What is publication bias? Why is this an important concept in systematic reviews and meta-analyses?
      • How is this investigated in the study?
    • What is a standard mean difference? Why was this the primarily reported outcome?
    • What are the main findings? Was an important level of heterogeneity identified, and how does that affect your interpretation of the findings?
    • How did the results change when investigated by different subgroups?
      • What is an ecological fallacy? How might individual patient data (instead of trial level) have strengthened the study findings?
    • What are the strengths and weaknesses of hemoglobin A1C as an outcome measure?3 Why do you think the authors investigated hemoglobin A1C?
    • Were the decreases in hemoglobin A1C "clinically meaningful"? Were they durable? How long was the follow-up of the included studies?
    • Are the patients in the primary studies comparable to patients in your practice?
    • Will this article change your practice? If so, how?

    References

    1. Stange KC, Miller WL, McLellan LA, et al. Annals Journal Club: It's time to get RADICAL. Ann Fam Med. 2006;4(3):196-197 http://annfammed.org/content/4/3/196.full.
    2. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ. 2009;339:b2700.
    3. Yudkin John S, Lipska Kasia J, Montori Victor M. The idolatry of the surrogate. BMJ. 2011;343:d7995.

  • Supplemental Appendixes

    Files in this Data Supplement:
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Get Permissions

Jump to section

Related Articles

Cited By...

More in this TOC Section

Similar Articles

Subjects

Keywords