Peer support for patients with type 2 diabetes: cluster randomised controlled trialBMJ 2011; 342 doi: https://doi.org/10.1136/bmj.d715 (Published 15 February 2011) Cite this as: BMJ 2011;342:d715
All rapid responses
Approaches to improve self management in long term conditions are
important research topics. But.... is a cluster RCT in general practice
the appropriate methodology to measure the effectiveness of a "simple"
intervention like "peer support" for any longterm condition (LTC) that
requires patients to self manage their condition between clinic visits?
The sample size calculation included an aspiration to reduce the
HbA1c by 0.9%. The reported baseline HbA1c was 7.2, just how low did the
authors want the HbA1c to go1?
Even if this study is "recognised as being well designed", the
research consists of a complex analysis of a simple intervention that took
years to complete and probably was expensive to undertake (I estimate at
least 350k). Meanwhile, 60 miles up the road in N. Ireland, the recorded
prevalence of diabetes increased by almost 33% during that time and the
evidence base for glucose control also changed!1
We need to continue to design and evaluate programs that improve self
management of LTCs and further develop the evidence base for the major
Public Health challenges associated with LTCs. The optimum way to do this
needs further debate and discussion but it can be done2,3.
1. Ismail-Beigi F, et al. Effect of intensive treatment of
hyperglycemia on microvascular outcomes in type 2 diabetes: an analysis
of the ACCORD randomised trial. Lancet 2010;376:419-30.
2. Aanand D, et al. Comparative effectiveness of goal setting in
diabetes mellitus clinics. Randomised Clinical Trial. Arch Intern Med;
171:5 (March 14 2011 ) Abstract
3. Hughes C.R et al, Sustained benefits of continuous subcutaneous
insulin infusion. Arch Dis Child 2010 Nov 25 (epub).
Competing interests: No competing interests
We welcome these comments on our study as they highlight the international
interest in this area of healthcare They also reflect many of the points
we made in our paper about the limited existing evidence base for peer
support and the difficulties demonstrating clinical improvement when
baseline control of risk factors was relatively good. We agree also that
there is evidence indicating that social support is beneficial for health,
however, it is unclear whether this type of peer support can be developed
by arranging for patients to meet as peers.
We would not agree with the description of our peer support intervention
as "periodic diabetes information groups". Our intervention falls within
the definition of peer support which is "the provision of social support
from an individual with experiential knowledge based on a sharing of
similar life experiences" (1) and was developed in the context of a strong
theoretical underpinning of social support following guidance from a
social scientist. The groups were designed to facilitate social support by
peers and were not designed to replicate health professional interactions.
This definition of peer support emphasises the sharing of experiential
knowledge and we would argue that the effective creation of individualised
behaviour change management plans goes beyond the concept of peer support.
We did not specifically train the peers in these types of professionalized
approaches as we were aware that excess training can dilute the peer
effect and create differences between the peers and those they support.
Fisher and Boothroyd question whether peer support was actually achieved
because of poor attendance. We address this in the discussion but as this
was a pragmatic trial designed to test the effectiveness of peer support,
issues around implementation and feasibility are highly important for
those considering introducing and financing peer support type systems for
all people with type 2 diabetes. Poor attendance in the voluntary aspects
of healthcare is an issue that merits further research and has to be
factored into recruitment and analysis of studies like ours. They also
question the limited contact between peer supporters and group
participants between meetings. This was not in any way discouraged by the
research team and project manager and there was some contact outside of
meetings between peer supporters and participants where both parties
agreed. However it did not occur in all groups as a matter of course. We
would again like to emphasise the additional features of the peer support
intervention which included the ongoing support given to peer supporters
by the project manager and the 'frequently asked questions' system both of
which could be considered as ongoing training and guidance. The peer
supporters themselves were busy people generously volunteering their time
to lead groups There will always be a need to balance the demands placed
on volunteers such as peer supporters and to take the inevitable attrition
into account in design of such studies. In our study, the majority of peer
supporters remained in their role for the duration of the intervention,
perhaps because we recognized the burden to be placed on them in the
design and execution of the study.(3)
Dr Farrell makes interesting comments regarding the nature of our
intervention and the remuneration to practices and peer supporters. The
changes she suggests would have fundamentally changed the intervention and
could be addressed by additional research. Her comment that programmes
that assist better self-management of long term conditions "intuitively
feel like the right thing to do" highlights why it is so important to test
the effectiveness of these interventions in pragmatic real world settings
before we invest heavily in their promotion.
There was no suggestion in the original paper of our evidence
"overwhelming" others. Our conclusion, based on this study only, which is
recognised as having been well designed, is that peer support should not
be widely used until further research is carried out. We welcome the
ongoing research being carried out by Peers for Progress and others in
this area of healthcare and look forward to their findings.
1. Dennis CL. Peer support within a health care context: a concept
analysis. International Journal of Nursing Studies. 2003;40:321-32.
2. Giblin PT. Effective utilisation and evaluation of indigenous health
care workers. Public Health Reports. 1989;104(361-368).
3. Paul G, Smith S, Whitford D, O'Kelly F, O'Dowd T. Development of a
complex intervention to test the effectiveness of peer support in type 2
diabetes. BMC Health Services Research. 2007;7(1):136.
Competing interests: No competing interests
A recent paper posted by the BMJ online, "Peer support for patients
with type 2 diabetes: cluster randomized controlled trial"1 reports
disappointing results of peer-led discussion groups. The prominence of the
report and its conclusion that "the results do not support the widespread
adoption of peer support" raise concerns for the field. However, details
of the intervention point up important differences from widely endorsed
features of peer support and call into question the generality of these
A vast amount of evidence links social support to health and well
being. Primates derive great advantage from the support of parents, kin
and familiars.2 Social support is strongly related to numerous health
indicators3 and its absence, social isolation, has been estimated to be as
lethal as smoking a pack of cigarettes per day.4-5 Reviews indicate
widespread benefits6 and promising results of peer support interventions
within important areas like diabetes management.7 Individual controlled
studies indicate substantial benefits of specific peer support
interventions, such as with "Lady Health Workers" reducing by half
prevalence of post-partum depression in Pakistan8 or "Asthma Coaches"
reducing rehospitalization for asthma by half among children with very low
income, single mothers in the U.S.9 Peer support programs are numerous
around the world, but are often poorly evaluated, reported, and
disseminated. Thus, how to organize and deploy peer support remains
The paper by Smith and colleagues1 shows null results in a cluster
randomized evaluation of periodic diabetes information groups led by peers
plus enhancements of clinical care, compared to enhanced clinical care
alone. A number of features of the study may have accounted for the
disappointing results. At entry, the clinical status of participants was
not remarkable (e.g., mean HbA1c = 7.2%), making demonstration of
improvement difficult. Additionally, the enhancement of clinical care
common to both conditions appears to have been appreciable, e.g.,
reduction of systolic blood pressure from 144 to 137 mm Hg in controls.
Most important, however, is the particular approach taken to
operationalizing peer support in this study. The intervention itself
focused on nine peer support group meetings spaced over two years. Based
on the description of "Specific topics discussed in the peer support
meetings" posted at BMJ.com, meetings addressed varied topics of interest
to those with diabetes (e.g., heart and vascular disease, blood sugar
levels, healthy eating, exercise, medications, foot care). "In general,
the groups followed and discussed the planned topics".p. 5,1 However,
meetings appeared not to include a focus on individualized plans for
behavior change and follow up of these plans, a feature commonly observed
to be important in achieving improvements in self management of diabetes
and other chronic diseases.10-12 The description of the "Peer supporter
training," also posted at BMJ.com, indicates there were two, 90-minute
training sessions for the peer supporters. Communication skills and role
play of them were covered only in the second session, which also addressed
lifestyle and medication issues, confidentiality, and support for the peer
supporters. From the report, there appears to have been no contact with
participants outside the group meetings. Those who failed to attend the
meetings were contacted by study nurses and the study manager but
apparently not by the peer supporters themselves.p. 5,1
Was peer support achieved? Average attendance was only five of the
nine meetings schedule over two years. Eighteen percent attended none. If
an intervention was intended to provide peer support but was only modestly
attended, one might question whether peer support has been provided
sufficiently to be tested.
From this important and well designed study, one can conclude that
spending time in an intervention led by a peer is not magic. Offering
adults with diabetes the opportunity to meet in occasional groups led by a
peer to discuss issues of common interest appears insufficient to effect
improvements in clinical status or well being. However, features of peer
support recognized in the field as important may not be well appraised by
this study. In particular, focus on adjusting management plans to the
specifics of individuals' lives, social and emotional support, linkage to
clinical care, individualized contact, ongoing support, and other features
common to successful peer support interventions13-16 do not appear to have
been emphasized in the intervention reported by Smith and colleagues.
Perhaps most important, the apparent limitation of contacts with the peer
supporters to the nine structured meetings appears to have eliminated the
easy availability of peer support often emphasized as a strength of social
It is surely correct that the results of the study of Smith and
colleagues "do not [in and of themselves] support the widespread adoption
of peer support".(abstract) However, it would be a tortured interpretation
of null findings to extend this observation to all of peer support. As
indicated above, there are many reasons to believe that peer support is
indeed highly effective. Rather than overwhelming that evidence, the
present results point to the challenges in developing ways to deliver peer
support that is responsive to the complex and dynamic set of emotional,
practical, and social needs of people with diabetes. Peers for Progress
(peersforprogress.org), a program of the American Academy of Family
Physicians Foundation, is dedicated to promoting global exchange to
identify effective and feasible peer support interventions. Surely the
field needs the "future research" for which Smith and colleagues call, but
the failure of this particular study should neither discourage that
research nor efforts to find effective and efficient ways to bring peer
support to the many who may benefit from it.
The authors are, respectively, Global Director, and Director of the
Program Development Center of Peers for Progress, a program of the
American Academy of Family Physicians Foundation dedicated to evaluating
and promoting peer support in health, health care and prevention around
1. Smith SM, Paul G, Kelly A, Whitford DL, O'Shea E, O'Dowd T. Peer
support for patients with type 2 diabetes: cluster randomised controlled
trial. BMJ 2011;342:d715.
2. Harlow HF. The nature of love. American Psychologist 1958;13:673-
3. Uchino BN. Social support and health: a review of physiological
processes potentially underlying links to disease outcomes. J Behav Med
4. House JS, Landis KR, Umberson D. Social relationships and health.
5. Holt-Lunstad J, Smith TB, Layton JB. Social relationships and
mortality risk: a meta-analytic review. PLoS Med 2010;7(7):e1000316.
6. Swider SM. Outcome effectiveness of community health workers: an
integrative literature review. Public Health Nursing 2002;19:11-20.
7. Norris SL, Chowdhury FM, Van Let K, Horsley T, Brownstein JN,
Zhang X, et al. Effectiveness of community health workers in the care of
persons with diabetes. Diabetic Medicine 2006;23:544-56.
8. Rahman A, Malik A, Sikander S, Roberts C, Creed F. Cognitive
behaviour therapy-based intervention by community health workers for
mothers with depression and their infants in rural Pakistan: a cluster-
randomised controlled trial. Lancet 2008;372(9642):902-9.
9. Fisher EB, Strunk RC, Highstein GR, Kelley-Sykes R, Tarr KL,
Trinkaus K, et al. A randomized controlled evaluation of the effect of
community health workers on hospitalization for asthma: the asthma coach.
Arch Pediatr Adolesc Med 2009;163(3):225-32.
10. Bodenheimer T, Davis C, Holman H. Helping Patients Adopt
Healthier Behaviors. Clinical Diabetes 2007;25:66-70.
11. Bodenheimer T, Handley MA. Goal-setting for behavior change in
primary care: an exploration and status report. Patient Educ Couns
12. Funnell MM, Brown TL, Childs BP, Haas LB, Hosey GM, Jensen B, et
al. National Standards for diabetes self-management education. Diabetes
Care 2011;34 Suppl 1:S89-96.
13. World Health Organization. Peer Support Programmes in Diabetes:
Report of a WHO Consultation 5-7 November 2007. Geneva, 2008.
14. Boothroyd RI, Fisher EB. Peers for progress: promoting peer
support for health around the world. Fam Pract 2010;27 Suppl 1:i62-8.
15. Funnell MM. Peer-based behavioural strategies to improve chronic
disease self-management and clinical outcomes: evidence, logistics,
evaluation considerations and needs for future research. Fam Pract 2010;27
16. Heisler M. Different models to mobilize peer support to improve
diabetes self-management and clinical outcomes: evidence, logistics,
evaluation considerations and needs for future research. Fam Pract 2010;27
17. Fisher EB, Jr. Two approaches to social support in smoking
cessation: Commodity Model and Nondirective Support. Addictive Behaviors
Competing interests: The authors are, respectively, Global Director, and Director of the Program Development Center of Peers for Progress, a program of the American Academy of Family Physicians Foundation dedicated to evaluating and promoting peer support in health, health care and prevention around the world.
The approach adopted by the authors raises a number of questions:
Is peer support a complex intervention or just an approach that is
difficult to measure the impact of the intervention?
Would the outcome have been different if the peer supporters had been
paid ?5000 and the participating practices paid ?300?
The study coincided with the introduction of structured diabetes care
in participating practices but in neither the control or intervention
groups were improvements in diabetes control observed....?
Should the study have compared peer support with Structured Patient
Education or an e-learning program to improve self management?
Programs to assist in better self management of long term conditions
between clinic visits intuitively feel to be the "right thing to do". Do
we need RCTs to confirm this?
Competing interests: No competing interests