Peer support for physical health improvement: recovering ‘stolen years’?


People with serious mental illness are at greater risk of premature death; they may die between 10 and 20 years earlier and these ‘stolen years’ are mostly attributable to cardiovascular, respiratory and metabolic diseases.

Lifestyle interventions, including physical activity and dietary advice, have been shown to improve the physical health of people with serious mental illness, with those that incorporate social support and motivational components showing most promise.

Incorporating peer support into health improvement interventions in other areas of chronic disease management has shown promising success. In the woodland, we are not strangers to the long and noble history of peer support in mental health, but also the controversy of packaging ‘peer support’ as an intervention and testing it in controlled experiments (see: No support for peer support including the insightful discussion underneath). How do these arguments fare when we look at peer support interventions for physical health improvement for people with serious mental illness?

The authors of this paper (Stubbs B. et al, 2016 – interestingly the lead author is a physiotherapist), based at the Maudsley in London, conducted a systematic review of the available evidence for peer support interventions (PSI) to improve physical health for people with severe mental illness. Specifically, the aims of the review were to establish if PSI improved:

  • Physical health parameters (e.g. metabolic risk factors)
  • Physical health appointment attendance, and
  • Outcomes, attendance and adherence to lifestyle interventions (e.g. physical activity, nutritional interventions).
Lifestyle interventions, including physical activity and dietary advice, have been shown to improve the physical health of people with serious mental illness.

Lifestyle interventions, including physical activity and dietary advice, have been shown to improve the physical health of people with serious mental illness.


The authors searched relevant databases and clearly explained the processes they adopted for selecting the studies for the review. Both interventional and observational studies were included; due to the small number of studies finally included in the review and the broad mix of study designs and interventions, the authors did not attempt meta-analysis (statistical pooling of data). Instead they identified the key results and limitations of each study to qualitatively synthesise the evidence for the effectiveness of peer support interventions.  


Seven unique studies, all conducted in the USA, were included in the review. Three were pilot randomised controlled trials which compared a peer support intervention against either treatment as usual or a waiting list control. The four remaining studies were pre- and post-test design studies, three of which also incorporated qualitative interviews with participants. In total, 220 participants took part in the seven studies. In all studies, ‘peer support’ had been formalised into a Peer Support Worker role.

Health improvement interventions

The duration of the interventions ranged from 6 to 24 weeks and encompassed a number of different approaches to physical health improvement:

  • Weight loss (2 studies)
  • Personalised fitness, healthy eating and weight loss (1 study)
  • Confidence in primary care encounters (1 study), and
  • Broad-based self management (3 studies), including elements such as exercise, healthy eating, medication management, sleep, working with a doctor, behaviour change and maintenance.

Role of Peer Support workers

The role of the Peer Support Workers was to share their personal experience with the study participants and lead, or co-lead with another person who had some kind of professional training, the delivery of the specific intervention.


In summary, there was a suggestion that peer support interventions had a positive impact on a wide range of health-related outcomes, but confidence in these outcomes was low:

  • Attendance at primary care appointments (2 studies, both significant)
  • Increased fitness and self-reported physical activity (3 studies, only one result was significant)
  • Weight loss (3 studies but no results were significant)
  • Increased reporting of physical complaints (1 study, not significant)
  • Improved physical health related quality of life (1 study, not significant)
  • Improved pain and pain causing interference (1 study, significance not specified)
  • Improved physical functioning (1 study, significance not specified)

Peer support showed some improvements in physical health, but confidence in the results was low.


There is inconsistent evidence to support the use of Peer Support Workers to improve the physical health and promote lifestyle change among people with serious mental illness.

The authors make this conclusion because the evidence base is poor; it is characterised by small studies with a mix of designs and interventions. They do acknowledge that this is perhaps unsurprising, given that Peer Support Interventions are generally still in their infancy. The suggestion of improvements across the seven studies indicates that further investigation is required, which is supported by the insights from the qualitative work where study participants endorsed the role and values of Peer Support Workers as helpful and a source of hope.

Strengths and limitations

The main limitation of the review method was that no formal assessment or grading of quality was made on the included studies; a step usually included in systematic reviews. But as no attempt was made to pool data for meta-analysis, the potential for the weaker studies to overly bias any outcomes is reduced. In fact, the authors take a measured approach to reporting the limited health improvements that were observed across the included studies.

The authors acknowledge that the Peer Support Worker role varied across the studies and was not always well described in the original papers. For example, for one of the included studies, all the review authors can offer is that ‘the Peer Support Worker played an integral role in the intervention’.  This variance and lack of description obscures our understanding about the unique contribution this role can make to physical health improvement.

Although the authors comment on the extent to which they consider outcomes can be attributed to peer support (in table 2 in the paper), it is not possible from this review to disentangle the effectiveness of the specific health improvement interventions and the potential influence of the peer support on any outcomes. Only one study reported that it was using an established (and presumably evidence-based?) intervention. For the other studies, it was not apparent whether the intervention had already been demonstrated to be effective. Moreover, in the controlled studies, similar interventions without Peer Support Workers were not used as the comparator.

The authors do offer some commentary on Peer Support Workers and which factors may influence how successful individuals can be in such roles. This recognition is important and, especially for those thinking about developing such roles in practice, another Mental Elf Blog can point to a myriad of factors that need to be taken into account.


Can we disentangle the quality of the intervention and the contribution of peer support on outcomes?


So while, as the authors rightly argue, there is an urgent need to tackle the scandal of premature mortality and address the physical health disparity for people with serious mental illness, they conclude there is insufficient evidence to recommend the widespread implementation of peer support interventions for physical health improvement.  

However, we have to remember that the review does not provide evidence that suggests PSIs are ineffective for health improvement. But rather there is a lack of well designed studies that have adequately investigated this approach. To address this paucity, I’m not sure that blandly stating ‘more research is needed’ is that useful. Researchers need to continue to think carefully about the real value of peer support and whether expecting it to have a measurable impact on specific health outcomes is the best way to go.

At the root of peer support is the, often serendipitous, sometimes magical, person-to-person relationship. And the benefits described by those experiencing peer support relate to diverse factors like personal confidence, self-esteem and empowerment, through to social inclusion, challenging stigma and discrimination and collective action (Faulkner and Kalathil, 2012). Trying to shoe horn the complexity of peer support into the neat models required for the formalised world of health intervention research seems at odds.

Rather than keep repeating studies that fail to due justice to peer support, we need to find a way of retaining the essential value base of peer support and carry this through to research methods that can explore its nuanced value to health improvement; physical or mental. 


Can peer support research address measurable physical health outcomes?


Primary paper

Stubbs B, Williams J, Shannon J, Gaughran F, Craig T. (2016) Peer support interventions seeking to improve physical health and lifestyle behaviours among people with serious mental illness: A systematic review. Int J Mental Health Nurs, 25: 484–495. doi:10.1111/inm.12256 [PubMed]

Other references

Faulkner A and Kalathil J. (2012) The freedom to be, the chance to dream: Preserving user-led peer support in mental health, Together for mental wellbeing, London [Full text]

Share on Facebook Tweet this on Twitter Share on LinkedIn Share on Google+