Is 'volunteering' a Public Health Intervention: A Systematic Review and Meta-Analysis (OP74)

Background Community participation interventions such as volunteering have been identified in the UK Government’s ‘Building the Big Society’ policy (2010) as a way of engaging people in their local communities and improving social capital, while the Marmot Review (2010) concluded they may have potential public health benefits such as improving wellbeing and decreasing health inequalities. Internationally, 2011 saw the European ‘Year of the Volunteer’, while the US Corporation for National and Community Service Strategic Plan (2011-2015) advocates supporting volunteer’s wellbeing and prioritising recruitment and engagement of underrepresented populations. Within this context, we examined the effect of formal volunteering on volunteer’s physical and mental health and survival, and explored the influence of volunteering type and intensity on health outcomes. Methods We sought evidence from experimental and cohort studies comparing the physical and mental health outcomes and mortality for a volunteering group (intervention) compared with a non-volunteering group (control). We searched electronic databases (Cochrane Library, Medline, Embase, PsychINFO, CINAHL, ERIC, HMIC, SSCI, ASSIA, Social Care Online, Social Policy and Practice) from inception to January 2013 and used citation tracking. No language, country or date restrictions were applied. Methodological quality was appraised and a risk of bias score generated. Data were synthesised using vote counting and random effects meta-analysis of mortality risk ratios. Results 40 papers were selected: 5 randomised controlled trials (RCTs, 7 papers); 4 non-RCTs; and 17 cohort studies (29 papers). While most cohort studies were large and well designed (25/29 low risk bias, 4/29 moderate risk), most RCTs (4 studies) were at moderate or high risk of bias (e.g. small samples). Cohort studies show volunteering has favourable effects on depression, life satisfaction, wellbeing but not on physical health. However, these findings were largely unconfirmed by experimental studies. Meta-analysis of 5 cohort studies found volunteers to be at lower risk of mortality (risk ratio: 0.78 [95% CI 0.66, 0.90]). There was insufficient evidence to demonstrate a consistent influence of volunteering type or intensity on outcomes. Conclusion Observational evidence suggests that volunteering benefits mental health and survival although the causal mechanisms are unclear and selection effects cannot be discounted. Consequently, there is limited robustly designed research to guide the development of volunteering as a public health promotion intervention. Future studies should explicitly map intervention design to clear health outcomes as well as use pragmatic RCT methodology to test effects