Review of how (and how not) to help poorer people improve their diet and do more exercise

Shouting with a megaphone

We know that type 2 diabetes is more prevalent in groups with lower socioeconomic status, and we know that improving diet and taking exercise can reduce the risk of type 2 diabetes.  So how can we help poorer people to act on diet and exercise advice?

A recent mixed methods systematic review looked at the effectiveness and acceptability of different diet and exercise interventions delivered in a community setting in the UK.


The reviewers conducted a comprehensive search for studies of low socio-economic status (SES), diet and physical activity.  They looked for qualitative and quantative studies of community-based physical activity or dietary interventions.

Two independent reviewers identified, assessed and extracted data from the studies found.

They conducted a narrative synthesis of the evidence.  They also cross-referenced the themes from the qualitative studies against the methods of the quantitative studies to see if there was evidence of an association between themes and quantitative outcomes.


An older man

It’s important that health workers have an understanding of their audience in delivering diet and exercise interventions

12 quantitative studies were found, three of which were RCTs.  All but two of the studies were rated as having a high risk of bias.

The studies looked at a range of different interventions, which the reviewers categorised as dietary/nutritional, food retail, physical activity and multi-component.

As the reviewers conducted a narrative synthesis, there was no attempt at meta-analysis.  Some highlights from the better quality studies were:

  • Behavioural counselling was more effective than nutritional counselling at increasing consumption of fruit and vegetables
  • Neighbourhood physical activity promotion led to sustained improvements in activity levels, but did not affect fitness measures
  • Mediterranean diet interventions led to a reduction in intake of high fat foods.

Many of the studies found no benefit, however.

15 qualitative studies were found that identified barriers and facilitators for people with low SES.  These included:

  • Resources, or lack therof
  • Lack of awareness or information about interventions
  • People often felt bombarded by confusing and contradictory information from different sources
  • Acceptability of interventions (attributes of the health workers, content and delivery of the intervention)
  • Existing attitudes of participants and their perceptions of their own capabilities.

Acceptability was increased where health workers had knowledge of the local community, where delivery included practical demonstrations and progressive small steps to change, and where single sex groups were used.  Social interaction and peer or family support played an important part in some successful interventions.


  • As the study did not attempt any meta-analysis, there is a danger of “cherry picking” the positive outcomes from the range of possible data present.
  • The individual studies themselves were of variable quality.  We need to look more closely at the details of the interventions.
  • The mixed methods analysis showed some indications of the types of activity are likely to be successful.  However, we should not read too much into this.  They should help design further prospective research, however.
  • The review was limited to a UK context.


The full text of this review is available from the journal Preventive Medicine: