Background Non-communicable diseases (NCDs) cause over two thirds of all global deaths. Poor diet generates more NCD and disability than tobacco, alcohol and physical inactivity combined. Worldwide, many different policy options have been proposed to promote healthier eating and prevent diet related NCDs. However, these options are diverse and potentially confusing. A key policy question is thus: how can diets best be improved to prevent NCDs nationally and globally? We reviewed the potential effectiveness of policy actions to improve healthy eating and thereby prevent NCDs.

Methods We searched six electronic databases (CDSR, CRD, MEDLINE, SCI, SCOPUS and the Campbell Library) using a pre-piloted search strategy. We focussed on systematic and non-systematic reviews dating from January 2004 onwards, containing quantitative outcomes. Retrieved studies were screened independently by two researchers. A PICOS approach (Participants, Interventions, Comparators, Outcomes and Study design) was used to assess the retrieved studies that fulfilled the criteria. Extracted data were categorised using an updated version of the ‘4Ps’ marketing framework: Price; Promotion; Provision; Product (adding Supply Chain, Trade and Investment; and Multi-component interventions).

Results After screening 1800 candidate publications, 57 systematic and non-systematic reviews were identified for inclusion in the review. Eight reviews provided information on more than one category. Thus, 19 papers provided data on food price interventions; 10 food promotion, 10 food provision, 2 food composition, 6 food labelling, 6 food supply chain, trade and investment, and 4 for multi-component interventions.

Price interventions appeared to be consistently very effective in improving healthy eating, both taxing unhealthy foods and subsidising healthy options such as fruit and vegetables. Product interventions, especially elimination of trans-fat and salt reduction, and comprehensive strategies involving multi-component interventions also proved powerfully effective in improving healthy eating and preventing NCDs.

Food Supply Chain and Trade and Investment studies reporting potential negative health effects of market liberalisation and globalisation appear potentially powerful and merit further research.

Other interventions, particularly reformulation, labelling, restrictions on the provision or marketing of unhealthy food also tended to improve healthy eating and prevent NCDs. However, their effectiveness was often limited and their sustainability was rarely studied.

Conclusion Increasingly strong evidence is now highlighting potentially powerful policies to improve diet and prevent NCDs, especially taxes, subsidies, reformulation, multi-component interventions and possibly also trade agreements affecting food supply chain. The other interventions appear to be, at best, modestly effective in the short term.