LAUSR

Background Obesity threatens to undo the improvements that have been made in life expectancy over the last two centuries. It disproportionately affects lower socioeconomic and ethnic minority groups and has become one of the most important global health challenges of the 21stcentury. Whilst obesity is not confined to city populations, cities are home to more than half of the world's population with concentrated groups at high risk of obesity. Cities have also long been the forefront of social and technological change that has led to our current obesogenic environment. The aim of this study was to systematically identify city-wide interventions to address obesity, from which recommendations for policy makers, health system leaders and political leaders in cities could be made. Methods Systematic review, conducted according to PRISMA guidelines, examining Embase, Ovid Medline, Central, Scopus, Campbell Library, CINALH, Health Business Elite; Health Management Information Consortium (HMIC), PyschINFO and Prospero. No restrictions on article type, date range or geographic location were applied. Along with classic academic sources, books and policy white papers were sought and reviewed. Studies that described a city-wide intervention to reduce obesity were included, irrespective of study design or perceived methodological quality. Only studies in English language were included. The primary outcome indicators that were sought and extracted were: reduction in obesity, reduction in weight and/or reduction in BMI. Where a primary outcome indicator was not stated, any other secondary impact measure was identified and recorded. This manuscript represents thematic analysis of a sub-set of data from the Prospero study, registration number: CRD42020166210 Findings Our search yielded 42,137 original citations of which 1614 met the inclusion criteria and 96 were coded as relating to obesity. The 96 citations, ranging in year of publication 1997 to 2019, were conducted in 36 cities, with 13 citations either not stating a city or covering multiple cities, across 5 continents. The highest proportion of publications were from North America (59 / 96) and in particular the USA (56/96) and New York City (23/96). Primary outcome indicators were only stated in one quarter of the identified studies (24/96). Overall, there was heterogeneity of study design, descriptive methodologies and publication types, with a majority being descriptive texts using qualitative instruments of assessment. Interpretation Multi-level and multi-component interventions, at the individual, community and city level, done in concert, are needed to address obesity. A composite of interventions that cities can utilise to address obesity is provided. These interventions will also be beneficial to the environment and make the case that personal health and planetary health are inextricably linked and should be considered as one. Funding None