Background Inconsistent associations between neighbourhood out-of-home food environment and dietary outcomes have been reported. One reason may be that previous studies have primarily focused on the location and type of out-of-home food outlets, rather than the healthiness of options served. We investigated associations between different measures of the neighbourhood out-of-home food environment, incorporating menu healthiness, with frequency of out-of-home meal purchases and diet quality.

Methods We used Ordnance Survey Points of Interest data, a dataset containing all food outlets in Great Britain (GB), to define various measures of participants’ out-of-home food outlet exposure. We linked this with cross-sectional survey data in adults living in GB from the International Food Policy Study in 2021 (n=3,523).

Exposures included availability, proximity, and relative composition of out-of-home food outlets in a number of different neighbourhood buffers around the home (i.e., 500–1600 m), and incorporated healthiness scores (0–12, with 0 the least healthy) based on menu attributes such as variety of vegetables sold. Outcomes were the number of meals purchased out-of-home in the past 7 days and the Healthy Diet Indicator (HDI) derived from a 24hr dietary recall. In multiverse analyses, where multiple analytical choices can be tested, we used generalised linear regression models with survey weights to investigate the impact of different exposure on outcomes, controlling for multiple testing.

Results GB adults had access to an average of 98 (95CI% 91, 104) out-of-home food outlets within 1600 m of their home, with an average healthiness score of 6.7 (95%CI 6.6, 6.7). The number of out-of-home food outlets available, regardless of their healthiness, was positively associated with the number of meals purchased out-of-home across all neighbourhood definitions; e.g. for every additional 100 out-of-home outlets within 1600 m of home, there was an increase of 0.10 (95%CI 0.02, 0.17) out-of-home meals purchased per week. Proximity, relative composition, and menu healthiness of neighbourhood out-of-home outlets were not associated with out-of-home meal purchase frequency. None of the exposure measures were associated with HDI.

Conclusion The only aspect of the neighbourhood out-of-home food environment associated with the number of meals purchased out-of-home was the number of out-of-home food outlets. Menu healthiness of out-of-home food outlets was not associated with how often adults living in GB purchased out-of-home meals or diet quality. Interventions focusing on mitigating the proliferation of out-of-home food outlets may be more effective in promoting healthy dietary behaviour than those that focus on food served. Our findings are limited by reliance on self-reported dietary data.