The association between different types of food outlets and obesity among adults living in UK

Abstract

It has been suggested that neighbourhood characteristics, such as access to unhealthy food establishments, could be related to the development of obesity. However, at the individual level, this may not reflect whether people living near restaurants eat more frequently at these outlets. Moreover, there was a lack of studies in United Kingdom (UK) that examined the extent to which energy intake from different types of food establishments contributes to total energy needs and whether these contributions were related to obesity among adults. This thesis aimed to investigate the association between obesity and frequency of use of different types of meal-serving food outlets (aside from supermarkets and groceries) at both the population and individual levels, and to see to what extent energy intake from different types of restaurants (Fast-Food Restaurants (FFRs) and Full-Service Restaurants (FSRs)) and food outlet services (delivery and takeaways) contributed to total energy requirements in UK adults. Also, I examined the impact of Coronavirus-19 (COVID-19) lockdown on BMI and the frequent use of different types of food outlets and food services (delivery and takeaways). Five studies were carried out. The first study (chapter 2) investigated the link between the densities of food outlets that serve meals, with body mass index (BMI), waist to hip ratio (WHR) and percentage of body fat among middle-aged adults using UK Biobank data. In this study, socioeconomic data, BMI, WHR and %fat of 456,079 individuals were obtained and averaged across 923 Postcode Districts (PDs). After adjusting the obesity measures for socioeconomic factors, the density of fish & chip shops per 1000 population per PD was significantly positively associated with obesity while FSRs, pubs, cafes and total food outlets were all negatively significantly associated with the obesity measures in males and females. The number of FFRs and delivery shops were not related to WHR and %fat, except BMI in males where the density of FFRs was significantly inversely associated. The second study (chapter 3) investigated the association between the consumption of meals bought from different sources including homemade meals and Body Mass Index (BMI) and %fat. After I adjusted the BMI and %fat in males and females for sociodemographic variables, I found that consuming takeaway and/or delivery meals at the individual level in middle-aged people was significantly positively related to obesity, and I also found that consuming meals prepared in restaurants and/or café were associated positively with adjusted BMI and adjusted %fat in males but not in females. The obesity measures were significantly inversely associated with home-cooked meals. The third (chapter 4) and fourth (chapter 5) studies investigated the association between the frequent weekly use of FFRs, FSRs, delivery, takeaways and Total Food Outlets combined (TFOs) and BMI among adults in Scotland before and during the lockdown that was implemented by UK government during the Coronavirus pandemic. Participants were contacted on 2 occasions: during the first lockdown between 26th March and 4th July 2020, and during the third lockdown between 29th March and 25th April 2021. The results showed that there was no association between the frequent use of the included food outlets and obesity before the pandemic in 2019 and during the first and third lockdown. However, during the third lockdown the BMI was significantly higher than the BMI before the Coronavirus pandemic. The fifth study (chapter 6) looked at the percent of contribution of energy intake from meals that were prepared at FFRs, FSRs, and restaurant services (delivery and takeaways) to total energy requirement in UK adults. I also investigated if these contributions were associated with obesity using seven-day food intake diaries, two-day 24hr recall surveys