According to the University of Bath Information Classification Framework, this project used highly restricted data and restricted data. We collected data on age, biological sex, height and weight (to calculate body mass index), ethnic origin, employment status, educational qualifications, marital status, and annual household income in the baseline questionnaire. These variables serve to describe the participant sample at an aggregate level in the associated journal article. Informed consent was obtained from participants for anonymised data to be preserved and reused for new purposes. Retained data will be preserved for at least 10 years from the end of the project or the publication date of any associated research findings. Measures already taken to address data protection and ethical issues in this dataset include changing indirect identifiers to another value or word that does not link back to anything and reducing the precision of some of the data (via recoding or categorisation): • Removed participants' date of birth (only their age in years at study entry remains) • Removed information on height and weight – only the calculated body mass index (in kg/m2) values remain • Redacted ethnicity to White versus Other • Redacted annual household income to Less than £25,000 versus Above or equal to £25,000 • Redacted employment status to Retired versus Other • Redacted education to University degree or higher versus Below degree • Redacted marital status to Married or living with a partner versus Other • No qualitative data are included in the dataset (the qualitative data underlying the associated journal article cannot be shared publicly for the privacy of individuals that participated in the study) • Removed any columns that were not necessary to replicate analyses presented in the associated journal article, including dates of data collection (these have been replaced with a Day_of_study variable, denoting the day number from 1 to 7) • New participant identifiers were created, and old ones destroyed to remove any linkage