2.3. Baseline Covariates

Physical activity (PA) was quantified using the short-form International Physical Activity Questionnaire (IPAQ) [25]. Metabolic Equivalent Task (MET)-minutes of PA/week was calculated by multiplying the MET value of activity by the number of minutes/week. PA was then classified as inactive (<600 MET min/week), active at the lower PA guideline (≥600 MET min/week), or active at the upper PA guideline range (≥1200 MET min/week). We adjusted for chronic illness using a dichotomous variable denoting the presence/absence of major cardiovascular disease (ICD-10 codes I00 to I99) or cancer (C00 to C97, excluding ill-defined, secondary, or unspecified neoplasms). Because using dietary energy intake as a proxy for dietary quality would result in a substantially smaller sample size, we used fruit and vegetable consumption (servings/day) as a proxy for dietary quality. Participants were asked to report the number of servings of cooked vegetables, salad and raw vegetables, fresh fruit, and dried fruit they consumed each day. For example, “On average how many heaped tablespoons of salad or raw vegetables would you eat per day?” One piece of fruit, such as a banana, or one heaped tablespoon of vegetables was considered one serving. Sleep was dichotomized based on participants’ responses to the question “About how many hours sleep do you get in every 24 h? (including naps)”, with participants whose sleep duration was within 7–9 h/day being classified as “adequate”. Sedentary behaviour was calculated by summing the total time spent watching television, using a computer screen or driving. Participants with an implausible sum (>24 h/day) of sedentary time, sleep, and PA were excluded from analysis. Smoking status comprised three categories: never, previous, and current smokers. We used the Townsend deprivation index as an indicator of socioeconomic status, which assigns each participant a score relative to the output area (the smallest UK census area) in which their postcode was located, with higher scores indicating greater socioeconomic deprivation [26].