G 112

Sociodemographic data included age, sex, and income. Low income was defined as having income in the lowest quartile among the entire Korean population. Comorbidities were included hypertension, diabetes, and dyslipidemia. Detailed definitions of comorbidities are presented in Supplementary Table S1. Body mass index (BMI) was defined as weight in kilograms divided by the square of height in meters (kg/m²). BMI as a categorical variable was defined as follow; underweight (< 18.5 kg/m²), normal weight (18.5 to < 23 kg/m²), overweight (23 to < 25 kg/m²), obese I (25 to < 30 kg/m²), and obese II (> 30 kg/m²)^{27 (link)}. History of smoking status, alcohol consumption, and exercise were obtained with a health checkup self-reported questionnaire. Smoking status was classified as a non-smoker, ex-smoker, and current smoker. A pack-year (PY) was used to quantify the amount of smoking. Alcohol consumption was categorized as none, mild alcohol consumption (< 30 g/day), and heavy alcohol consumption (≥ 30 g/day). The questionnaire form and the assumptions for calculating the amount of alcohol consumption based on a questionnaire are described in the Supplementary Methods^{15 (link)}. The frequency of alcohol intake per week (0 to 7 per week) and the amount of alcohol consumption per each drinking session (0 to ≤ 32 g, 32 to ≤ 56 g, 56 to ≤ 112 g, and > 112 g) were also collected^{15 (link)}. From the self-reported questionnaire in the national health examination provided by Korean NHIC, the intensity and frequency of physical activity were obtained. The structured questionnaire was based on the International Physical Activity Questionnaire (IPAQ) which was developed by the World Health Organization^{28 (link),29 (link)}. Both the reliability and the validity of the Korean version of IPAQ short form were proven^{30 (link),31} . Light intensity of exercise was defined as ≥ 30 min per day of walking slowly or sweeping carpets. Moderate physical activity was defined as ≥ 30 min per day of brisk walking, dancing, or gardening, and strenuous physical activity was defined as ≥ 20 min per day of running fast, cycling, or aerobic^{32 (link)}. The number of moderate or vigorous physical activities per week was also collected from the questionnaire. Regular exercise was defined as performing a moderate physical activity at least 5 times per week or strenuous physical activity at least 3 times per week^{33 (link)}.
To assess the impact of clustering of unhealthy lifestyle factors, we divided the cohort into 8 groups by the prespecified combinations of unhealthy lifestyle habits, including current smoking, heavy alcohol consumption, and lack of regular exercise, as follows: (1) subjects without any unhealthy lifestyle factors; (2) those with a single unhealthy lifestyle factor among the following: current smoking, heavy alcohol consumption, or lack of regular exercise; (3) those with combinations of 2 unhealthy lifestyle factors, including current smoking with heavy alcohol consumption, lack of regular exercise with current smoking, and lack of regular exercise with heavy alcohol consumption; and (4) subjects with all 3 unhealthy lifestyle factors.

Lifestyle factors were based on self-reported information, except for height and weight in one of the studies (Whitehall II). Smoking status included categories of current smoker and non-smoker. We defined high alcohol consumption as a weekly consumption exceeding 112 g of absolute alcohol for both men and women, according to 2016 guidelines in the UK.²⁴ Moderate alcohol consumption referred to a weekly consumption of more than 0 g and 112 g or fewer. Heavy episodic drinking was defined as a participant reporting having passed out at least once because of heavy drinking during the past 12 months (data available in FPS and HeSSup). BMI was categorised as lower than 18·5 kg/m² (underweight), 18·5–24·9 kg/m² (normal weight), 25·0–29·9 kg/m² (overweight), and 30·0 kg/m² or higher (obesity). Leisure-time physical activity was a dichotomous variable, including low physical activity versus intermediate and high physical activity. Details of lifestyle factors are provided in the appendix (p 2).
Sickness absence was measured as the number of diagnosis-specific sickness absence days per year derived from health registers in all cohorts and summed up for the follow-up period, which was either until death, old-age pension, or the end of follow-up of the respective studies. Details of sickness absences and covariates in each cohort are provided in the appendix (pp 3–5).

In total, 2 masterbatches of polymer blends, namely, PBS + 2% SLNPs and PBS + 20% thymol, were initially prepared. The first masterbatch (MB) was prepared by mixing 2 g SLNP powder and 98 g PBS pellets and then gradually loaded the mixture into a co-rotating twin screw extruder (LABTECH Twin-screw extruder; 20 mm screw diameter; length to diameter ratio of 32). The barrel temperature profile from feeding to the die zone was set at 110, 130, 140, 150, 154, 155, 155, and 160 °C with a screw speed of 0.112 g-force. The produced extrudate passed through a cooling water, was air-dried, and then palletized to around 2.5 mm long. The second MB was prepared using an internal mixer (HAAKE™, Rheomex OS, Bremen, Germany; 310 cm² capacity; 0.8 fill factor; roller rotor) to minimize the vaporization of the volatile thymol (vapor pressure of 53.33 Pa at 25 °C) and to produce good dispersion and homogeneity of the thymol in the PBS matrix [32 (link),33 (link)]. The rotor speed was set at 0.038 g-force and the temperature of the mixing chamber was set at 120 °C. An amount of 250 g PBS pellets were gradually fed into the internal mixer until melt phase then a 62 g thymol powder was introduced into the mixer and allowed to mix for 12 min. The produced bulk polymer mixture was crushed into pellets. Proportions from the two masterbatches were pre-mixed to produce the final formulations, as reported in Table 1.
The mixtures were formed into composite films by using a blown film, single-screw extruder (Thermo Scientific HAAKE™ Rheomex OS, Bremen, Germany; 19 mm screw diameter; length to diameter ratio of 25) with a screw speed set at 0.038 g-force and temperature profile of 150, 155, 160, and 160 °C for the feeding, compression, metering, and die zone, respectively. Neat PBS film which served as the control for analyses was also formed using the same blown film, single-screw extruder with the same setting.

A dilution ratio of 28 g MgP/L-water (7 g salt in 250 mL water) [25 (link)] was initially tested, subsequently applying a four-fold increase up to 112 g MgP/L-water (7 g salt in 62.5 mL water). Two different acids were tested to dissolve the MgP salts: citric acid (C) (4.5 N) and nitric acid (N) (5 N). Dissolution tests were carried out at an acid addition rate of 0.5 mL/min. A titration curve was plotted showing the evolution of the pH against the total amount of protons added. To verify the degree of dissolution of the MgP salts, the remaining total suspended solids (TSS) were measured once pH 3.0 was reached.