Table 1 Description of measures

Wellbeing Outcomes

Psychological Distress. Kessler 10 (K10) [19]

10 items. Measure of psychological distress amongst adolescents, assessing symptoms over the past 4 weeks

The range for the K10 is 0–40. The summary score used is a total of all 10 items. A higher score indicates greater psychological distress

General Quality of Life.Youth Quality of Life Instrument-Short Form (YQoL-SF) [20].

15 items. Multidimensional tool that asses general quality of life of adolescents aged 11 to 18 years. Responses to each item are on a ten-point Likert scale

YQoL-SF is recoded to a 100-point scale, a higher score indicates better quality of life. The summary score used is a mean of the recoded items

Health-related Quality of Life.The Pediatric Quality of Life Inventory Physical function sub-scale (PedsQL-PF) [21].

8 items. A health-related quality of life scale which assesses physical ability and symptoms over preceding 30 days. Responses to each item are on a five-point Likert scale

The physical functioning sub-scale is recoded to a 100-point scale, a higher score indicates better quality of life. The summary score used is a mean of the recoded items

Risk Factors: All risk factors were dichotomised with 1 indicating risk definition met

Adiposity Domain

High Body Mass Index (BMI)

2 items. Mean height and weight used to calculate BMI (weight in kilograms divided by height in meters squared). Age-sex-adjusted BMI Z-score computed using World Health Organization (WHO) 2007 reference chart [22, 23].

Risk was defined as BMI z-score > 1, which corresponds to the + 1 SD defined as overweight by WHO [23].

High Waist Circumference

1 item. Mean waist circumference was calculated from the two nearest waist measurements

Risk was defined as waist circumference ≥ 90 cm for males; ≥ 80 cm for females [24].

Substance use Domain

Tobacco use. Adapted from Global Youth Tobacco Survey (GYTS) [25].

2 items. Questions relating to current cigarette smoking, and frequency of smoking

Risk was defined as smoking cigarettes at a frequency of weekly or more

Alcohol use. Adapted from YRBSS [26].

2 items. Questions relating to lifetime and current frequency of alcohol use in the past 30 days

Risk was defined as at least one drink of alcohol in the past 30 days

Physical inactivity Domain

Lack of daily moderate to vigorous physical activity. Adapted from HBSC [27, 28].

1 item. Question on frequency and duration of physical activity over the past week. Equivalent to WHO PA guideline for 11–17-year-olds [29].

Risk was defined if an individual did not get 60 min of moderate to vigorous physical activity daily

Lack of vigorous physical activity. Adapted from HBSC [27, 28].

1 item. Question on frequency of vigorous physical activity over the past month

Risk was defined if an individual did vigorous physical activity less than twice per week

Sedentary Domain

High sedentary activity. Adapted from HBSC [27, 28].

1 item. Questions relating to time spent watching TV (on average)

Risk was defined as watching TV more than 2 h per day

High sedentary activity. Adapted from HBSC [27, 28].

1 item. Questions relating to time spent using a laptop and playing video games (on average)

Risk was defined as gaming/ computer use more than 2 h per day

Unhealthy diet Domain

Insufficient fruit and vegetable intake. Adapted from HBSC [27, 28].

2 items. Questions relating to weekly consumption of fruits, and of vegetables

Risk was defined as fruits or vegetables consumed less than daily

Increased sweets or soft drink intake. Adapted from HBSC [27, 28]

2 items. Questions relating to weekly consumption of sweets, and of soft drinks

Risk was defined as sweets or soft drinks consumed at least daily

Risk Count

Co-occurring risk exposure: Count from 5 risk domains

A count of the risk domains that an individual had ≥ 1 risk/s present. For example, an individual with risk factors tobacco use, alcohol use, and high BMI would have a domain count of 2: substance use and adiposity

Range 0–5. A higher score indicates more co-occurring risk factors across multiple domains. The binary variable was defined as those with risk factors in ≥ 3 domains

Biomarkers:All risk factors were dichotomised with 1 indicating risk definition met. Biomarkers were selected, and risk cut points defined, based on the international diabetes federations’ criteria for metabolic syndrome in children and adolescents, and published in the study protocol [17, 24].

Risk Biomarkers

Raised triglycerides

1 item. Biomarker assays analysed, triglyceride level reported

Risk defined as triglyceride level of ≥ 1.7 mmol/l [24].

Reduced high-density lipoprotein (HDL) cholesterol

1 item. Biomarker assays analysed, HDL cholesterol level reported

Risk defined as HDL level of < 1.03 mmol/l in males, < 1.29 mmol/l in females [24].

Raised blood pressure

2 items. Mean systolic and mean diastolic blood pressure

16-17yrs risk defined as systolic or diastolic blood pressure ≥ 90th centile for age, sex, height [30]. 18yrs, risk defined as systolic ≥ 130 mmHg or diastolic ≥ 85 mmHg [31].

Raised glycated haemoglobin (HbA1c)

1 item. Biomarker assays analysed, HbA1c % reported

Risk defined as HbA1c ≥ 5.7% [24].

Biomarker Risk Count

Co-occurring biomarkers exposure: Count from 5 biomarkers

A count of 0–5 was assigned for the presence of high waist circumference, high triglycerides, low HDL, high blood pressure, and high blood sugar

Range 0–5. A higher score indicates the presence of biomarkers for risk. The binary variable was defined as those with ≥ 3 risk biomarkers