Table 2.
Summary of measurements undertaken within WAVES study and their associated outcome variables
| Measurements | Time points | Instrument | No of measures at each time point | Method of assessment | Outcome variables* | |||
|---|---|---|---|---|---|---|---|---|
| Baseline | 1st follow-up | 2nd follow-up | 3rd follow-up | |||||
| Weight | Yes | Yes | Yes | Yes | Tanita bioimpedance monitor (Tanita SC-331S; Tanita, Tokyo, Japan) | Once | Barefoot and in light clothing | Body mass index (BMI) z score Overweight or obese (BMI ≥85th centile or ≥95th centile (both using UK 1990 BMI reference curves for children33) |
| Height | Yes | Yes | Yes | Yes | Leicester height measure | Twice (third measure if difference >0.4 cm)† | Barefoot and in light clothing | |
| Demographic data (sex and date of birth) | Yes | No | No | No | Parent questionnaires | NA | Parent report and school records | |
| Body fat % | Yes | Yes | Yes | Yes | Tanita bioimpedance monitor (Tanita SC-331S; Tanita, Tokyo, Japan) | Once | Barefoot and in light clothing using two limb (legs) bioelectrical impedance technology | Body fat % |
| Waist circumference (to nearest 0.1 cm) | Yes | Yes | Yes | Yes | Flexible, non-stretch, cloth tape measure | Twice (third measure if difference >0.4 cm)† | Measured at iliac crest | Waist circumference z score using UK 1990 BMI reference curves for children33 |
| Skinfold thickness (biceps, triceps, subscapular, suprailiac, and thigh) | Yes | Yes | Yes | Yes | Holtain Tanner/Whitehouse Skinfold Caliper (Holtain, UK) | Twice (third measure if difference >0.4 mm)‡ | Measured on non-dominant side | Sum of four skinfolds§ (biceps, triceps, suprailiac, and subscapular) |
| Dietary intake | Yes | Yes | Yes | Yes | Child And Diet Evaluation Tool (CADET) (a validated 115 item 24 hour food tick list22 completed for seven distinct time periods) | Once (24 hours) | Completed by trained researchers in school, and parent/carer at home (with instructional DVD) | Dietary daily total energy intake (kJ in 24 hours), fat, sugar, fibre (g/day), and fruit and vegetable intake (g/day and portions) |
| Physical activity | Yes | Yes | Yes | No | Actiheart (Cambridge Neurotechnology, Papworth, UK) | Once (worn continuously for five days, including a weekend) | Fitted in school by trained researcher | Daily physical activity energy expenditure (kJ/kg body weight/day)¶, and time spent being sedentary and undertaking at least moderate intensity activity (min/24 hours) assessed by Actiheart |
| Blood pressure | Yes | Yes | Yes | Yes | Automated oscillometric monitor (BpTRU BPM-100, British Columbia, Canada) | Twice (third measure if error reading, or if one value outside normal range)** | Three minutes seated rest before and between readings | Systolic and diastolic blood pressure |
| Quality of life | Yes | Yes | Yes | Yes | Pediatric quality of life inventory (PedsQL) | NA | Researcher administered questionnaire | Self reported health related quality of life |
| Social acceptance | Yes | Yes | Yes | Yes | Kidscreen-52 health questionnaire for children and young people | NA | Researcher administered questionnaire | Social acceptance |
| Body image dissatisfaction | Yes | Yes | Yes | Yes | Child’s body image scale (CBIS) | NA | Researcher administered questionnaire (score derived from sex specific 7 point child’s body image scale) | Body image dissatisfaction |
| Demographic data (date of birth, sex, ethnicity, postcode (proxy measure for deprivation)) | Yes | No | No | No | Parent questionnaires | NA | Parent report and school records | Sex, ethnicity, deprivation (index of multiple deprivation) |
Measurements were carried out by trained research staff using standard protocols.
Primary outcomes (from trial protocol)=difference in BMI z scores between arms at first follow-up and second follow-up, all other outcomes are trial protocol secondary outcomes but some of the measures of obesity (proportion of children overweight/obese between arms, waist circumference z score, sum of four skinfolds, and body fat %) were included as primary outcomes in the trial registration.
Where two values were within ≤0.4 cm, a definitive measurement value was calculated as the average of the two. For individuals with three values recorded, a definitive measurement value was calculated as the average of the closest pair (within ≤0.4 cm) or average of all three readings (if there were no two closest readings, but the differences between values were ≤0.4 cm). When none of the three values were within 0.4 cm of each other no definitive measurement value was calculated.
Where two values were within ≤0.4 mm, a definitive measurement value was calculated as the average of the two. For individuals with three values recorded, a definitive measurement value was calculated as the average of the closest pair (within ≤0.4 mm) or average of all three readings (if there were no two closest readings, but the differences between values were ≤0.4 mm). When none of the three values were within 0.4 cm of each other no definitive measurement value was calculated.
Skinfold thickness was measured at five different sites (biceps, triceps, thigh, suprailiac, and subscapular), as detailed in the protocol. Compared with the other sites, however, the children found the measurement of thigh skinfold thickness more intrusive, resulting in a lower level of data availability for this compared with the other sites. The skinfold thickness summary measure was therefore calculated excluding the thigh measurement.
Children with less than 24 hours of valid data were excluded. In addition, to ensure representation across the whole 24 hour period, for those with 24 hours of valid data, only those with a distribution of at least six hours in each quadrant of the day (morning; 3 am-9 am, noon; 9 am-3 pm, afternoon; 3 pm-9 pm, and midnight; 9 pm-3 am) were included.
Readings with a systolic or diastolic, or both, value 20 mm hg above the 99.6th centile of the UK age specific and sex specific reference data34 were excluded as implausible values. Pairs of readings for which an error message was returned for either the systolic or diastolic value were excluded. Subsequent to these exclusions, systolic and diastolic values that remained were treated independently. For individuals with only one systolic or diastolic value this was taken as the definitive measurement value. For individuals with two remaining systolic or diastolic values the definitive measurement value was taken as the average of the two values. For individuals with three remaining systolic or diastolic values, provided there was a closest pair of values, the definitive measurement value was taken as the average of these two values, or, in instances of no closest pair, the average of all three values.