Study population and recruitment strategy
Participants were self-selected from two regional locations in Shepparton, northern Victoria, and Colac, south-west Victoria (estimated resident population 40–70 years at 30 June 2015 was 23,437 and 13,055, respectively). Census data were used to ensure that a representative sample of the age and sex distribution of each regional community between 40 and 70 years were recruited. The sampling strategy involved large scale promotion of the study to key local workplaces, community groups (e.g. Rotary) and Local Members of Parliament, community engagement at local markets and stalls, direct and indirect postal invitations, radio and newspaper advertising and social media (e.g. Facebook). Individuals responded to participate and a brief screening questionnaire administered via telephone or face-to-face was initially used to determine age and history of CVD or other chronic disease. Eligible participants made an appointment to undergo screening to identify MetS from 26 September 2014 and recruitment was completed on 1 April 2016. The study protocol conformed to SPIRIT guidelines and a full CONSORT diagram will be finalised upon trial completion.
Procedures
A self-report questionnaire posted to participants in the week prior to their scheduled visit collected information regarding:
socio-demographic factors including marital and work status, primary language spoken, income, education and ethnicity;
health behaviours including smoking, diet and alcohol via the Dietary Questionnaire for Epidemiological Studies (DQES v2) [
19], sleep and physical activity and sedentary behaviour via the International Physical Activity Questionnaire (IPAQ) [
20] and;
general health including history of CVD, associated conditions or other serious conditions, family history of CVD or diabetes, prescribed medications and adherence via the Medication Adherence Questionnaire (MAQ) [
21], signs and symptoms of angina via the Rose Angina questionnaire [
22], health related quality of life via the Assessment of Quality of Life – 8 Dimension (AQoL-8D) [
23], perceived risk and health belief evaluation via item specific self-efficacy [
24] and locus of control [
25].
An on-line, discrete choice experiment during the week prior to their scheduled visit at baseline and at study end evaluated participant preferences for the attributes of a health and lifestyle management program they consider of greatest importance affecting their choice to participate in such programs.
The screening assessments were undertaken at a dedicated clinic or on-site at workplaces (for few organisations with large numbers of workers) by registered nurses according to a standardised protocol at baseline, prior to randomisation. The same registered nurse who conducted the screening assessment remained responsible for the follow-up assessments of each participant until study completion; due to standardising the management of participants and frequency of follow-up, a different nurse that managed participants from only one study group was not required to alleviate bias.
Waist and hip circumference were measured using a Figure Finder® Tape Measure (Novel Products, IL, USA) in accord with the World Health Organization (WHO) STEPwise approach to surveillance (STEPS) procedure [
26] in the horizontal plane whilst standing; the level mid-way between the lowest rib and iliac crest at the end of a gentle expiration was taken for waist circumference and the level at the maximum extension of the buttocks defined hip circumference. Body mass index (BMI, kg/m
2) was determined from height, measured using a portable stadiometer (seca®, Hamburg, Germany) and weight, using digital weighing scales (A&D Medical, SA, Aus), after the removal of shoes and heavy garments. Total cholesterol (TC), high- (HDL-C) and low- (LDL-C) density lipoprotein cholesterol, triglycerides and HbA
1C in capillary/fingerstick whole blood were analysed by the reflectance photometry technique using a calibrated Afinion™ AS100 analyser (Alere, MA, USA). Sitting blood pressure (BP) was measured in the brachial artery with a suitable cuff size using a calibrated Omron HEM-907 automated monitor (Omron Healthcare Co. Ltd., Kyoto, Japan) after 5 min of rest; the average of two measurements separated by a one-minute interval were analysed provided BP did not vary by ≥ 10/≥ 5 mmHg, in which case another reading was taken and the closest two readings were analysed. Calculation of CVD risk via the Framingham Risk Equation [
27], and T2DM risk via the AUSDRISK tool [
28], was estimated from applicable health and lifestyle risk factors and results from the thorough screening assessment to predict risk of a cardiovascular event [
29] and developing T2DM [
28], respectively, over the next 5 years.
If MetS was indicated and subjects agreed to be randomized into a study group, additional information was collected including: electrocardiography (ECG) via a 12-lead Universal ECG™ in accordance with standard electrode placement and recorded using Office Medic™ Software (QRS Diagnostic, MN, USA); biochemistry assessment from venous blood of creatinine to estimate glomerular filtration rate (eGFR) using the Modification of Diet in Renal Disease study equation, and high-sensitive C-reactive protein (hs-CRP) as an inflammatory marker of CVD risk; actigraphy for 7-day continuous quantitative measurement of sleep, physical activity and sedentary behaviour using GTX3 accelerometers (ActiGraph, FL, USA) and; arterial stiffness from ankle brachial (ABI) and cardio ankle vascular pressure indexes (CAVI) using the VaSera™ VS-1500 N vascular device (Fukuda Denshi Co Ltd., Tokyo, Japan).
Qualitative evaluation of intervention fidelity [
30] to assess 1) adherence to key aspects of the protocol being implemented, and 2) competence in delivering the program in terms of communication, technical abilities and skills in responding to participants was established by video recordings of nurse/participant clinic visit interactions that were later evaluated by two trained observers, with feedback of results to improve nurse’s performance. For selection of a random set of observations to provide a representative sample for measuring intervention fidelity, nurses were advised (3–5 days prior) by the study co-ordinator that all participants scheduled to attend clinic visits in the following week will be asked to provide written informed consent to video recording their clinic encounter. Fidelity monitoring continued until the target number of 2 observations from each of the two trial arms (4 in total) per nurse at baseline and mid (12 month) study visits was completed.