Details on the participants’ baseline characteristics, study site, design, and procedures are reported in our published work [31]. Briefly, this was a randomized control trial design carried out in adults with MetS. The participants were recruited from an outpatient clinic of faith-based hospital in Nairobi, while follow-up took place in the community for 15-months. MetS was determined using the International Diabetes Federation (IDF) diagnostic criteria [32]. Having MetS was considered, if a person had central obesity (waist circumference of ≥ 94 cm for men and ≥ 80 cm women), and at least 2 of the following CVD risk factors (reduced HDL-C, high BP, raised TGs and high FBG). Two research assistants involved in the recruitment of the study subjects, while assignment into the groups was carried out by the primary investigator. The research assistants as well as the participants were totally blinded during the randomization. The physical, clinical and biochemical values were measured by the research assistants who were not aware of the groups. To minimize contamination, subjects from the same family and who live together were assigned to the same group.
Eligible individuals were assigned to either lifestyle intervention group (n = 176) or usual care group (n = 176) using a block stratified randomization technique. Details of the randomization have been reported in an earlier article [31]. Those in the intervention arm received world Health Organization (2011-a) recommended lifestyle modification intervention comprised of physical activity, dietary counselling and behavioral intervention. While the control group received usual clinic care involved lab investigations, drug treatment and general health advice. The study was designed into 3 phases namely: baseline, intervention phase and evaluation phase (Fig. 1). At each phase, anthropometric, clinical and biochemical parameters were measured. Using chi-square test, categorical data were compared between the groups before and after the intervention. Continuous data were compared between and within the groups using independent t-test and paired t-test, respectively. The primary outcome measures included change in CVD risk factors: Body weight, BMI, WC, WHR, WHtR, FBG, HDL-C, TGs, systolic and diastolic BP.
2.1 Study Intervention
The lifestyle intervention involved health education and awareness creation on healthy lifestyles. It consisted three face-to-face sessions (at baseline, 9 and 15 months) of counselling for dietary intake patterns and physical activity. Between the visits, additional counseling through telephone calls was provided.The intervention was administered by a registered nurse who had over 7 years work experience.
According to the WHO guidelines (WHO, 2011-a), they were educated about the recommended intake of sugar, salt, fruits, vegetables, legumes and nuts. They were advised to restrict salt intake to less than 1 teaspoon (5grams) and sugar intake to less than 5 teaspoons (25 grams) per day as well as to avoid or limit processed/fast foods and saturated fats intake [33]. They received specific information on the recommended standard drinks and frequency of alcohol consumption as well as the negative effect of excessive drinking on CVDs. They were also educated to involve in adequate level of physical activity (≥ 150 minutes of moderate or ≥ 75 minutes of vigorous-intensity physical activity in a week) [9]. The Dietary Approaches to Stop Hypertension [34] diet was demonstrated using a plate indicating the recommended rations of vegetables/fruits proteins, and carbohydrate intake. Furthermore, the health education intervention was accompanied with the Kenya diabetes and hypertension prevention flip charts [35]. Dietary and physical activity goals were set for each participant according to their specific needs. At each visit, the planned goals were reviewed and barriers to follow the recommended lifestyle practice were discussed.
The control group received routine medical care as per the hospital’s protocol. This includes laboratory investigations, treatments with drugs and discharge general health education. For example, if an individual is diagnosed with diabetes, the person receives anti-diabetic medication and general lifestyle advice. No specific health education intervention is provided on the recommended dietary intake patterns (salt, sugar, fat, fruits, vegetables ect..), alcohol consumption and physical activity. Individuals who were not aware their health status, but confirmed to have diabetes (FBG ≥ 7.0 mmol/L) and/or hypertension (BP ≥ 140/90 mmHg) were informed to visit the diabetic/hypertension clinic of further evaluations and follow up.
2.3 Assessment of behavioural characteristics
Details of the assessment of behavioural characteristics including dietary intake patterns, physical activity, alcohol intake and cigarette smoking as well as validity and reliability of the study tools have been reported in our previous published work [31].
2.4 Assessment of cardio-metabolic risk factors
Cardio-metabolic risk factors including body weight, waist circumference, blood pressure, fasting blood glucose, HDL-C and TGs, and BP were obtained at baseline, midline and end-line. Data regarding physical and clinical parameters were measured by trained nurse research assistants. Biochemical measurement including fasting blood glucose, triglycerides and high density lipoprotein cholesterol were determined by the hospital’s laboratory technicians who were blinded to the groups.
2.4.1 Anthropometric measurements
Anthropometric parameters including height, weight, waist circumference (WC), and hip circumference (HC) were measured and evaluated before the intervention and at 9 and 15-months after the intervention. The weight was measured using a Sohenle mechanical weighing scale in light clothes. This was done to the nearest 0.1 kg. The participant's height in meters was measured using a standard stadiometer, while they were standing upright on a flat surface without shoes. BMI was obtained as a ratio of body weight (kgs) to height in meters squared. Measurement of the WC was taken at the midpoint between the lower margin of the last palpable rib and the top of the iliac crest (hip bone) [36]. Using a flexible tape measure, the participants’ HC was measured at the largest circumference of the buttocks [37]. Waist/hip ratio (WHR) was determined as the ratio of WC to HC.
2.4.2 Blood pressure (BP) measurements
The respondents were requested to a private room and allowed to be seated for approximately 5 minutes to relieve from anxiety. While the person was sitting, the BP was measured on the right hand using an automatic blood pressure measuring tool (OMRON: Model: M3; HEM-141-E, Serial. No: 20170916247VG, Japan). The BP was measured twice with an interval of at least five minutes and the mean was recorded. Measurement of systolic BP ≥ 140 mmHg and/or diastolic BP ≥ 90 mmHg was considered as hypertension. Subjects whose BP values below 140/90 mmHg, but reported taking antihypertensive drugs for at least two weeks before the baseline survey were considered as having hypertension [38].
2.4.3 Biochemical measurements
A sample of 3 milliliters of blood was collected (8-hours fasting) for determination of HDL-Cholesterol and TGs. After the sample was taken from the brachial vein, it was spun for 5 minutes to get enough serum. Then, the serum was analyzed using the Cobas Integra technique. As an element of metabolic syndrome, raised TGs was considered as ≥ 1.7 mmol/L regardless of gender. While reduced HDL-Cholesterol was considered as < 1.29 mmol/L and < 1.03 mmol/L in females and in males, respectively [32]. The FBG concentration was taken from the participants’ fingers and analyzed using HemoCue® B-Glucose photometer (1995). As an element of metabolic syndrome, elevated FBG level was set at ≥ 5.6 mmol/L [32]. Whereas, FBG levels of 5.6–6.9 mmol/L and ≥ 7mmol/L were considered as pre-diabetes and diabetes, respectively [39].
2.4 Outcome measures
The primary outcomes were changes in eating and physical activity behaviors, cardio-metabolic risk factors including body weight, BMI, WC, WHR, WHtR, systolic and diastolic BP, blood glucose and lipid levels.
2.5 Data analysis
The IBM SPSS software (ver: 22) was used to analyze the data. Frequencies and proportions were obtained for categorical variables, whereas, means and standard deviations were expressed for continuous variables. The chi-square test was used to establish any significant differences in socio-demographics between the two groups. For continuous variables, the paired t-test was used to determine within-group changes of CVD risk factors between baseline and follow-up. While, the between-group difference in these variables was performed using the t-test for independent samples. The impact of intervention was evaluated as the average difference in cardio-metabolic risk factors between the two groups. A p-value of < 0.05 was considered statistically significant.