ORIGINAL RESEARCH 1 SAJSM VOL. 30 NO. 1 2018 The association between being overweight/obese and blood pressure in rural South African women living in the Tshino Nesengani (Mukondeleli) village P J-L Gradidge, PhD1; M Phaswana, BSc (Hons)1; E Cohen, PhD2 1 Centre for Exercise Science and Sports Medicine, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa 2 MRC/Wits Developmental Pathways for Health Research Unit, Department of Paediatrics, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa Corresponding author: P J-L Gradidge (philippe.gradidge@wits.ac.za) The prevalence of both abdominal and whole body obesity is expected to increase dramatically in developing countries, such as South Africa, with recent data demonstrating that black South African women have the highest prevalence of obesity within sub-Saharan Africa.[1] South African women living in urban settings certainly have higher rates of obesity compared with men and rural women;[2] however, the prevalence of excess adiposity in women living in both settings is higher than the global averages.[1] A consequence is that the risk of associated cardiometabolic diseases, such as type 2 diabetes and hypertension, is high in African women.[3] The obesity crisis in South Africa is made more complex by the dynamic rural-urban shift, which may include an extension of the urbanisation concept into the rural setting as observed in other countries experiencing nutritional transition.[4] This phenomenon may explain the high prevalence of excess adiposity in both settings as learned obesogenic practices such as sedentariness and unhealthy eating may be adopted by rural communities. Research indicates that the global consumption of sugar- sweetened beverages (SSB) has increased and is associated with weight gain, elevated blood pressure (BP) and other cardiometabolic diseases risk factors.[5] Despite black South African females having the highest prevalence of obesity in the sub-Saharan region, the results of the South African Demographic and Health Survey show that other population groups have higher tobacco and alcohol consumption patterns.[6] This suggests that there may be other environmental factors driving the high prevalence of obesity and cardiometabolic diseases in African women. These other environmental factors have received far less attention than diet or physical activity. They include smoking and sleep behaviour. Smoking is well known as an independent risk factor for several cardiovascular diseases, while sleep duration has a negative association with BMI,[7] although the mechanisms underlying this effect are largely unknown. The aim of this study was therefore to determine whether other environmental factors (e.g. physical activity, sedentary behaviour, consumption of SSBs, smoking, smokeless tobacco consumption or ‘snuff’, and sleep duration) correlate independently or with the measurements of fat and BP in a cohort of rural African women. Methods Sample This was a cross-sectional study of a convenience sample of rural black South African women living in the Tshino Nesengani (Mukondeleli) village, Limpopo Province, South Africa. Potential participants who were pregnant, aged <18 years, non-Black, and living outside of the village were excluded from the study. Ethical approval was granted by the Human Research Ethics Committee (Medical), University of the Witwatersrand (ethics certificate number: M170377), and all participants gave written consent. The questionnaires were administered to the participants and the details communicated in the participants’ home language when necessary to ensure correct completion of the questionnaires. Physical activity The Global Physical Activity Questionnaire (GPAQ) was used to determine self-reported total moderate-vigorous physical activity (MVPA) and the estimated sitting time. The GPAQ is reliable and has been validated for use in Africa.[8] Active in the GPAQ was defined as taking part in: moderate physical activity for a total of 150 minutes per week (≥five days per week); or vigorous physical activity for 60 minutes per week (≥three days per week); or 600 metabolic minutes per week (≥five days MVPA).[8] In addition, walking for travel purposes, as a domain of light physical activity, was determined using the GPAQ. Background: The purpose of this cross-sectional study was to investigate whether bio-behavioural factors are associated with blood pressure and body composition in rural black South African women. Methods: Data were collected on 200 African women living in the Tshino Nesengani (Mukondeleli) village, Limpopo Province, using simple anthropometry, blood pressure, and self-reported questionnaires for sleep, physical activity, and sugar-sweetened beverage (SSB) consumption. Results: Six patterns of SSB consumption were determined by principal component analysis. Regression analysis showed that longer sleep duration patterns (≥nine hours/night) was associated with lower systolic and diastolic blood pressure; whilst the principal components (beer, wine, and sweetened tea) were associated with a higher body mass index. Conclusion: These findings highlight novel bio-behavioural contributors of blood pressure and body anthropometry in rural African women. Keywords: African, BMI, waist circumference, sugar- sweetened beverages S Afr J Sports Med 2018; 30:1-5. DOI: 10.17159/2078-516X/2018/v30i1a5066 http://dx.doi.org/10.17159/2078-516X/2018/v30i1a5066 ORIGINAL RESEARCH SAJSM VOL. 30 NO. 1 2018 2 Beverage intake The beverage intake questionnaire (BEVQ-15) is a 15-item, seven day recall on SSB used to measure the quantities and amounts of habitual beverage consumption.[9] The BEVQ-15 includes 15 categorised beverage items to estimate total kilocalories (kcal) of consumed beverages: water, regular soft drinks, 100% fruit juice, juice drinks (other than fruit juice), full cream milk, low-fat milk, skim (fat-free) milk, sweetened tea, coffee or tea with milk and sugar, black coffee or tea without sugar, light beer, regular beer, mixed alcoholic drinks, wine (red or white), meal replacement drinks and energy drinks. The total of SSB calorie consumption is calculated from the estimated energy consumption for each item. Participants were asked to recall the amount and frequency of each item. Sleep patterns The Pittsburgh Sleep Questionnaire Index (PSQI) was used to determine the overall quality of sleep.[10] The PSQI is a self- reported questionnaire consisting of seven components of sleep that evaluate sleeping duration, sleep disturbance, sleep latency, habitual sleep efficiency, daytime dysfunction, use of sleeping medicine, and sleeping quality. The total score ranging from 0-21 is summed from these items. A total score >five indicates poor sleep quality for the global PSQI index, and a score of ≤five indicates good sleep quality. Socioeconomic status and education Household asset ownership was used as a proxy measure of socioeconomic status (SES).[11] The questionnaire included eleven household items, ranked in order of value from lowest to highest: (1) radio, (2) computer/laptop, (3) refrigerator, (4) washing machine, (5) television (TV), (6) telephone/landline, (7) cell phone/mobile, (8) internet, (9), electricity, (10) digital satellite TV, and motor vehicle. The score of these commodities was summed to give a total SES index ranging from 0 to 66. Tertiles of SES score were created for further analysis: low (SES score: <29), moderate (SES score: 29-36), and high SES (SES score ≥36) categories. The levels of education were captured as follows: 0 for no schooling, 1 for primary school, 2 for incomplete high school, and 3 for completion of high school. Participants were also queried on their employment status. Anthropometry All measurements were performed with participants in light clothing and without shoes. Body weight was measured using an electronic digital weighing scale to the nearest 0.1 kg (Seca, USA). Height was measured to the nearest 0.1 cm using a stadiometer (Seca, USA). Body mass index (BMI) was calculated as weight (kg)/height (m2). Waist and hip circumferences were measured using a spring tape to the nearest cm. Waist circumference was measured between the lowest ribs and the iliac crest, and hip circumference was measured at the greatest protuberance just below the gluteal line. Blood pressure Resting BP was measured using a digital BP monitor (Omron M6 version HEM-7001-E, Omron, Kyoto, Japan). Participants were seated for a minimum of five minutes before the first BP measurement was done. Two subsequent measurements were taken with rest periods in between. The average of these latter two readings was used to determine mean resting BP. Statistical analysis Descriptive statistics were presented as mean ± standard deviation, median (interquartile range), or percent in tables. A principal component analysis (PCA) was performed with all BEVQ-15 items except water and diet soft drinks (due to zero and low caloric values) to determine patterns of SSB consumption. The following steps were followed: (1) the covariance matrix was applied, (2) the variance loadings were rotated using the varimax with the Kaiser normalisation orthogonal method. These authors could perform a PCA on the covariance matrix of SSB variables because the Bartlett’s test of sphericity was good (p<0.0001) and Kaiser-Meyer-Olkin measure was acceptable (0.524). Six principal components (PC) on SSB matrix consumption were initially retained based on eigenvalues ≥1 and the scree plot observation. PC1 (hot tea/coffee with added sugar and energy drinks) explained 15.8% of the variance, while PC2 (beer and iced tea) and PC3 (low-fat milk and spirits) explained 12.1% and 10.9% of the variance respectively. The associated eigenvalues were 2.05, 1.58 and 1.41, respectively. The first four components also had the greatest factor loading with correlations ≥0.30 in the same direction, resulting in PC5 and PC6 being removed from an additional analysis. Bivariate models were created to determine the association of plausible lifestyle behaviours variables with BMI and waist circumference. Those variables with p<0.20 were included in the initial backward stepwise multivariable linear regressions models. Thus the independent variables for the BMI model included age, employment status, hypertension, short sleep (