The relationship between physical activity and risk factors for non-communicable diseases of a population in transition : the PURE study
Van Niekerk, Tershia
MetadataShow full item record
Non-communicable diseases (NCDs), also known as chronic diseases of lifestyle, cause the greatest burden of disease globally. The major risk factors for NCDs are hypertension, hyperglycaemia, high cholesterol, tobacco smoking, alcohol abuse, overweight/obesity and physical inactivity. NCDs in South Africa are increasing in black South Africans with the transition from rural to urban areas. The transitions have resulted in a change in lifestyle. Regular moderate intensity physical activity (PA) has many health benefits and decreases the risk for NCDs. PA is often determined by means of questionnaires, motion sensors (pedometers and accelerometers), heart rate and accelerometry combined. Within the South African context PA has traditionally been determined with internationally composed questionnaires adapted for South Africa. In South Africa the relationship between PA and risk factors for NCDs has not been investigated in populations in transition, and limited information on the relationship between change in PA and the change in risk factors in a South African population is available. The objectives of this study was to determine the correlation between the adapted Baecke physical activity questionnaire and the International Physical Activity Questionnaire (short version) (IPAQ-S), the changes in PA and how the changes relate to changes in BMI, and finally the relationship between the changes in PA and the changes in the risk factors for NCDs of black South Africans. The study forms part of the baseline and five year follow-up of the South African leg of the Prospective Urban and Rural Epidemiological (PURE) study. This study is a longitudinal study of which the baseline data was allocated in 2005 and the five year follow-up allocated in 2010. 2 000 participants aged 30 years and older were recruited for the initial study – 1 000 urbanised (from Ikageng), and 1 000 rural black adults (from Ganyesa, Moswana and Tlakgameng). Data, including the PA questionnaires (Baecke & IPAQ-S) were collected by a specialised multidisciplinary team. After signing an informed consent form, questionnaires were completed during individual interviews and conducted by extensively trained fieldworkers in the language of the participants’ choice. The variables used in this study were anthropometric measurements, blood pressure, serum lipids and fasting blood glucose. BMI was calculated from the body weight divided by the height squared. Statistical analysis was performed using SPSS for windows (Version 21.0). Descriptive statistics were performed to determine the characteristics of the participants. The relationship between Baecke and IPAQ-S was determined by means of a partial correlation adjusting for age and BMI. Linear regression analyses were used to determine the relationship between the changes in PA (dependent variable) and BMI (predictor variable) and then adjusted for gender and age. Where a significant relationship was observed (in the case of setting, urban versus rural) separate analyses were performed for the rural and urban participants; likewise the relationship between the changes in PAI (dependant variable) and the change in the risk factors for NCDs (predictor variable) was determined by means of linear regression analysis, and also adjusted for gender, change in age and setting (urban/rural). The results from this study indicated that a weak but significant relationship was found between the Baecke and IPAQ-S (Spearman r = 0.243; p = 0.00) when adjusted for age and BMI. Significant differences were found between rural and urban participants for age and BMI in 2005, where the urban participants where older and reported a higher BMI compared to the rural participants. Rural women gained significantly more weight than the urban women. The PAI in the urban participants increased from 2005 (6.40 ± 1.84) to 2010 (7.50 ± 1.40), but decreased in rural participants from 2005 (8.21 ± 1.48) to 2010 (5.10 ± 1.54). Change in BMI was significantly inverse associated with change in PA for the urban population after adjusting for gender, setting (rural/urban) and change in age (β = -0.10; p = 0.004). Significant differences were found for resting systolic blood pressure (SBP) for the rural (129.72 ± 23.30) and urban (137.33 ± 25.14) participants as well as the diastolic blood pressure (DBP) of rural (86.16 ± 14.48) and urban (89.28 ± 14.46), fasting glucose of rural (4.88 ± 1.23) and urban (5.10 ± 1.86), triglycerides of rural (1.21 ± 0.64) and urban (1.38 ± 0.92) and physical activity index (PAI) of rural (8.21 ± 1.48) and urban (6.40 ± 1.84) in 2005. There were significant changes in the high density lipoprotein cholesterol (HDL-C) and in the low density lipoprotein cholesterol (LDL-C). Although the overall PAI decreased from 2005 (7.30 ± 1.90) to 2010 (6.46 ± 1.85), it increased in urban participants (6.40 ± 1.84 – 7.50 ± 1.40) and decreased in rural participants (8.21 ± 1.48 – 5.10 ± 1.54). A significant negative relationship between changes in PAI and changes in blood pressure (systolic and diastolic), total cholesterol and LDL-cholesterol was found when adjusted for gender. When changes in PA and changes in risk factors were separated according to gender, a significant negative relationship was found between PA and diastolic blood pressure (β -0.63; p 0.02) in the male population, and a significant negative relationship for females between PA and systolic blood pressure (β -1.05; p 0.002), diastolic blood pressure (β -0.59; p 0.003), total cholesterol (β -0.05; p 0.01) and LDL-cholesterol (β -0.07; p 0.00). In conclusion, the study found that the low but significant correlation between PAI assessed with the Baeck questionnaire and IPAQ-S makes both questionnaires applicable for the South African context, however the Baecke questionnaire is based on various domains for PA, while the IPAQ-S report on time spent in physical activity. Over the 5-year period PA decreased in this black South African population with a concomitant increase in BMI. Biological risk factors for NCDs increased from 2005 to 2010. The change in PA was inversely related to changes in total blood pressure. Future in PA interventions would be beneficial in the management of hypertension in the at risk South African black population.
- Health Sciences