Physical activity status, chronic stress, cardiovascular risk factors and telomere length in an urban South African teachers' cohort : the SABPA study

Abstract

The dose-response relationship between physical activity (PA), disease and mortality has primarily been obtained from self-report questionnaires in Western populations. A major limitation of self-reported PA is the likelihood of measurement error and these recordings cannot account for all 24-h activities, thus negating the influence of sedentary time and daily light intensity activity. Modern-day studies using objective measures of PA are highly controversial in the description of PA, as well as reliable wear time of these objective devices to accurately assess PA behaviour. The aim of the research presented in this thesis was to ascertain the associations between seven-day objectively measured PA (expressed as time spent in four different metabolic equivalent of task (MET) categories), cardiovascular disease risk factors (24-h ambulatory blood pressure and central obesity), chronic stress (General Health Questionnaire total score and serum cortisol) and DNA damage (leukocyte telomere length) in a cohort of African and Caucasian school teachers recruited from the Dr Kenneth Kaunda Education District in the North West Province of South Africa. All parameters were objectively measured (the GHQ was only added for thoroughness on measures of cognitive perceived stress) in the study population. The Africans (n=96) were younger than the Caucasians (n=107) (48.33 versus 51.06 years, p=0.024), but presented with slightly higher waist circumferences, significantly higher 24-h ambulatory systolic blood pressure (SBP, p≤0.000), diastolic blood pressure (DBP, p≤0.000) and mean arterial pressure (MAP, p≤0.000); significantly higher perceived stress scores (GHQ total scores, p=0.001) and significantly shorter telomeres (p≤0.000). The hypertensive participants in the total group (Africans and Caucasians combined) recorded 2.2 hours (12.4%) more daily awake sedentary time than the normotensive participants (p=0.004) and sedentary time was also a slightly better predictor of hypertension than moderate and vigorous activity time (Odds ratio=1.00, p=0.006). Irrespective of race and sex, 24-h SBP and DBP measurements were respectively associated with daily awake sedentary time (ß=0.17, p=0.018 and ß=0.18, p=0.020), light activity time (ß=-0.15, p=0.043 and ß=-0.16, p=0.041), waist circumference (ß=0.45, p≤0.000 and ß=0.33, p≤0.000) and log serum gamma glutamyl transferase (γ-GT, alcohol use) (ß=0.18, p=0.018 and ß=0.24, p=0.004). An older age (ß=-0.28, p≤0.000), higher alcohol consumption (ß=-0.21, p=0.003) and increased central obesity (ß=-0.17, p=0.017) were associated with shorter telomeres. Attenuated cortisol levels (ß=-0.12, p=0.068) showed a tendency towards associations with longer telomeres that may indicate possible cortisol down regulation to protect against DNA damage. Time spent in the different MET-categories showed no direct associations with either cortisol or telomere length. However, a sensitivity analysis indicated that daily light intensity activity time was significantly correlated with lower waist circumference (r=-0.21, p=0.004); a parameter associated with both cortisol (ß=-0.22, p=0.003) and telomere length (ß=-0.17, p=0.017). The thorough recording of PA during the true awake time of 24-h cycles over a period of seven days ensured that the beneficial effect of light intensity activities, as well as the detrimental effect of sedentary time, was highlighted by this study. The average awake time of all ethnic and sex groups were around 17 hours per day, which was more than most previous studies using objective measures of PA. The exclusion of participants who did not comply through wearing the Actiheart for a full seven days (n=143, 40%) did, however, have a negative impact on sample size that may have affected the statistical power for uncovering some significant associations and the high participant burden of the Actiheart device became clear. Therefore, the researchers used the data of the full seven-day recordings to also determine the minimum number of consecutive days the Actiheart device could be worn to accurately estimate energy expenditure and PA. The two-day combination of Wednesday-to-Thursday did not differ from the weekly average TEE, as well as for all MET-categories in all ethnic and sex groups. This two-day combination is practically convenient and would lessen participant burden. Future researchers are urged to test this combination in other populations to standardize Actiheart wear time. It can be concluded from the findings in this study that less daily awake sedentary time, more light intensity activity time, as well as lower alcohol consumption favour improved health as it is beneficial to 24-h ambulatory blood pressure and helps to maintain a healthy waist circumference, which ultimately influence telomere shortening. Furthermore, the two-day combination of Wednesday-to-Thursday seems to be sufficient to accurately estimate weekly energy expenditure and habitual PA with the Actiheart apparatus.