A cross-sectional study of physical activity and sedentary behaviours in a Caribbean population: combining objective and questionnaire data to guide future interventions

Christina Howitt, Soren Brage, Ian R. Hambleton, Kate Westgate, T. Alafia Samuels, Angela MC Rose, Nigel Unwin

Current understanding of population physical activity (PA) levels and sedentary behaviour in developing countries is limited, and based primarily on self-report. We described PA levels using objective and self-report methods in a developing country population. PA was assessed in a cross-sectional, representative sample of the population of Barbados (25-54 years), using a validated questionnaire (RPAQ) and individually calibrated combined heart rate and movement sensing monitors. The RPAQ collects information on recalled activity in 4 domains: home, work, transport, and leisure. Physical inactivity was defined according to World Health Organization (WHO) guidelines; sedentary lifestyle was defined as being sedentary for 8 h or more daily; PA overestimation was defined as perceiving activity to be sufficient, when classified as 'inactive' by objective measurement. According to objective estimates, 90.5 % (95 % CI: 83.3,94.7) of women and 58.9 % (48.4,68.7) of men did not accumulate sufficient activity to meet WHO minimum recommendations. Overall, 50.7 % (43.3,58.1) of the population was sedentary for 8 h or more each day, and 60.1 % (52.8,66.9) overestimated their activity levels. The prevalence of inactivity was underestimated by self-report in both genders by 28 percentage points (95 % CI: 18,38), but the accuracy of reporting differed by age group, education level, occupational grade, and overweight/obesity status. Low PA was greater in more socially privileged groups: higher educational level and higher occupational grade were both associated with less objectively measured PA and more sedentary time. Variation in domain-specific self-reported physical activity energy expenditure (PAEE) by educational attainment was observed: higher education level was associated with more leisure activity and less occupational activity. Occupational PA was the main driver of PAEE for women and men according to self-report, contributing 57 % (95 % CI: 52,61). The most popular leisure activities for both genders were walking and gardening. The use of both objective and self-report methods to assess PA and sedentary behaviour provides important complementary information to guide public health programmes. Our results emphasize the urgent need to increase PA and reduce sedentary time in this developing country population. Women and those with higher social economic position are particularly at risk from low levels of physical activity.