Socioeconomic status across the life course and cancer risk behaviors among older adults in rural South Africa

Authors: Kobayashi LC

Category: International Issues in Cancer, Lifestyles Behavior, Energy Balance & Chemoprevention
Conference Year: 2018

Abstract Body:
Socioeconomic status (SES) at various points across the life course may influence cancer risk behaviors in later- life, although this relationship has rarely been investigated in rural, low-income settings. Using interview data from “Health and Aging in Africa: A Longitudinal Study of an INDEPTH Community in South Africa” (HAALSI), a population-based study of adults 40 years in rural South Africa in 2015 (n=5059), we aimed to estimate the associations between life course SES indicators (father’s occupation during childhood, education, and current household wealth) and cancer risk behaviors. The associations between each life course SES indicator and engaging in <210 min/week of moderate-to-vigorous physical activity (MVPA), having an overweight/obese BMI, drinking alcohol 5 days/week, and current smoking were estimated in mixed effects logistic regression models adjusted for age, sex, and country of birth, allowing for clustering by village of residence. Overall, 57% of participants engaged in <210 min/week of MVPA, 59% had an overweight/obese BMI, 6% drank alcohol 5 days/week, and 9% were current smokers. Low-skill father’s occupation (vs. high-skill) was associated with increased odds of low MVPA (OR=1.33; 95% CI: 1.17-1.52), but no other risk behaviors. Having no education (vs. any) was associated with increased odds of low MVPA (OR=1.27; 95% CI: 1.11-1.49) and drinking alcohol 5 days/week (OR=1.75; 95% CI: 1.31-2.32), and reduced odds of overweight/obese BMI (OR=0.86; 95% CI: 0.74- 1.01). Being in the poorest household wealth quintile (vs. richest) was associated with reduced odds of low MVPA (OR=0.78; 0.65-0.94) and overweight/obese BMI (OR=0.34; 95% CI: 0.27-0.42), and increased odds of smoking (OR=6.24; 95% CI: 4.23-9.20) and drinking 5 days/week (OR=2.93; 95% CI: 2.29-3.73). Behaviors did not cluster within villages, except for MVPA. Results indicate high prevalence of overweight/obesity and low physical activity among older, rural South Africans. These problems are concentrated among those of higher SES, opposite to trends previously observed in high-income countries and urban areas. Future work should investigate how to best support older adults living in rural, low-income areas in sub-Saharan Africa to engage in cancer-protective behaviors.

Keywords: socioeconomic status; rural; South Africa; behavior