Childhood deprivation and later-life cognitive function in a population-based study of older rural South Africans

Lindsay C Kobayashi; M Maria Glymour; Kathleen Kahn; Collin F Payne; Ryan G Wagner; Livia Montana; Farrah J Mateen; Stephen M Tollman; Lisa F Berkman

doi:10.1016/j.socscimed.2017.08.009

Childhood deprivation and later-life cognitive function in a population-based study of older rural South Africans

Soc Sci Med. 2017 Oct:190:20-28. doi: 10.1016/j.socscimed.2017.08.009. Epub 2017 Aug 14.

Authors

Lindsay C Kobayashi¹, M Maria Glymour², Kathleen Kahn³, Collin F Payne⁴, Ryan G Wagner⁵, Livia Montana⁴, Farrah J Mateen⁶, Stephen M Tollman³, Lisa F Berkman⁷

Affiliations

¹ Harvard Center for Population and Development Studies, Cambridge, MA, USA. Electronic address: kobayashi@hsph.harvard.edu.
² Harvard Center for Population and Development Studies, Cambridge, MA, USA; Department of Epidemiology and Biostatistics, University of California, San Francisco, San Francisco, CA, USA.
³ MRC-Wits Rural Public Health and Health Transitions Research Unit (Agincourt), School of Public Health, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa; INDEPTH Network, Accra, Ghana; Epidemiology and Global Health Unit, Department of Public Health and Clinical Medicine, Umeå University, Sweden.
⁴ Harvard Center for Population and Development Studies, Cambridge, MA, USA.
⁵ MRC-Wits Rural Public Health and Health Transitions Research Unit (Agincourt), School of Public Health, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa; INDEPTH Network, Accra, Ghana.
⁶ Harvard Medical School, Boston, MA, USA.
⁷ Harvard Center for Population and Development Studies, Cambridge, MA, USA; MRC-Wits Rural Public Health and Health Transitions Research Unit (Agincourt), School of Public Health, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa.

Abstract

Rationale: Little research has evaluated the life course drivers of cognitive aging in South Africa.

Objectives: We investigated the relationships of self-rated childhood health and father's occupation during childhood with later-life cognitive function score and whether educational attainment mediated these relationships among older South Africans living in a former region of Apartheid-era racial segregation.

Methods: Data were from baseline assessments of "Health and Aging in Africa: A Longitudinal Study of an INDEPTH Community" (HAALSI), a population-based study of 5059 men and women aged ≥40 years in 2015 in rural Agincourt sub-district, South Africa. Childhood health, father's occupation during childhood, and years of education were self-reported in study interviews. Cognitive measures assessed time orientation, numeracy, and word recall, which were included in a z-standardized latent cognitive function score variable. Linear regression models adjusted for age, sex, and country of birth were used to estimate the total and direct effects of each childhood risk factor, and the indirect effects mediated by years of education.

Results: Poor childhood health predicted lower cognitive scores (total effect = -0.28; 95% CI = -0.35, -0.21, versus good); this effect was not mediated by educational attainment. Having a father in a professional job during childhood, while rare (3% of sample), predicted better cognitive scores (total effect = 0.25; 95% CI = 0.10, 0.40, versus unskilled manual labor, 29% of sample). Half of this effect was mediated by educational attainment. Education was linearly associated with later-life cognitive function score (0.09; 95% CI = 0.09, 0.10 per year achieved).

Conclusion: In this post-Apartheid, rural South African context, older adults with poor self-reported childhood health or whose father worked in unskilled manual labor had relatively poor cognitive outcomes. Educational attainment strongly predicted cognitive outcomes, and appeared to be, in part, a mechanism of social stratification in later-life cognitive health in this context.

Keywords: Aging; Cognitive aging; Cognitive function; Education; Rural; Self-rated health; Socioeconomic conditions; South Africa.

Publication types

Research Support, N.I.H., Extramural

MeSH terms

Adult
Age Factors
Aged
Aged, 80 and over
Aging / psychology*
Cognition*
Factor Analysis, Statistical
Female
Health Status Disparities
Humans
Linear Models
Longitudinal Studies
Male
Middle Aged
Psychosocial Deprivation*
Risk Factors
Rural Population / statistics & numerical data
Self Report
Socioeconomic Factors
South Africa

Grants and funding

P01 AG041710/AG/NIA NIH HHS/United States