Approaches for assessing the role of household socioeconomic status on child anthropometric measures in urban South Africa

Authors

  • Zoë A. Sheppard,

    Corresponding author
    1. Department of Human Sciences, Loughborough University, Loughborough, Leicestershire, LE11 3TU, United Kingdom
    • Department of Human Sciences, Loughborough University, Loughborough, Leicestershire LE11 3TU, United Kingdom
    Search for more papers by this author
  • Shane A. Norris,

    1. MRC Mineral Metabolism Research Unit, Department of Pediatrics, University of the Witwatersrand, Johannesburg, South Africa
    Search for more papers by this author
  • John M. Pettifor,

    1. MRC Mineral Metabolism Research Unit, Department of Pediatrics, University of the Witwatersrand, Johannesburg, South Africa
    Search for more papers by this author
  • Noël Cameron,

    1. Department of Human Sciences, Loughborough University, Loughborough, Leicestershire, LE11 3TU, United Kingdom
    2. MRC Mineral Metabolism Research Unit, Department of Pediatrics, University of the Witwatersrand, Johannesburg, South Africa
    Search for more papers by this author
  • Paula L. Griffiths

    1. Department of Human Sciences, Loughborough University, Loughborough, Leicestershire, LE11 3TU, United Kingdom
    2. MRC Mineral Metabolism Research Unit, Department of Pediatrics, University of the Witwatersrand, Johannesburg, South Africa
    Search for more papers by this author

Abstract

The objectives of this article were to compare the variance explained in anthropometric outcomes when using individual measures of socioeconomic status (SES) versus different approaches to create SES indices within the urban African context, and to examine the influence of SES measured during infancy on child anthropometric outcomes at 7/8 years. Data from the 1990 Birth-to-Twenty cohort study set in Johannesburg-Soweto, South Africa, were used (n = 888). Linear regression models were used to investigate the association between SES (individual and index measures) during infancy and anthropometric measures at age 7/8 years, controlling for sex, age, and population group. Both individual and index measures of SES explained similar proportions of the variance for each anthropometric outcome. SES measured during infancy influenced weight more than height at age 7/8 years in Johannesburg-Soweto. Positive associations were found between SES and the anthropometric measures––ownership of a car, telephone, and having an inside flush toilet were the most significant SES variables. The similarities observed in the variance explained relating to the anthropometric outcomes suggest that researchers who want to adjust for SES in analyses could use an SES index to make statistical models more parsimonious. However, using such indices loses information relating to the specific socioeconomic factors that are important for explaining child anthropometrics. If the purpose of the research is to make policy recommendations for the improvement of child growth, individual SES variables would provide more specific information to target interventions. Am. J. Hum. Biol., 2009. © 2008 Wiley-Liss, Inc.

Ancillary