We are grateful for Nick Ruktanonchai’s comments (1) on our paper on health-related selection and social causation as explanations for socioeconomic inequalities (2). Most of the research in this domain uses self-reports of health to examine this issue. We contend that social patterning in reporting may introduce bias in such results. Our approach was to focus on objectively measured health indicators in order to reduce this threat to validity.
In his letter, Ruktanonchai notes that our data did not allow us to completely test the relative strengths of the social causation and health selection hypotheses during childhood. This is correct: Repeated measures of socioeconomic status (SES) and objective health indicators were only available across adulthood, since Whitehall II participants were aged 35–55 years upon entry into the study. Indeed, there are presently very few epidemiologic studies that track participants across the entire life course.
We found that retrospectively reported hospitalizations in childhood and low birth weight were associated with adulthood SES after adjusting for father’s social class. This finding is consistent with the health selection hypothesis. We have previously shown that father’s social class, a proxy measure of childhood SES, is generally not a strong predictor of health in adulthood, as indicated by coronary risk factor levels (3). This analysis suggests that social causation from childhood did not play a major role in our cohort. Further research is required to examine whether prospective studies starting from childhood replicate our findings.
Our design was strong in relation to studying changes in SES and health during adulthood. We were able to use independent indicators of SES (employment grade) and objective physiologic measures of cardiometabolic factors during adulthood, measured at baseline and at 5- and 10-year follow-ups. However, after pointing out the modest population-level differences observed in single risk factors between SES groups, Ruktanonchai suggests that the clinical relevance of these differences may be limited. We would argue that further analyses to complement our data on single risk factors are still needed in order to address this issue. Because cardio- and cerebrovascular diseases and diabetes have multifactorial etiologies, the risk of these diseases can be assessed reliably only with risk prediction algorithms combining multiple risk factors (4, 5). Such analyses in relation to social causation and health selection are currently not available. A well-documented illustration of the relevance of this principle is the European Cardiovascular Disease Prevention Guidelines, which note that a person with excessively raised cholesterol levels may, in fact, have a much lower 10-year risk of cardiovascular disease than a person with normal lipid levels but modestly elevated values for several other risk factors (4).
Taken together, we consider these findings and those of our paper to be a step forward in testing social causation and health-related selection using objective health data. To better understand the ways in which these processes might operate in adulthood, we will continue this line of inquiry in the Whitehall II study using risk algorithms (such as the Framingham scores), preclinical disease indicators (such as prediabetes), and manifest disease. We encourage other investigators to do the same with their data and so move towards a better understanding of the roles of social causation and health selection in health.
The Whitehall II study has been supported by grants from the United Kingdom Medical Research Council; the British Heart Foundation; the United Kingdom Health and Safety Executive; the United Kingdom Department of Health; the US National Heart, Lung, and Blood Institute (grant HL36310); the US National Institute on Aging (grants AG13196 and R01 AG034454); the US Agency for Health Care Policy and Research (grant HS06516); and the John D. and Catherine T. MacArthur Foundation Research Networks on Successful Midlife Development and Socio-economic Status and Health. M. E. is supported by the Work Environment Fund and the Academy of Finland (grant 128002); M. K. is supported by the BUPA Foundation, the Academy of Finland, and the European Union NEW OSH ERA Research Programme; G. D. B. is a Wellcome Trust Research Fellow; A. S.-M. is supported by a “EURYI” award from the European Science Foundation; J. E. F. is supported by the Medical Research Council; M. J. S. and M. H. are supported by the British Heart Foundation; and M. G. M. is supported by a Medical Research Council Research Professorship.
Conflict of interest: none declared.