Does subjective social status predict health and change in health status better than objective status?

• Published in: Psychosomatic medicine Vol. 67; no. 6; p. 855
• Main Authors: Singh-Manoux, Archana, Marmot, Michael G, Adler, Nancy E
• Format: Journal Article
• Language: English
• Published: United States 01.11.2005
• Subjects: Adult; Age Factors; Cohort Studies; Female; Health Status; Humans; Linear Models; Longitudinal Studies; Male; Middle Aged; Morbidity; Probability; Social Class; Sociometric Techniques; Surveys and Questionnaires; United Kingdom
• ISSN: 1534-7796
• DOI: 10.1097/01.psy.0000188434.52941.a0

To examine, among middle-aged individuals, if subjective socioeconomic status (SES) predicts health status and change in health status over time better than objective SES. Data are from the Whitehall II study, a prospective study of British civil servants. SES data are drawn from Phase 5 (1997-1999) of the study and health data from Phases 5 and 6 (2000-2001). Physical and mental component scores from the Short Form 36, the General Health Questionnaire, and self-rated health were used to assess health status. Multiple linear regressions were used to examine the relationship between SES and health and change in health status. Complete data were available on 5486 people. Results show both measures of SES to be global measures of SES. Both measures of SES were significantly associated with health outcomes and with decline in health status over time. However, when both objective and subjective measures of SES are entered simultaneously in the model to predict change in health status, it was only the latter that continues to be significantly associated with health and changes in health. Subjective SES is a better predictor of health status and decline in health status over time in middle-aged adults. These results are discussed in terms of three possible explanations: subjective SES is a more precise measure of social position, the results provide support for the hierarchy-health hypothesis, and the results could be an artifact of common method variance.