Abstract
Socioeconomic position is most commonly operationalised as education, social class, or income. Socioeconomic position can also be operationalised as occupational complexity. Studies of health inequalities in the general population show that the choice of indicators influences the results. However, less is known about the influence of indicator choice in older populations. We explored whether the strength of the association between socioeconomic position and health in old age (69–88) varied by the indicator of socioeconomic position used.
Data from two nationally representative Swedish surveys (LNU and SWEOLD) were combined, providing 20 years of follow-up. Average marginal effects were analysed to compare the association between the indicators of socioeconomic position (education, social class, occupational complexity and income) and two late-life outcomes: mobility limitations and psychological distress.
All indicators were associated with both outcomes in late-life; differences were small. Income was independently and most strongly associated with both outcomes and contributed the most to model fit. Social class contributed the least to both outcomes. The results indicated that occupational complexity was at least as strongly associated to health in old age as social class and education.
If the primary objective of including an indicator of socioeconomic position is to adjust the model for socioeconomic differences rather than to analyse these inequalities, income may be the preferable indicator, as it seems to capture the most variation in late life health. If, on the other hand, the primary objective of a study is to examine health inequalities, the choice of indicator should be theoretically guided.