The choice that matters: the relative influence of socioeconomic status indicators on chronic back pain- a longitudinal study.

BACKGROUND

In health research, indicators of socioeconomic status (SES) are often used interchangeably and often lack theoretical foundation. This makes it difficult to compare results from different studies and to explore the relationship between SES and health outcomes. To aid researchers in choosing appropriate indicators of SES, this article proposes and tests a theory-based selection of SES indicators using chronic back pain as a health outcome.

METHODS

Strength of relationship predictions were made using Brunner & Marmot's model of 'social determinants of health'. Subsequently, a longitudinal study was conducted with 66 patients receiving in-patient treatment for chronic back pain. Sociodemographic variables, four SES indicators (education, job position, income, multidimensional index) and back pain intensity and disability were obtained at baseline. Both pain dimensions were assessed again 6 months later. Using linear regression, the predictive strength of each SES indicator on pain intensity and disability was estimated and compared to the theory based prediction.

RESULTS

Chronic back pain intensity was best predicted by the multidimensional index (beta = 0.31, p < 0.05), followed by job position (beta = 0.29, p < 0.05) and education (beta = -0.29, p < 0.05); whereas, income exerted no significant influence. Back pain disability was predicted strongest by education (beta = -0.30, p < 0.05) and job position (beta = 0.29, p < 0.05). Here, multidimensional index and income had no significant influence.

CONCLUSIONS

The choice of SES indicators influences predictive power on both back pain dimensions, suggesting SES predictors cannot be used interchangeably. Therefore, researchers should carefully consider prior to each study which SES indicator to use. The introduced framework can be valuable in supporting this decision because it allows for a stable prediction of SES indicator influence and their hierarchy on a specific health outcomes.