This is the first study to examine community SES and individual SES measures simultaneously as predictors of rOA and corresponding symptoms in a community-based study of African Americans and Caucasians. Our study results indicate that individual SES measures (education and occupation) influence pain and physical function in a cohort of rural, community-dwelling adults with knee rOA. Further, community SES (block group poverty rate) is associated with pain among a subset of individuals with symptomatic knee OA. These data suggest that individuals with knee rOA who are at the highest risk of developing disability and pain are those who have lower SES.
In our analyses, significant associations with pain and physical function remained independently associated with individual and community SES measures for persons with rOA even when the three SES variables were simultaneously introduced as the main explanatory variables in multivariable analyses. Further, our results remained after adjusting for hip pain, which is also a major contributor to disability. In examining the association with educational attainment more specifically, persons with less than 12 years of education were more likely to have worse WOMAC Function and WOMAC Total scores compared to individuals with higher educational attainment, after adjusting for occupation type and poverty. These results are similar to those found in a recent study by Lopez-Olivio et al. which reported that among individuals who underwent knee replacement, those with less than a high school education had worse WOMAC Pain scores and Function scores when compared with those who had at least a high school education [39
]. An additional study using data from the National Health and Nutrition Examination Survey I (NHANES I) found that, among individuals self-reporting knee OA, a low educational level was associated with both more severe radiographic findings as well as more pain [40
We also observed that persons working in non-managerial occupations were more likely to have worse WOMAC Pain scores compared to individuals in managerial positions. Further, persons living in high poverty communities (>25% poverty rate) tended to have worse WOMAC Function and WOMAC Total scores compared to people living in low poverty communities, although not at a level of statistical significance. A recent study reported that adults with chronic knee and/or hip pain who lived in deprived areas were more likely to develop disability [41
] while another study reported that patients who lived in areas that were socioeconomically deprived benefitted less from knee replacement surgery than those who did not live in deprived areas [42
]. Community poverty may affect those with OA differently, which may lead to greater disability. Communities with high poverty rates often have limited resources including fewer clinics, safe options for public transportation, community centers and safe places to exercise, as well as poorly kept sidewalks and less access to safe streets, all resources that can often contribute to the improvement of function, pain and disability in individuals with OA [43
Fewer significant associations between SES and pain and disability outcomes were found within the subgroup of individuals with symptomatic knee OA, which may be due to the small subsample size (n
471). In this subsample, multivariable analyses revealed that living in a high poverty community was significantly associated with WOMAC Pain scores, and associations with WOMAC Function scores and WOMAC Total scores approached statistical significance. We are unsure why we found little association between most SES measures and physical function measures. However, our results for this sub-population are in line with those from another study that failed to find an association for education with disability [44
Overall these results for the associations of SES on disability among those with knee OA are similar to previous findings between SES and disability among those with OA of the hip [23
]. This suggests that a broader rOA relationship may exist between SES measures and physical function, pain and stiffness that is not localized to specific joints, further expanding clinical application of the data. This is especially true for individual SES (education and occupation), but may be equally important for community SES as more sensitive and detailed measures of community SES are tested.
To date, the exact mechanisms of how SES leads to disability remain unclear. While physical features such as deformity and muscle weakness are important aspects, other factors may also have a role in disease progression to disability. A number of cross-sectional studies have demonstrated that psychological [48
], demographic [51
], clinical and biomechanical [52
] factors are associated with disability among patients with knee OA. These sociodemographic characteristics may influence health behaviors that lead to OA diagnosis (i.e., seeking medical care) or enhance progression of the disease [53
]. Although many of these factors were controlled for in our analyses, it is possible that the associations seen may be due to other circumstances associated with SES, such as lifestyle choices (smoking, diet, obesity, physical activity, etc.), demographic characteristics (age and race), or community and psychological factors (perceived helplessness, social support and perceived discrimination) [54
]. For instance, obesity and low physical activity levels have both been shown to have both direct and indirect effects on physical function and disability [25
]. Further, SES may also have an effect on level of medical care itself. Individuals with lower SES levels often do not have access to affordable medical services and interventions or access to quality healthcare, which are important determinants of health outcomes in OA [58
There are some limitations to our study that warrant mention. Our study was cross-sectional in design so we can only demonstrate an association and not a causal relationship. Also, our measures of knee disability and functional limitations using the WOMAC instrument, while widely used, are not perfect. However, the WOMAC instrument can detect differences between better vs. worse outcomes when comparing those with mild or moderate disease vs. those with severe knee OA [59
]. Additionally, although the measures are still self-reported, WOMAC is designed specifically to measure pain and function of individuals with lower body OA and is among the most sensitive of all instruments used in the assessment of OA of the knee or hip [29
] and has been used extensively in observational studies and clinical trials. Further, studies have shown that the WOMAC pain and function subscales exhibit comparable or greater responsiveness to change than corresponding SF-36 subscales [59
]. Our measure of community SES (block group poverty) is a somewhat crude measure of community influence. However, area-level poverty has been shown to be a good measure for community-level SES and allows for the assessment of the impact of one’s local environment [36
]. Finally, our outcome measure data were self-reported, which may lead to some overestimation of socioeconomic differences in outcomes, compared with more objective clinical measures of impairment.
Strengths include that we carried out our study in a large community-based cohort of individuals with radiographically-confirmed OA instead of self-reported arthritis, with the inclusion of several SES measures and data for comorbid conditions and hip symptoms. In addition, our study is unique in reporting results among a subset of individuals with symptomatic OA to account for pain. Most studies investigating the effectors of SES on disability have only one measure of SES, usually education. However, here we report results that include three SES measures that we evaluated for independent effects. Importantly, we have included a community-level measure of SES in addition to typical individual-level measures.