Exploring the Pathways Between Socioeconomic Status and ESRD

There is strong evidence of an association between low socioeconomic status (SES) and end-stage renal disease (ESRD).1, 2, 3 In this issue of AJKD, Ward examines this association more profoundly than previous studies, by comparing the relationship between SES and ESRD due to 3 different forms of kidney disease.4

While the relationship between low SES and kidney disease is evident, the pathways of this association are unclear. Few studies have investigated the actual ways in which SES might affect the development, progression, and complications of kidney disease. In his study, Ward addresses this issue by comparing the SES association with all primary renal diseases and then separately with 3 causes of ESRD, diabetes mellitus, systemic lupus erythematosus, and autosomal dominant polycystic kidney disease (ADPKD). Considering that low SES might influence kidney disease through decreased access to effective treatment, comparing the SES-ESRD relationships by causes that vary with respect to availability of such treatments might better distinguish the pathways that lead from low SES to ESRD.

Ward found strong SES-ESRD associations for ESRD due to diabetes mellitus, some association with ESRD due to lupus, and no apparent association for ESRD due to ADPKD. The author indeed expected the strongest SES-ESRD association for those with ESRD due to diabetes mellitus. Of the 3 conditions leading to ESRD, diabetes can be most readily controlled by effective treatment and thus the risk of ESRD can be prevented or mitigated. In discussing his findings, Ward suggests some ways in which low SES might impede effective treatment of diabetes, including factors related to inadequate access to antidiabetic medication and to lower patient education and health literacy.

The influence of low SES on health is likely to be complex and multifaceted. Physical factors, such as access to treatment, probably interact with psychological factors including low optimism5 and reduced coping skills.6 While it is difficult, and perhaps impossible, to tease out the exact mechanism of this relationship, comparing types of disease that might be more influenced by treatment, as Ward has done, is an insightful way to examine some of these pathways.

Factors aside from health care access and utilization might also explain the observed stronger association between low SES and ESRD due to diabetes. Negative health behaviors and conditions, such as obesity,7 smoking,7, 8, 9 lack of availability of healthy foods,10 and reduced physical activity,7 are associated with low SES and also associated with the development of diabetes, and thus, potentially, with subsequent diabetes-induced ESRD. Since diabetes-induced ESRD is likely more influenced by environmental and external factors than ESRD due to lupus nephritis and ADPKD, the stronger association of diabetes with low SES is understandable, and possibly due to a combination of factors listed above.

The strength of Ward’s study is its use of a uniquely comprehensive national database, the United States Renal Data System (USRDS), estimated to include about 93% of US adults diagnosed with ESRD. There are some limitations with this study, however, that include limited measures of SES. Ward acknowledges that neighborhood SES is better measured at a more meaningful socioeconomic level such as census tract or block group, rather than zip codes. Ward also mentions the lack of individual SES measures in his study as a limitation, in that area SES measures are imprecise indicators of individual SES measures. However, while some studies have found that individual SES associations are stronger in their effects on kidney disease than area SES measures,11 a recent analysis of older adults found stronger associations for neighborhood SES measures.12 Moreover, these studies found an independent relationship between kidney disease and neighborhood SES after adjusting for individual SES.11, 12 After accounting for individual socioeconomic means, it is likely that a low socioeconomic environment can influence the development of kidney disease. Research suggests that a myriad of environmental factors might influence kidney disease, including exposure to lead, stressful exposure to high crime levels, limited access to health information and resources, and limited availability of healthy foods and recreational resources.13, 14, 15

Despite some limitations with the data, the USRDS database presents a unique opportunity for future study related to SES and ESRD. The inverse relationship observed between SES and treated ESRD is of particular interest given the common means of health insurance shared by the study participants (ie, Medicare). While Ward’s study focused on conditions causing ESRD, future study on ESRD patients and their outcomes might shed light on pathways beyond health care access that might still be contributing to excess risk for those living in low SES areas and/or with limited socioeconomic means. The complex effects of SES despite health insurance may include delayed health care16 and waiting list entry for transplantation,17 and exposure to environmental nephrotoxins.18, 19

In the United States, race is an integral part of any SES-health relationship. Since race and SES are inextricably linked, it is difficult to tease apart this relationship.20 Ward attempts to examine this by exploring the SES-ESRD association in separate race/ethnic groups; he found a weaker association among African Americans, compared to whites, Asian/Pacific Islanders, Native Americans, and Hispanics, although the relative risks of diabetes-induced ESRD for the lowest SES quartile versus the highest were still statistically significant for African American men and women (see Ward, Fig 1). Ward suggests a number of reasons for the weaker association among African Americans, including genetic susceptibility to diabetes-induced renal damage, and residual confounding due to poor SES measures. However, since Ward used race-specific quartiles of SES, it is not possible to directly compare SES-ESRD associations between race/ethnic groups. This trend is also not clear in the literature; some studies have similarly found weaker neighborhood SES-kidney disease associations among African Americans,11, 12 while another study examining neighborhood poverty and ESRD found stronger socioeconomic differentials among African Americans,21 although these studies did not distinguish between the types of kidney disease as Ward does.

The weaker SES-kidney disease associations among African Americans found in Ward’s analyses and in other studies11, 12 are perhaps due to inadequate SES measures, as Ward suggests. For instance, it is possible that other factors such as racial discrimination and segregation are strongly associated with disease development in the African American population, and are not captured by conventional SES measures.

Despite the weaker SES differentials, however, in absolute terms higher rates of ESRD still plague the African American population. Based on the USRDS data for 2005, ESRD incidence rates were 268 per million population among whites and 991 per million among African Americans.22 It is imperative that future research continues to assess these racial differentials in the incidence of earlier stages of kidney disease and to address the high rates of ESRD incidence in the African American population, especially with regard to ESRD due to diabetes. Since studies, including Ward’s, find persistent racial disparities despite adjusting for SES,1, 2, 4 it is likely that issues extending beyond conventional SES contribute to these racial disparities. One such barrier, medication adherence, was recently found to be a significant predictor of low glycemic control for African Americans compared to whites after adjusting for several other hypothesized mechanisms for racial differences in glycemic control.23 It is possible that the barriers leading to racial disparities in diabetic kidney disease consist of social and cultural factors that extend beyond SES, including confidence and trust in the health care system and its providers.24, 25

Future research should continue to explore not only the familiar low SES-ESRD association, but also the pathways that might underlie this relationship with ESRD and earlier stages of kidney disease. Ultimately, the findings of such work should inform policies to implement interventions to reduce socioeconomic disparities in kidney disease.