Martin Luther King Jr once said, “Of all the forms of inequality, injustice in health is the most shocking and the most inhumane.”¹ More than 50 years later, health care disparities have yet to be eradicated. It has long been recognized that race and socioeconomic status are associated with incidence of disease, receipt of health care, and health outcomes. These disparities are well documented in pancreatic cancer, of which pancreatic ductal adenocarcinoma (PDAC) is the most common subtype. The study by Tsilimigras et al,² in this issue, contributes to the growing body of research on racial and socioeconomic disparities in PDAC treatment and outcomes.

This study examined the relationship of socioeconomic factors, including social vulnerability, with quality of care in the treatment of metastatic PDAC using the SEER-Medicare database. The authors applied a quality score—originally proposed for metastatic urothelial cancer—that incorporates receipt of systemic chemotherapy, receipt of palliative care, and cancer-specific survival >1 year.³ They found that 93.3% of patients with metastatic PDAC met at least one of these metrics, thus satisfying the criteria for the quality score. When evaluating individual components of the quality score, the authors found that 62.2% of patients received systemic therapy, 83.3% received palliative care/hospice services, and 13.7% had cancer-specific survival >1 year. Lower social vulnerability, married status, and higher income were independently associated with meeting the quality score. Social vulnerability was independently associated with lower rates of palliative care and systemic chemotherapy utilization, whereas married status and higher income were both independently associated with greater systemic chemotherapy utilization and improved survival. In evaluating the components of the social vulnerability index used in this study, the authors found that racial/ethnic minority status and lower socioeconomic status were independently associated with failing to meet the quality metric.

The study by Tsilimigras et al² builds on prior research evaluating disparities in treatment and outcomes for metastatic PDAC using national databases. For example, using the National Cancer Database (NCDB), Khan et al⁴ demonstrated that lower socioeconomic status and non-White race were independently associated with decreased utilization of palliative care for metastatic PDAC, both before and after Medicaid expansion. Another analysis using the NCDB found that non-White patients were less likely to receive any treatment for PDAC, regardless of stage and independent of other variables,⁵ and yet another NCDB analysis—from the same group as behind the current study—similarly demonstrated that Black race and lower income were independently associated with decreased guideline-concordant care for metastatic PDAC.⁶ Interestingly, in a separate analysis of the SEER-Medicare dataset, it was previously reported that race and ethnicity were not independently associated with receiving chemotherapy and radiation for metastatic PDAC.⁷ Notably, the prior study assessed race and ethnicity at the individual patient level, whereas the current study analyzed social vulnerability, of which racial/ethnic minority population was one factor, in the area where each patient resided.

Many limitations of the study Tsilimigras et al² are common to other studies of this type and are well acknowledged by the authors, including the use of large, retrospective registry databases. Additionally, although SEER-Medicare is robust in identifying Medicare beneficiary patients with cancer, it only captures patients aged ≥65 years, and thus the findings might not be representative of all patients, especially those younger in age. Limitations that are specific to this study are more nuanced. In measuring quality of care for metastatic PDAC, it remains unclear whether a patient should be considered as having received “quality care” simply because that patient survived beyond 1 year, even when systemic chemotherapy and palliative care are omitted. Extended survival in the absence of treatment is likely due to favorable tumor biology rather than the quality of care received. Moreover, identifying patients who can be expected to have such an outcome early in the disease course—when decisions about treatment are made—is inherently difficult. Nevertheless, this was not a major factor in the study’s findings given that only 0.2% of all patients did not receive either systemic chemotherapy or palliative care yet survived beyond 1 year. Another limitation is that the authors do not report on whether treatments were offered but not given or refused. A recent study found that Black patients were more likely than White patients to decline cancer-related surgical care.⁸ Insight into the reasons why patients do not ultimately receive treatment and whether it is related to access, the options offered to them, or the options that they choose to pursue will be helpful in understanding and addressing disparities. Additionally, the extremely short survival in this study (median survival, 3.3 months) may have influenced results, because some patients may have died before palliative care could be arranged.

Some of the ways in which race and socioeconomic variables are analyzed in this study are common in the field but bear mentioning. For instance, income and social vulnerability were assessed at the census tract level as opposed to the individual patient level, meaning that these factors were inferred based on residential location. It is well established that there is important intersectionality of individual and neighborhood-level characteristics in health care disparities, including those seen in PDAC treatment and outcomes.⁹ Additionally, as noted earlier, race was only evaluated as a component of the census tract social vulnerability rather than as an independent variable at the patient level. Future studies may benefit from assessing other vulnerable populations not explicitly examined in this study. Furthermore, although the social vulnerability index is widely used in health disparities research, it was originally designed to assess community resilience to natural disasters, and therefore might be a crude metric.¹⁰ Whether other metrics specifically designed for health services research might be better suited to evaluate health disparities remains to be seen.

In all, the authors should be commended for the contribution to our knowledge of health care disparities in metastatic PDAC. However, future research must move beyond identifying disparities and toward elucidating the mechanisms that drive them. These mechanisms are likely complex as they may relate to numerous factors, including biology, societal constructs, economics, policy, geography, and environment, as well as historic and contextual inequities. The current methodology for such research remains crude in large part due to limitations of available data. Going forward, efforts must also focus on mitigating inequities in health care. This will require not only understanding the mechanisms underlying disparities but also developing interventions to improve access, reduce barriers, and ensure equitable care for diverse patient populations.