The Mediating Effects of Racial and Social Health Disparities on Cancer Urgent Care Outcomes: An Urban Cancer Center Experience

Background

Therapy and disease related complications are common among patients with cancer, leading to increased acute care resource use during the treatment period. Previously, we published an analysis supporting the efficacy of Henry Ford Cancer Institute’s (HFCI) cancer urgent care clinic in reducing excess emergency department (ED) visits in patients with hematologic malignancies. Keeping in mind our diverse patient population, we sought to assess whether racial and socioeconomic variables impacted this observed benefit.

Methods

We conducted a retrospective review of patients with active hematologic malignancies seen at HFCI’s cancer urgent care clinic between January 1, 2021, to December 31, 2022. As in our prior analysis, determination of whether cancer urgent care interventions prevented a subsequent ED visit, denoted at the end of each visit note by the performing provider, was collected. Patient demographics, malignancy, treatment characteristics, performance status (ECOG), Charlson Comorbidity Index (CCI), reason for urgent care visit, and interventions performed therein were collected for each patient along with Area Deprivation Index (ADI), insurance status, and race. A multivariable regression analysis was conducted to assess the impact of these variables in whether ED disposition was prevented by urgent care interventions.

Results

Both initial patient visits (n=141) and repeat presentations (n=269) to cancer urgent care during the study period were analyzed; 51.8% of patients were male and 48.2% were female. 51.8% of patients identified as black, 29.8% as white, and 18.4% as other/non-white. 35% of patients had private insurance while 54.3% and 10.7% had Medicare and Medicaid respectively. 21.3% of patients were ECOG 0, while 48.9%, 22.0%, and 7.8% were ECOG 1, 2, and 3 respectively. Mean CCI was 4.1 (SD 2.24), mean state ADI was 6.2 (SD 3.21), and mean national ADI was 70.1 (SD 26.7).

86.5% of patients were on active treatment at the time of visit, 96.6% of whom were on standard of care regimens with the remainder on clinical trials. Of patients on any treatment, 41.5% presented to urgent care within the first two cycles of therapy while 58.5% presented after the first two cycles. The most represented malignancies were multiple myeloma and high-grade lymphoma at 33.3% and 19.1% of cases respectively.

Of the 138 initial cancer urgent care visits in which complete data were available, a subsequent ED visit was prevented in 103 cases. In multivariable regression analysis, no difference was seen in the odds of preventing an ED visit when accounting for patients’ race, ADI, or insurance status while controlling for all other variables; however, ECOG score and patient presentation during the first two cycles of therapy were statistically significant mediators of this outcome. The odds of preventing an ED visit were decreased by 86% (95% CI [0.03-0.67]) for patients ECOG 2 compared to ECOG 0 and 94% (95% CI [0.01-0.60]) for patients ECOG 3 compared to ECOG 0 (p=0.018). A visit to urgent care was 3.21 times (95% CI [1.01-10.24]) more likely to prevent a subsequent ED visit for patients presenting after their first two cycles of therapy compared to those presenting during their first two cycles while controlling for all other variables (p=0.008). These findings were re-demonstrated when analyzing repeat patient presentations to cancer urgent care during the study period (n=269).

Discussion

Significant morbidity is imposed on patients undergoing treatment for hematologic malignancies. Our data align with existing literature demonstrating that dedicated cancer urgent care centers can reduce ED resource use while ameliorating disease related complications. In our analysis, indices of patient morbidity such as ECOG and proximity to initiation of therapy predictably affected the efficacy of cancer urgent care preventing ED disposition. That CCI was not seen as a significant mediator of this outcome is likely due to the low CCI of our patient population with narrow standard deviation.

The benefit of cancer urgent care in preventing ED disposition was otherwise seen equally across our patient population, regardless of ADI, race, or insurance status. These findings suggest that dedicated cancer urgent care centers are socially equitable resources to offload the burden of acute care in diverse patient populations with hematologic malignancy.