Investigating Neighborhood Socioeconomic Status and Acuity of Illness Presentation in Children with Acute Leukemia

Background

Despite improvements in pediatric leukemia treatment, there continue to be disparities in outcomes among disadvantaged groups as lower socioeconomic status (SES) and minority race/ethnicity are associated with lower overall survival. We have previously shown that higher acuity of illness presentation is associated with mortality during induction chemotherapy. We hypothesized that this association was due to differences in access to healthcare, as patients who experience difficulties accessing healthcare would likely present with more severe illness manifestations. We further hypothesized that pediatric patients with acute myeloid leukemia (AML) and acute lymphoblastic leukemia (ALL) who resided in lower SES neighborhoods were more likely to have higher acuity of illness at initial disease presentation compared to patients who resided in higher SES neighborhoods.

Methods

This retrospective cohort study included patients aged 0 to 21 diagnosed with AML and ALL from 2012 to 2022 at University of California, San Francisco hospitals. The exposure variable was neighborhood SES (nSES), which was used as a proxy for a neighborhood resident's access to healthcare. Neighborhood SES was defined at the census tract level based on a patient's address at diagnosis and is presented as tertiles of statewide California neighborhoods. We employed a widely used composite measure from the California Neighborhoods Data System that includes education index, proportion with a blue-collar job, proportion above 200% of the poverty line, proportion employed, median rental cost, median value of owner-occupied housing, and median household income. Chart review was conducted to collect demographics, laboratory values, Intensive care unit (ICU) admission, and ICU interventions as markers of illness acuity, such as vasopressor support and mechanical ventilation. The primary outcome of interest was ICU admission within 72 hours of hospital admission. Race, ethnicity, and insurance status were included in a multilevel regression to estimate risk ratios and confidence intervals evaluating ICU admission within the first 72 hours by nSES.

Results

Of 154 patients, 87% of patients were diagnosed with ALL and 13% of patients were diagnosed with AML. Patients were primarily Hispanic (48%) and publicly insured (63%). 58% of patients were in the lowest nSES tertile (N=90), 21% were in the middle nSES tertile (N=33), and 20% were in the highest nSES tertile (N=31) based on California nSES data. There were no significant differences in imaging findings and laboratory abnormalities between nSES tertiles. 27% of patients (N=41) were admitted to the ICU within 72 hours of hospital admission.

Patients residing in neighborhoods in the lowest nSES tertile were 3.10 times more likely to be admitted to the ICU in the first 72 hours than patients from neighborhoods in the highest nSES tertile (confidence interval [CI]: 1.01-9.51). After adjusting for race, ethnicity, and insurance status, patients residing in neighborhoods in the lowest nSES tertile were 3.08 times more likely to be admitted to the ICU compared to patients residing in neighborhoods in the highest nSES tertile (CI: 0.99-9.57). Patients residing in neighborhoods in the middle nSES tertile were 3.44 times more likely to be admitted to the ICU compared to patients residing in the highest nSES tertile (CI: 1.06-11.20). After adjusting for race, ethnicity, and insurance status, patients residing in neighborhoods in the middle nSES tertile were 3.55 times more likely to go to the ICU (CI: 1.05-11.97) compared to patients residing in neighborhoods in the highest nSES tertile.

Conclusion

This is the first study to investigate the association between acuity of disease presentation and neighborhood SES for pediatric patients with leukemia. Our data showed that living in the lowest and middle nSES tertile neighborhoods was significantly associated with requiring ICU admission within the first 72 hours of hospital admission. These findings support the hypothesis that disparities in pediatric leukemia outcomes are related to barriers in access to care. Interventions focused on improving access to care, such as health navigators, may improve early outcomes for disadvantaged groups with pediatric leukemia.

No relevant conflicts of interest to declare.