AUTHORS
Savinnie Ho, BA, Maria Gonzalez Hinojosa, BS2, Nadia Bourdoud, MPH3, Michael Seid, PhD4, Lena E. Winestone, MD, MPH5
AFFILIATIONS
1 University of California, San Francisco, San Francisco, California, United States
2 University of California, San Francisco, San Francisco, California, United States
3 UCSF Benioff Children’s Hospitals, San Francisco, California, United States
4 Cincinnati Children’s Hospital, Cincinnati, Ohio, United States
5 UCSF Benioff Children’s Hospital, San Francisco, California, United States, UCSF Benioff Children’s Hospitals, San Francisco, California, United States
ABSTRACT
Background
Despite improvements in treatments and survival, there continue to be disparities in outcomes among children with acute leukemia. Socioeconomic disadvantages may contribute to these disparities by limiting access to healthcare resources and increasing barriers to navigating complex health systems. Few studies have directly evaluated caregiver-reported household income and barriers to care in the pediatric population. We evaluated associations of caregiver-reported annual household income and insurance status with barriers to care and markers of disease acuity at presentation. We hypothesize that lower socioeconomic status may be linked to greater barriers to care and disease acuity at presentation in children with acute leukemia.
Methods
We conducted a cross-sectional study of caregivers of pediatric patients aged 0–21 with acute lymphoblastic leukemia (ALL) or acute myeloid leukemia (AML) diagnosed between 2020 and 2026, treated at two children’s hospitals. Caregivers completed the validated Barriers to Care Questionnaire (BCQ), comprising five subscales: pragmatics, skills, expectations, marginalization, and knowledge and beliefs. Total and subscale scores were reverse-coded and rescaled to 0-100, with higher scores representing fewer perceived barriers. Laboratory markers at presentation, including thrombocytopenia, acute kidney injury (AKI), transaminitis, hyperbilirubinemia, and hyperuricemia, were retrospectively collected from the electronic health record.
Annual household income (<$50,000 vs. ≥$50,000) and insurance status (public vs. private) were dichotomized and evaluated as measures of SES. Multivariable linear regression assessed associations between SES and BCQ scores after adjusting for primary language, race and ethnicity, and primary care provider (PCP) status. PCP status is defined as having a source of usual primary care prior to diagnosis. Sensitivity analyses modeled household income as an ordinal variable with 6 categories (rather than binary). Exploratory analyses evaluated associations between SES measures and laboratory markers of disease acuity using prevalence ratios and 95% confidence intervals.
Results
Of the 79 caregivers who completed the BCQ survey, 72 (91%) reported annual household income. After adjustments, caregivers from households earning <$50,000 annually had significantly lower adjusted total BCQ scores than those from higher-income households (β = −10.70, 95% CI −21.35 to −0.06, p=0.049). In sensitivity analyses, treating income as a 6-category ordinal variable, each one-category increase in caregiver-reported household income was associated with higher Skills scores (β = 6.07, 95% CI 2.17–9.98, p=0.003), higher Knowledge & Beliefs scores (β = 6.13, 95% CI 1.66–10.60, p=0.008), and higher total BCQ scores (β = 4.37, 95% CI 0.82–7.92, p=0.017), demonstrating progressively fewer caregiver-reported barriers with increasing household income. Public insurance status was associated with lower Skills (81.3 vs. 90.6, p=0.014) and Knowledge & Beliefs (93.8 vs. 100, p=0.038) scores in unadjusted analyses. However, these associations were no longer significant after multivariable adjustment.
Exploratory analyses demonstrated higher prevalence of ≥1 laboratory abnormality among children from lower-income households (PR 2.05, 95% CI 0.88–4.79, p=0.111) and among public-insured children (PR 2.04, 95% CI 0.88–4.74; p=0.117). Children from households making <$50,000 also had higher prevalences of thrombocytopenia, AKI, transaminitis, hyperbilirubinemia, and hyperuricemia. However, these exploratory analyses were underpowered and did not reach statistical significance.
Conclusion
Lower household income was associated with greater caregiver-reported barriers to care, specifically in skills and knowledge and beliefs, with fewer barriers observed across increasing household income categories. Compared to insurance status, household income may be a stronger driver of barriers to care and unmet supportive care needs. Exploratory analyses also suggested a trend toward greater laboratory evidence of disease severity at presentation among children with lower SES. By collecting caregiver-reported SES and barriers to care, these findings support the hypothesis that SES is related to barriers to care and disparities in early disease presentation. Interventions aimed at mitigating barriers to care, such as caregiver navigation programs, may improve outcomes in pediatric acute leukemia.
DISCLOSURES
Savinnie Ho, BA: No relevant conflicts of interest to declare.