Socioeconomic inequality in acute myeloid leukaemia treatment in England: What drives this inequality Free

Introduction: Socioeconomic inequalities in survival of acute myeloid leukaemia (AML) exist. In a private healthcare context such as the United States, socioeconomic factors such as household income, marital status, type of insurance and race/ethnicity are known to be associated with differential treatment patterns and survival among patients with AML. Few studies from countries with universal healthcare coverage have looked at differential AML treatment patterns based on patients' socioeconomic status. We aimed to quantify socioeconomic inequalities in access to intensive chemotherapy (ICT) among patients with AML in England, and understand what may drive such inequalities.

Methods: The national, population-based cancer registry of England was used to identify patients with AML. Linkage with the national chemotherapy dataset (SACT) and hospital inpatient/outpatient records (HES) provided information on access to treatment. The study included patients aged 25 to 80 years, diagnosed with de novo AML in England between January 1, 2015 to December 31, 2022. All patients had a minimum potential follow-up of one year from the date of AML diagnosis. Patients with preceding haematological conditions (e.g. myelodysplastic syndrome) and those with previous exposure to chemotherapy or radiotherapy were excluded. The explanatory variable was deprivation, derived from the Income Domain of the Index of Multiple Deprivation, which measured relative deprivation at a small-area level across England. Deprivation was grouped into quintiles, from Q1 (least deprived) to Q5 (most deprived). The outcome variable was receipt of ICT. Confounding variables included age at diagnosis, sex, and comorbidities. The analyses adjusted for patient characteristics and the hospital (National Health Service (NHS) Trust) where patients began treatment. The likelihood of receiving ICT across deprivation levels was first estimated using a generalised estimating equation. Then, a mixed-effects probit regression model was used to predict the number of patients who might have received ICT, had they been in Q1, while keeping their original other characteristics (i.e., age, sex, comorbidities, and NHS Trust allocation).

Results: Between 2015 and 2022, 11,125 patients aged 25–80 were diagnosed with de novo AML in England. The median age at diagnosis was 67 years, 57% were male, and 32% had at least one comorbidity.

Treatment data were available for 94% of the cohort. Overall, 52% received ICT, with a decline by age at diagnosis, from 82% among those aged 25–59 years, to 64% in ages 60–69 years, and 19% in ages 70–80. Patients in Q5 were 4.2% less likely to receive ICT than those in Q1 (Risk difference: –4.2%, 95% confidence interval: –6.9% to –1.6%). The inequalities between Q5 and Q1 persisted before and after the coronavirus disease-19. The adjusted probability of receiving ICT for male patients without comorbidities differed for Q5 and Q1 in ages between 51 to 72, with the largest inequality observed at age 67. The difference in the probability of ICT between Q5 and Q1 also varied across NHS Trusts. If we assumed that all patients had access to ICT similarly to Q1, while keeping the rest of the patient characteristics unchanged, the model predicted that 240 more patients would potentially receive ICT. However, once we accounted for the effect of NHS Trust allocation, the model predicted that only 17 more patients would receive ICT.

Conclusion: The analysis of the England national, population-based data suggested that more deprived patients had a lower chance of receiving ICT than less deprived patients, even after adjusting for the confounding effects of age, sex and comorbidities. The effect of deprivation on ICT use was attenuated after adjusting for both patient and NHS Trust allocations. This suggests that the NHS Trust characteristics such as institutional culture, workforce capacity, and resource availability, may be the key drivers of the deprivation-associated inequality in AML treatment, rather than the deprivation itself.