No abstract

BACKGROUND: The impact of treatment delays on cancer outcomes has been well documented, but few published studies explicitly attempt to identify subgroups which would benefit the most from potential interventions improving timeliness of care. Using the state of North Carolina and breast cancer as an example, we sought to quantify race-specific associations between frontline treatment delay and breast cancer mortality, and estimate the potential impact of improvements in timeliness by racial sub-group. METHODS: We conducted a retrospective cohort study, utilizing multipayer insurance claims linked to cancer registry data from the Cancer Information and Population Health Resource, of females diagnosed with stage I-III breast cancer from 2004 to 2014. Our exposure of interest was treatment delay, defined as initial cancer directed therapy received ≥ 60 days after diagnosis. Our outcome was death certificate identified breast cancer mortality within 5-years of follow-up. To quantify the association between treatment delay and mortality we estimated hazard ratios (HR) and 95% confidence limits (CL) where non-breast cancer deaths were treated as competing events. We then simulated the effect of treating all patients within 60 days using inverse probability of treatment weighting (IPTW), and estimated the potential impact by comparing the simulated data with observed data. Models were adjusted for age and stage at diagnosis, tumor grade, hormone receptor status, and modality of first treatment. Models were stratified by race, dichotomized as Black or non-Black, and geographic subregion for evaluation. RESULTS: Our analytic cohort comprised 21,200 patients, of whom 19% were Black. Treatment delays >60 days were twice as frequent among Black compared with non-Black patients (13.5% vs. 6.5%). We found a positive association between treatment delay and breast cancer mortality in the overall cohort (HR=1.3; CL: 1.1, 1.6). The association differed between Black (HR=1.4; CL: 1.1, 1.8) and Non-Black (HR=1.2; CL: 0.9, 1.6) patient subgroups. For both racial groups, the strength of the relationship between treatment delay and mortality varied across geographic subregions. The estimated potential improvement in 5-year cumulative breast cancer mortality from delivering timely care to all patients differed in magnitude across sub-populations. The magnitude of potential improvement in breast cancer mortality at 5 years in the overall analytic cohort was 0.3% (5-year risk of breast cancer mortality in observed cohort: 7.2% vs 6.9% with simulated universal timely treatment). In the Black subcohort, the potential decrease in mortality for timely treatment was 1.0% (observed: 12.3% vs simulated: 11.4%), while in the non-Black population the potential decrease was 0.1% (observed: 6.1% vs simulated: 5.9%). CONCLUSIONS: The magnitude of association between delayed treatment of stage I-III breast cancer and breast cancer mortality differed between Black and non-Black patients, highlighting that interventions to avert treatment delays may have a meaningful impact for this group. Further research is needed to identify factors underlying this difference, which may include a direct effect of differences in timeliness, or downstream differences in the intensity and quality of cancer care beyond the 60-day landmark for delayed patients. Our findings suggest that comprehensive intervention at the state level to improve the timeliness of breast cancer treatment could potentially reduce breast cancer mortality, and that targeting such intervention to patient groups with larger projected benefit, such as Black patients, may be a more efficient use of resources. We have illustrated how a counterfactual approach may be useful in identifying subgroups where focused intervention efforts may yield greater improvements in outcomes. Citation Format: Bradford Jackson, Stephanie Wheeler, Juan Yanguela, Matthew LeBlanc, Tzy-Mey Kuo, Christopher Baggett, Katherine Reeder-Hayes. IMPACTS OF TREATMENT DELAY ON BREAST CANCER MORTALITY AND BENEFIT OF TIMELY CARE IN BLACK AND NON-BLACK WOMEN [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P1-06-04.

The Affordable Care Act has significantly reduced the number of uninsured individuals in the U.S., yet millions of insured Americans lack adequate coverage or face substantial health-related financial risks. In this study, we conduct an analysis of a nationally representative sample to generate up-to-date evidence on the association between insurance types and demographic factors and the probability of cost-related underinsurance among continuously insured, lower-income adults in the U.S. We extend the previously established definition of underinsurance by including a measure of subjective financial burden. Our estimates suggest that 33.2% of lower-income individuals in our sample were underinsured, with a fourth of them meeting two or three criteria. We also find that Medicaid beneficiaries were significantly less likely than others to experience underinsurance. Our findings provide evidence that underinsurance is a major concern, and expansion of coverage alone may not suffice to protect individuals from financial risks and cost barriers to health care use.

6549 Background: While we have seen impressive improvements in breast cancer (BC) survival over the last few decades, substantial racial disparities persist. Inequities in the receipt of guideline-concordant treatment contribute to worse survival in Black BC patients. Previous studies have shown that patient-level disparity reduction interventions (e.g., patient navigation) can significantly increase guideline-concordant treatment receipt among Black patients, with positive spillover effects in nonBlack patients. The aim of this study was to estimate the potential impact of disparity reduction interventions on racial inequities in survival after BC diagnosis in North Carolina. Methods: We used data from the Cancer Information and Population Health Resource, which links multipayer claims to North Carolina’s cancer registry for biological women diagnosed with BC in 2004-2017. We calculated Black/nonBlack disparities in the receipt of chemotherapy (CTx) within 4 months of diagnosis for patients ages <70 with hormone-receptor negative stage Ib-III BC (CTx cohort; N=2223) and the receipt of endocrine therapy (ET) within 12 months of diagnosis for stage I-III hormone receptor-positive BC (ET cohort=16220). We then simulated the potential increase in the proportion of patients receiving CTx and ET if proven patient-level disparity-reduction interventions were implemented across the state. Based on the literature, we assumed that the effects of these interventions in increasing treatment receipt would be same in the CTx and ET cohorts. We estimated the effect of this potential increase in CTx and ET receipt on 10-year overall survival using cohort-, race-, and treatment receipt-stratified Markov models. We report confidence bounds representing 95% of simulation results. Results: Over the 2004-2017 period, 72.9% and 70.8% of Black patients in our cohorts received CTx and ET, respectively. This was significantly lower than among nonBlack patients (p<.05). State-wide implementation of disparity reduction interventions could increase CTx and ET receipt among Black patients to 86.7% (80.9-92.5%) and 84.2% (78.5-89.9%), respectively. As a result, the racial 10-year survival gap would decrease from 9.4 to 8.1 (5.7-10.5) percentage points in the CTx cohort, and from 6.2 to 5.4 (4.0-6.7) percentage points in the ET cohort. Conclusions: Patient-level interventions can reduce disparities in guideline-concordant-treatment receipt. However, increases in treatment receipt alone would not be enough to close the racial survival gap, suggesting that, in isolation, patient-level interventions are unable to eliminate disparities. Action on the structural factors that disadvantage Black BC patients is needed. Our team is conducting community-engaged work to share and interpret these findings.