Citations

Abstract Background: Social determinants of health (SDOH) are critical drivers of cancer care access and outcomes. In multiple myeloma (MM), access to cellular therapies including autologous stem cell transplantation and CAR-T therapy often depends on formal or informal caregiver support. Traditional SDOH measures (e.g., race, income, or education) are often used as proxies for access to care; however, these measures may overlook individual realities pertaining to cellular therapies where a caregiver is generally required. In response to feedback from a HealthTree Patient Advisory Board questioning the practice of inferring caregiver support, this study evaluated a simple survey question by a large cross-sectional survey. Methods: After development of a dedicated single-question (“What best describes your current caregiving support?”) through iterative feedback from the Patient Advisory Board and MM experts, four key answers were settled on: (1) cohabitation with dedicated caregiver(s) to support care (“family support”); (2) presence of non-cohabiting caregiver(s) to provide dedicated support (“external support”); (3) lack of dedicated caregiver, hence support is arranged when needed (“arrangeable support ”); or (4) lack of dedicated caregiver, hence finding any support is a struggle (“inadequate support”). We deployed this new demographic question from February through July 2025 through HealthTree. Responses were analyzed across SDOH variables using Fisher exact or chi square tests, and random forest classifiers were employed for predictive modeling. Results: Of 1231 respondents, 71% had dedicated caregivers (68%, n=839, with family support, and 3%, n=41, with external support). In contrast, 23% (n=287) of patients routinely arranged support while 5% (n=64) had inadequate support (struggle to find support). The inadequate-support group had significant demographic differences from the sample as a whole: a higher proportion were women (78% vs 56%, p = 0.04), Black (9% vs 4%, p < 0.01), lived >50 miles from their treatment center (17% vs 6%, p < 0.01), and had an annual income <$50,000 (17% vs 6%, p = 0.02). This group had a numerically higher proportion of patients whose employment status was disabled (12% vs 5%, p=0.08) as well. No proportion difference in age (67 ± 9 yr), military status, or severity of diagnosis - precursor disease (17%) v. multiple myeloma (81%). Rates of inadequate support were overall low in patients who lived with only their spouse (1%) or lived with their spouse & children (3%). In contrast, rates of inadequate support were higher (than 5%) in patients who lived alone (21%), divorced (20%), lived with adult children (14%), lived with friends (13%), separated (11%), or widowed (9%) (p < 0.001). However, the arrangeable-support group comprised 12% of patients who lived only with their spouse, 19% of patients who lived with their spouse and children, and 61% of patients who lived with their adult children (p < 0.001). Finally, a random forest classifier failed to identify any individual in the inadequate-support category, suggesting that unmeasured or more complex patterns may underlie this classification. Conclusions: To our knowledge, this is the largest analysis of caregiver support and its relationship with other SDOH in patients with MM to date. Despite the importance of caregiving to transplantation and CAR-T therapy, we found that 5% of patients with MM struggle to find caregiver support; this figure likely underestimates the true prevalence, given potential selection bias among patients who are digitally engaged through the HealthTree platform. Even among patients who lived with their spouses, an additional 10-20% of patients routinely had to arrange caregiving support on a visit-by-visit basis; for patients who lived with their adult children, this proportion was over 60%. The requirement to arrange caregiving support can create time toxicity and financial toxicity for patients, although these factors were not directly queried in our study. Finally, caregiver availability cannot be routinely inferred from traditional SDOH variables. Our data highlight the importance of including caregiver status as a standard demographic question in clinical and research settings. Further interventions are needed to improve caregiver support for vulnerable patients with MM as well as to reduce the duration of caregiver requirements for procedures such as CAR-T therapy.

Abstract Background Accurate reconstruction of electronic health record (EHR) treatment sequences is essential for leveraging real-world data (RWD) in clinical research and trial-eligibility platforms in multiple myeloma (MM). However, inconsistencies in RWD treatment documentation and the complexity of line-of-treatment (LOT) assignments in MM hinder reproducibility and limit comparisons against published trials. To address these limitations, we developed HEAL-MM, an automated rule-based algorithm that systematically parses structured medication data to calculate LOTs and segment treatment into standardized, clinically meaningful phases. The algorithm is designed to be interpretable, adaptable to evolving therapeutic frameworks, and compatible with widely adopted EHR data standards. Here, we report its real-world validation against expert-adjudicated treatment sequences to assess concordance, identify edge-case failure points, and support RWD integration. Methods A stratified cohort of 100 MM patients (9,136 structured treatment events) was curated from 114 unique healthcare systems compiled from the multi-institutional HealthTree Foundation harmonized registry. Cohort selection was randomized to ensure coverage across diagnostic eras (pre-2010, 2015–2020, 2022–2025) and treatment exposures, including CAR T-cell therapies, bispecific antibodies, and anti-CD38-based regimens. HEAL-MM sequenced MM regimens chronologically and applied predefined logic rules to identify LOT transitions, the initiation of a new therapeutic course based on drug changes, treatment gaps, and procedural anchors such as stem cell transplant or CAR T. Within each LOT, the algorithm segments and labels therapy into phases, clinically distinct intervals (e.g., induction and maintenance) using rule-based clinical logic. Outputs were benchmarked against expert-labeled adjudications. Concordance was evaluated using concordance index (C-index), stratified by patient, LOT depth and diagnostic era. Results: HEAL-MM demonstrated high concordance with expert-labeled treatment sequences: 0.94 for LOT identification, 0.98 for treatment phase segmentation, and 0.96 for phase label assignment. Achieving complete agreement with all expert annotations in 90% of patients for LOT assignment, 94% for the number of treatment phases identified, 77% for the specific labels applied to each phase, and 75% across all three dimensions simultaneously. Among discordant patients, the absolute difference in final LOT count was 0.5 ± 0.9. Performance was highest in early lines (LOTs 1–3; C-index: 0.97), with seven patients showing discrepancies at this stage. Concordance declined in later lines (LOT ≥4; C-index: 0.87), where 50% of the discordant cases (n=4) were attributable to misclassifications in earlier LOTs, such as medication restarts following >3-month gaps, demonstrating the compounding effect of early errors on longitudinal accuracy. Temporal stratification revealed improved accuracy in post-2020 cases (C-index: 0.94) compared to 2015–2020 cohorts (0.91), reflecting the increasing adoption of structured treatment standards and EHR data quality improvements. To assess HEAL-MM performance at clinically significant junctures, we analyzed treatment transitions that initiated a new LOT or marked a change in phase, achieving C-indices of 0.97 for LOT transitions, 0.97 for treatment phase shifts, and 0.94 for phase label assignment. Conclusions: HEAL-MM provides a validated, scalable solution for standardizing MM treatment structures in RWD environments, as it accurately detects the exact treatment entry of a new LOT 97% of the time. This rule-based framework enables reproducible LOT and phase assignment across heterogeneous EHR systems, supporting real-time cohort identification across institutions and EHR platforms. The algorithm's modular design allows for continuous integration of novel therapeutic categories and evolving treatment logic. While the algorithm maintained high fidelity overall, early-stage misclassifications consistent with increased heterogeneity and documentation ambiguity occasionally propagated downstream, underscoring the compounding effect of upstream errors in longitudinal reconstruction. With its generalizable structure and compatibility with industry-standard data models, HEAL-MM lays the foundation for structured treatment reconstruction in MM across RWD infrastructure and serves as a transferable blueprint for other hematologic malignancies.