Deep learning and, specifically, convoltional neural networks (CNN) represent a class of powerful models that facilitate the understanding of many problems in computer vision. When combined with a reasonable amount of data, CNNs can outperform traditional models for many tasks, including image classification. In this work, we utilize these powerful tools with imagery data collected through Google Street View images to perform virtual audits of neighborhood characteristics. We further investigate different architectures for chronic disease prevalence regression through networks that are applied to sets of images rather than single images. We show quantitative results and demonstrate that our proposed architectures outperform the traditional regression approaches.

Multiple myeloma (MM) is a common hematologic malignancy, but due to its incurable nature, patients experience many relapses in their lifetime and hence face unique challenges. This mixed-methods study consisting of an online survey and subsequent focus groups aimed to understand how social and identity experiences affected the diagnostic, treatment, and care journey for patients with MM. Twenty-three adult patients with MM participated in this study. The survey participants identified common determinants negatively impacting their health, including mental health concerns (experienced by 90.5% of respondents), worries about food shortage (42.9%), and transportation concerns (28.6%). Focus group participants described high physical and mental health burdens associated with MM. Frequent monitoring, fear of a relapse, and unpredictable side effects contributed to high anxiety. Participants indicated that MM differed from other types of cancer and chronic health conditions in many ways, particularly how and where the diagnosis was made, disease progression and relapse, treatments and side effects, and financial concerns. Most participants (65.0%) reported ≥1 social need that negatively impacted health outcomes including lack of knowledge about MM, financial instability, and lack of insurance, transportation, and social support. The findings reveal that patients with MM continually experience patient-specific mental and physical health burdens indicating high unmet needs throughout the disease journey.

Patients living with multiple myeloma (MM) have a substantial disease burden and face multiple barriers to care. Building upon our previous research using mixed methods, this focus group research aimed to identify patients’ priorities regarding specific social and identity-related needs, map these prioritized needs to the disease journey, and describe patient-generated ideas to improve patient support. Participants noted that patients with MM need a range of emotional, social, and financial support throughout the disease journey. They identified initial MM diagnosis and treatment adherence as two critical points in the MM journey where patients need the most support and assistance. The findings of this research suggest that overall, patients with MM need comprehensive support, ideally from a multidisciplinary team consisting of health care providers, patient advocates, social workers, and psychologists to help patients understand their disease and treatment options, make informed treatment decisions, adhere to treatment, and ultimately reduce their disease burden and improve outcomes. This research revealed that patients with MM need varying types and levels of support, with the most common needs including information on disease and treatment, connections to financial resources and support systems, assistance with navigating insurance options, and transportation and logistical support for medical appointments.

To improve health inequities, it is necessary to understand the impact of social determinants of health (SDoH), or social risk factors, including race and economic status, on multiple myeloma (MM) treatment patterns. To identify SDoH factors leading to gaps in frontline treatment in Medicare beneficiaries with newly diagnosed MM (NDMM). This retrospective study used data from various sources, including claims data, individual-level SDoH measures (eg, race, dual-eligibility [DE] status for Medicare and Medicaid, low-income subsidy [LIS] status, and special needs plan eligibility) from the Humana Research database population during the time frame from 2016 to 2023, and community-level SDoH measures from the Agency for Healthcare Research Quality database. Treatment pattern outcomes included treatment within 90 days of first MM diagnosis, time from diagnosis to frontline treatment, frontline treatment regimen type, daratumumab-containing frontline regimens, and duration of frontline therapy. Multivariable regression was used to evaluate the association between SDoH factors and MM treatment patterns. Of 4,483 individuals identified with NDMM, 31.9% were Black race and 24.1% had DE/LIS status. More than half of individuals in the study resided in areas that were above the national median for receiving public assistance, having less than high school education, having no health insurance, and having no Internet. In the overall cohort, 1,941 (43.3%) patients had no treatment within 12 months of diagnosis, 811 of whom had no evidence of symptomatic disease (ie, asymptomatic smoldering MM). Median time to treatment initiation (TTI) from diagnosis was 2.7 months, and 51.2% of patients received treatment within 90 days of diagnosis. Lower odds for treatment initiation within 90 days were observed for Black patients (vs White patients; odds ratio [OR] = 0.865 [CI = 0.752-0.995]), DE/LIS patients (vs non-DE/LIS; OR = 0.696 [CI = 0.599-0.809]), and by special needs plan enrollment (vs nonenrollment; OR = 0.717 [CI = 0.547-0.940]), but community-level SDoH was generally not independently associated with TTI. Among 2,523 patients who received frontline treatment within 12 months of diagnosis (treated cohort), TTI and duration of treatment were similar between the overall cohort and DE/LIS and non-White subgroups. Secular trends were observed in frontline treatment regimens, which were mostly triplets, and evolved over time to comprise fewer doublet regimens and more quadruplets, with an increase in daratumumab-based regimens. Inequities in timely frontline NDMM treatment were observed for non-White patients and those with DE/LIS status. Combinations of community-level SDoH, but no one single factor, may underlie these inequities.

Introduction: Patient perspectives and preferences are important when establishing quality and cost metrics for value-based payment (VBP) models, to ensure patient-centered cancer care and VBP incentives are aligned. Methods: Insights were gathered about value-based care (VBC) models from patients with Medicare and one of 4 common cancers (multiple myeloma, bladder, lung, and prostate) included in Medicare's Enhancing Oncology Model. This study included 4 virtual 2-hour focus groups and a 2-hour co-creation session. Aspects of cost and quality important to cancer care were identified (value is defined as the ratio of quality/cost). Results: Insights from the focus groups (n = 20) and co-creation session (n = 6) highlighted a lack of awareness about VBC and concerns about the effect of financial incentives on oncology care. Focus group participants identified shared decision-making and treatment effectiveness outcomes as the most important elements of quality. Co-creation participants highlighted a need for improved awareness about the implications of VBP incentives for oncology care decisions. Conclusion: In this exploratory qualitative research, patients with cancer expressed a need for transparency of VBC dynamics to help preserve the patient–physician relationship and their ability to receive innovative, affordable, and accessible care.