Underserved Patients Research Articles

Telehealth utilization barriers among Alabama parents of pediatric patients during COVID-19 outbreak

BackgroundTelehealth can improve access to evidence-based care at a lower cost for patients, especially those living in underserved and remote areas. The barriers to the widespread adoption of telehealth have been well documented in the literature. However, the barriers may not be the same for pediatric patients, who must rely on their parents or guardians to make healthcare decisions. This paper presents some of the leading barriers parents or guardians of pediatric patients report in using telehealth to meet their children’s healthcare needs.MethodsThis cross-sectional survey was conducted in a tertiary care pediatric Emergency Department (ED) at a children’s hospital in Alabama between September 2020 to December 2020. The parents or guardians of pediatric patients were asked about their reasons for not using telehealth despite having healthcare needs for their children, whether they canceled or rescheduled healthcare provider visits and facility visits, and whether the child’s health conditions changed over the past three months. Descriptive analyses were conducted that explored the distribution of telehealth use across the variables listed above.ResultsFive hundred ninety-seven parents or guardians of pediatric patients participated in the survey, and 578 answered the question of whether they used telehealth or not over the past three months. Of them, 33.1% used telehealth, 54.3% did not, and 12.6% did not have healthcare needs for their child. The leading reason for not using telehealth was that the doctor or health provider did not give them a telehealth option, the second main reason was that they did not know what telehealth is, and the third leading reason was that the parents did not think telehealth would help meet healthcare needs for their child.ConclusionsThis study highlights the telehealth utilization barriers among underserved pediatric populations, including the need for physicians to proactively offer telehealth options to parents or guardians of pediatric patients. Improving health literacy is of paramount importance, given that a substantial proportion of parents were not familiar with telehealth. Policymakers and healthcare organizations should raise awareness about the benefits of telehealth which can improve healthcare access for underserved pediatric patients.

153-LB: Early Impact of a Hybrid Chronic Care Model on A1C Levels among Underserved Patients with Diabetes in Primary Care

Underserved populations are at higher risk of developing diabetes-related complications. This study aimed to evaluate changes in A1c among undeserved patients who received care through a hybrid Chronic Care Model (h-CCM). This was a retrospective observational study conducted in two satellite clinics of a federally qualified health center. h-CCM was delivered collaboratively via in-person and telehealth visits conducted by family physicians, nurse practitioners and clinical pharmacists. Baseline sociodemographic of patients >40 years old with diabetes who received care through h-CCM for ≥ 3 months between Oct 2020 and Nov 2022 were included. Patients with any missing data were excluded. Descriptive analysis and paired t-test were used accordingly. A total of 104 (53.0%) patients were analyzed. The average age was 60.7± 9.8 years with the majority being Hispanic (72.1%), female (57.7%) and with a mean h-CCM visits of 2.2± 1.8 over 3 months. The average ASCVD risk at baseline was 20.0± 16.6% with hypertension (88.5%) being the most common comorbidity. Average A1c decreased from 9.4± 2.1% at baseline to 8.7±1.9 at 3 months (-0.7%, p <0.001) with improvements observed in all levels of baseline A1c (p<0.001, Figure 1). h-CMM was effective in achieving timely A1c reduction among underserved patients with diabetes. Greater improvements were observed in patients with higher baseline A1c. Disclosure J. Y. Lee: None. J. T. Nguyen: None. M. Seo: None. T. Nguyen: None. N. Patel: None. V. Rodriguez: None. J. Mayorga: None.

700-P: Outcomes of a Diabetes Support Coach Intervention Implemented during the COVID-19 Pandemic

During the COVID-19 pandemic, people with diabetes reported increased diabetes distress and difficulty maintaining glycemic control. These challenges were exacerbated for racial minorities and those with low socioeconomic status. As part of a Project Extension for Community Healthcare Outcomes (ECHO) program, an intervention using 28 diabetes support coaches (DSCs), adults with diabetes who served as peer mentors and healthcare navigators, was implemented with underserved patients with type 1, type 2, and other forms of diabetes at 41 federally qualified health centers in FL and CA (intervention n=74, control n=363). Patients self-enrolled and engaged with DSCs via 1-1 weekly text messages, phone calls, or virtual meetings. Patients’ HbA1c was measured at baseline and 6-month follow-up. The Diabetes Distress Scale (DDS-17), Coronavirus Anxiety Scale (CAS-19), and a survey of diabetes technology use and COVID-19 outcomes were also administered. Descriptive statistics were computed for all metrics (n, %; median IQR for skewed data). Between-group comparisons were evaluated via Chi-Square tests for categorical data and t-tests for continuous data. Diabetes distress was significantly lower among the intervention group [DDS-17 score median=1.6] than controls [DDS-17 score median=2.1] (p=.0077). More patients in the intervention regularly used a continuous glucose monitor (CGM) (69.9% vs. controls: 38.8%, p<.001). There were no significant differences between groups in insulin pump use, HbA1c, COVID-19 anxiety, or other COVID-19 outcomes. The latter results were likely due to the COVID-19 survey evaluating structural outcomes like financial stability that were beyond the scope of the DSC intervention. Overall findings demonstrate that DSCs reduced diabetes distress and improved CGM use among underserved patients with diabetes. Support coaching is valuable for this population; further research is needed to improve their HbA1c levels and COVID-19 outcomes. Disclosure J.Maizel: None. S.C.Westen: None. K.K.Hood: Consultant; Cecelia Health. J.J.Wong: None. B.Dixon: None. S.L.Filipp: None. M.J.Gurka: None. A.F.Walker: None. M.J.Haller: Board Member; SAB Biotherapeutics, Inc., Consultant; Sanofi, MannKind Corporation. D.M.Maahs: Advisory Panel; Medtronic, LifeScan Diabetes Institute, MannKind Corporation, Consultant; Abbott, Research Support; Dexcom, Inc. A.Addala: None. R.Lal: Advisory Panel; Provention Bio, Inc., Consultant; Abbott Diabetes, Biolinq, Capillary Biomedical, Inc., Deep Valley Labs, Gluroo, Tidepool, PhysioLogic Devices, Morgan Stanley. L.Figg: None. M.Hechavarria: None. K.G.Malden: None. E.Sheehan: None. Funding The Leona M. and Harry B. Helmsley Charitable Trust (G-2005-03934)

1034-P: Access to Continuous Glucose Monitoring Systems in a Primary Care Clinic for Underserved Patients with Type 2 Diabetes—A Pilot Study

B-Disparities in diabetes care are well documented. Continuous Glucose Monitoring Systems (CGMs) have improved patient experience without the need for finger sticks and have been shown to effectively lower Hemoglobin A1c (HbA1c), decrease clinical inertia and decrease the number of hypoglycemic events. Medically underserved and historically marginalized persons are disproportionately affected by diabetes, yet limited data exist on use of CGMs in this population. M-We report a pilot initiative that brought CGM access (FreeStyle Libre 2) to a Federally Qualified Community Health Center (FQHC). Patients with type 2 diabetes and an HBA1c >8.0% were referred to be enrolled in CGM. Each patient had an initial nurse visit for device placement of FreeStyle Libre (FSL) and several follow up visits to review glycemic trends and make medication changes with a provider. Baseline HBA1c was collected and repeated at the end of the study. Provider and patient surveys were distributed at each visit. R-Fifty participants were enrolled in the CGM study and had FSL placed. The CGM active time was greater than 50% in 76% of participants and those who continued with CGM had an increase in active time. Initial mean A1C for the entire cohort was 9.64% and average glucose after 2 weeks was 195 mg/dl. Hypoglycemia, especially overnight, reduced overtime. The use of CGM data was associated with clinician-initiated medication changes. Provider surveys revealed an increase in likelihood of medication changes based on glycemic trends and patients reported better understanding of their disease process. C-This pilot study demonstrates the feasibility of utilizing diabetes technology (FreeStyle Libre 2) in underserved patients at a primary care clinic staffed with internal medicine residents, physicians, and nurse practitioners. CGM is beneficial in under resourced primary care outpatient settings to determine glycemic trends and make appropriate medication recommendations. Disclosure R.G.Sgroi: None. S.Gumireddy: None. W.Tan: None. R.Kaplan: None. Y.Milambwe: None. L.H.Rosoph: None. D.A.Salama: None. C.R.Feher: None. N.J.Rennert: None.

Healthcare Professionals’ Emotional Labor and Management of Workplace Violence with Underserved Patients in the Safety Net Context

ABSTRACT Healthcare providers (HP) work in high-stress situations, interacting with patients and families who are often in crisis. HPs who work in safety net clinics, which provide care to uninsured, Medicaid recipients and other vulnerable populations, interact with patients who are frequently frustrated by long wait times, extensive paperwork, short appointments, and have generally lower health literacy. Many patients have chronic conditions and substance use disorders which has been associated with higher likelihood to be perceived as verbally aggressive and/or perpetrate workplace violence (WPV). Using interviews with 26 HPs at safety net clinics, we investigated how HPs manage interactions with aggressive patients and avoid burnout. Findings are based on emotional labor constructs describing why and how workers use emotion management strategies to smooth communication and relationships with clients/patients. According to our participants, HPs perform emotional labor to de-escalate interactions, prevent WPV, and to develop relationships with patients who might become regular clinic patients. We found that HPs perceive an influence of the clinic context on patient aggression management, hold initial perceptions that shape engagement with aggressive patients, and report emotional labor and burnout that came from interacting with aggressive patients to prevent WPV. We offer implications that extend research on emotional labor and burnout, provide guidance to healthcare organizations, and offer directions for future theory and research.

Reality of Family Caregiving for Hispanics With Alzheimer's Disease and Related Dementias: A Qualitative Analysis.

Background: Underserved Hispanic patients often experience unmet palliative care (PC) needs, particularly those with noncancer diagnoses such as Alzheimer's disease and related dementias. Most caregivers for Hispanic patients are family relatives who are less likely to use health care and community resources and experience high caregiver burden. We adapted a culturally tailored patient navigator (PN) intervention to provide support and improve PC outcomes for Hispanics with Alzheimer's disease and related dementias and their family caregivers (FCGs). Objectives: To explore Hispanic FCGs' experiences and perceptions of caregiving for a loved one, and how our PN intervention impacted their needs. Design: Qualitative descriptive. Setting/Subjects: FCG participants (n = 10) from our randomized control trial's intervention group were recruited from academic and safety net hospitals and community-based clinics across urban and rural Colorado in the United States. Measurements: Data obtained from individual, semistructured, 30-minute telephone interviews were recorded, transcribed, translated, and analyzed using NVivo and qualitative thematic analyses. Results: Four major themes emerged: Methods of Support, Cultural Expectations and Varying Family Contributions, Lack of Self-Care, and Awareness. Subthemes highlighted differing definitions of "contributing," role resentment, and interpersonal issues. Varying familial expectations underscore FCG strain when the burden of caregiving is not shared. Participants used various coping strategies as necessary support and gained awareness through education, guidance, and referrals to resources. Conclusions: PNs helped FCGs and patients beyond the intervention's scope. Providing support and awareness to FCGs, and incorporating cultural beliefs, may improve PC access to disparate populations and guide future interventions. Clinical Trial Registration Number NCT03181750.

Collaborative depression care sensitive to the needs of underserved patients with cancer: Feasibility, acceptability and outcomes

Purpose A single-arm trial evaluated the feasibility, acceptability, and outcomes of COPE-D, a collaborative care intervention for underserved cancer patients with depression. Methods Bilingual (Spanish and English) care managers provided counseling and/or medication management in consultation with physicians. Outcomes were treatment improvement (≥ 5-point reduction in PHQ-9), treatment response (≥ 50% reduction in PHQ-9), suicidal ideation resolution, and changes in depression (PHQ-9), anxiety (GAD-2), sleep disturbance (PSQI), global mental and physical health (PROMIS), social isolation (PROMIS), and qualitative feedback. Results 193 patients consented to participate. 165 initiated and 141 completed treatment, with 65% and 56% achieving treatment improvement and response, respectively. Outcomes did not differ by ethnicity (31% Hispanic), cancer stage (71% stages III-IV), income, or education. Suicidal ideation, depression, anxiety, sleep disturbance, and social isolation also improved. Qualitative feedback was largely positive. Conclusion COPE-D improved depression and quality of life among underserved patients, with acceptable retention rates.

Requirement for Discharge in the Care of a Responsible Adult in Procedural Sedation in the Emergency Department: Necessity or Potential Barrier to Health Equity?

BackgroundProcedural sedation is commonly practiced by emergency physicians to facilitate patient care in the emergency department (ED). Although various guidelines have modernized our approach to procedural sedation, many procedural sedation guidelines and practices still often require that patients be discharged into the care of a responsible adult. DiscussionSuch requirement for discharge often cannot be met by underserved and undomiciled patients. Benzodiazepines, opioids, propofol, ketamine, “ketofol,” etomidate, and methohexital have all been utilized for procedural sedation in the ED. For patients who may require discharge without the presence of an accompanying responsible adult, ketamine, propofol, methohexital, “ketofol,” and etomidate are ideal agents for procedural sedation given rapid onsets, short durations of action, and rapid recovery times in patients without renal or hepatic impairment. Proper pre- and postprocedure protocols should be utilized when performing procedural sedation to ensure patient safety. Through the use of appropriate medications and observation protocols, patients can safely be discharged 2 to 4 h postprocedure. ConclusionThere is no pharmacodynamic or pharmacokinetic basis to require discharge in the care of a responsible adult after procedural sedation. Thoughtful medication selection and the use of evidence-based pre- and postprocedure protocols in the ED can help circumvent this requirement, which likely disproportionally impacts patients who are of low socioeconomic status or undomiciled.

CDK2 regulates collapsed replication fork repair in CCNE1-amplified ovarian cancer cells via homologous recombination.

CCNE1 amplification is a common alteration in high-grade serous ovarian cancer and occurs in 15-20% of these tumors. These amplifications are mutually exclusive with homologous recombination deficiency, and, as they have intact homologous recombination, are intrinsically resistant to poly (ADP-ribose) polymerase inhibitors or chemotherapy agents. Understanding the molecular mechanisms that lead to this mutual exclusivity may reveal therapeutic vulnerabilities that could be leveraged in the clinic in this still underserved patient population. Here, we demonstrate that CCNE1-amplified high-grade serous ovarian cancer cells rely on homologous recombination to repair collapsed replication forks. Cyclin-dependent kinase 2, the canonical partner of cyclin E1, uniquely regulates homologous recombination in this genetic context, and as such cyclin-dependent kinase 2 inhibition synergizes with DNA damaging agents in vitro and in vivo. We demonstrate that combining a selective cyclin-dependent kinase 2 inhibitor with a DNA damaging agent could be a powerful tool in the clinic for high-grade serous ovarian cancer.

Factors Associated With Liver Transplant Referral Among Patients With Cirrhosis at Multiple Safety-Net Hospitals

A high proportion of underserved patients with cirrhosis receive care at safety-net hospitals (SNHs). While liver transplant (LT) can be a life-saving treatment for cirrhosis, data on referral patterns from SNHs to LT centers are lacking. To identify factors associated with LT referral within the SNH context. This retrospective cohort study included 521 adult patients with cirrhosis and model for end-stage liver disease-sodium (MELD-Na) scores of 15 or greater. Participants received outpatient hepatology care at 3 SNHs between January 1, 2016, and December 31, 2017, with end of follow-up on May 1, 2022. Patient demographic characteristics, socioeconomic status, and liver disease factors. Primary outcome was referral for LT. Descriptive statistics were used to describe patient characteristics. Multivariable logistic regression was performed to evaluate factors associated with LT referral. Multiple chained imputation was used to address missing values. Of 521 patients, 365 (70.1%) were men, the median age was 60 (IQR, 52-66) years, most (311 [59.7%]) were Hispanic or Latinx, 338 (64.9%) had Medicaid insurance, and 427 (82.0%) had a history of alcohol use (127 [24.4%] current vs 300 [57.6%] prior). The most common liver disease etiology was alcohol associated liver disease (280 [53.7%]), followed by hepatitis C virus infection (141 [27.1%]). Median MELD-Na score was 19 (IQR, 16-22). One hundred forty-five patients (27.8%) were referred for LT. Of these, 51 (35.2%) were wait-listed, and 28 (19.3%) underwent LT. In a multivariable model, male sex (adjusted odds ratio [AOR], 0.50 [95% CI, 0.31-0.81]), Black race vs Hispanic or Latinx ethnicity (AOR, 0.19 [95% CI, 0.04-0.89]), uninsured status (AOR, 0.40 [95% CI, 0.18-0.89]), and hospital site (AOR, 0.40 [95% CI, 0.18-0.87]) were associated with lower odds of being referred. Reasons for not being referred (n = 376) included active alcohol use and/or limited sobriety (123 [32.7%]), insurance issues (80 [21.3%]), lack of social support (15 [4.0%]), undocumented status (7 [1.9%]), and unstable housing (6 [1.6%]). In this cohort study of SNHs, less than one-third of patients with cirrhosis and MELD-Na scores of 15 or greater were referred for LT. The identified sociodemographic factors negatively associated with LT referral highlight potential intervention targets and opportunities to standardize LT referral practices to increase access to life-saving transplant among underserved patients.

Improving the treatment of functional seizures through a public specialist outpatient clinic

ObjectivesWe performed an audit of the first 12 months of clinical operations to assess the feasibility of a newly established public outpatient clinic for the assessment and treatment of functional (psychogenic nonepileptic) seizures (FS). MethodClinical notes for the first 12 months of the FSclinic weresystematicallyreviewed with data compiled onreferral pathways, clinic attendance, clinical features, treatments, and outcomes. ResultsOf eighty-two new FS patients referred to the clinic, over 90% attended. Patients were diagnosed with FS after comprehensive epileptological and neuropsychiatric review, mostly with typical seizure-like episodes captured during video-EEG monitoring, and most accepted the diagnosis. Most had FS at least weekly, with little sense of control and significant impairment. The majority of individuals had significant psychiatric and medical comorbidity. Predisposing, precipitating, and perpetuating factors were readily identified in >90% of cases. Of 52 patients with follow-up data within12 months, 88% were either stable or improved in terms of the control of their FS. ConclusionThe Alfred functional seizure clinic model, the first dedicated public outpatient clinic for FS in Australia, provides a feasible and potentially effective treatment pathway for this underserved and disabled patient group.

P9: Teaching Health Disparities as Regards Cardiopulmonary Support Technologies for Underepresented/ Underserved Patients

Background: There is a growing interest in addressing health care disparities faced by underrepresented/ underserved (U/U) populations across all fields of medicine. To that end, my goal was to introduce to the undergraduate (UG) and graduate (G) bioengineering classes that I teach the challenges faced by U/U patients as regards state-of-the art cardiopulonary support technologies (CPST) for treatment of end-stage cardiac or pulmonary failure. Methods: At the University of Pittsburgh (UoP) I teach two separate UG artificial organs (AO) classes; one class focuses on cardiac assist devices and the other on pulmonary assist devices. I also teach the cardiac assist module for a G cardiovasuclar organ replacement class. Through a long-standing NSF-funded collaboration with North Carolina Agricultural and Technical State University (Greensboro, NC), a Historically Black College and University (HBCU), I also teach my AO classes to NCAT students. The NCAT courses are taught remotely (via ZOOM). For the past several years, the term projects for my UG and G artificial organs classes at UoP and NCAT have focused on i) investigating health inequities/ disparities in CPST as applied to U/U patients; and, ii) proposing innovative improvements to current CPST that benefit, particularly, U/U patients. Students are asked to address the following in their term papers and class PPT presentations: 1) current state-of-the art in CPST; 2) limitations of current CPST as applied to U/U patients; 3) unmet clinical needs for U/U patients; 4) the feasibility of the proposed innovative improvements from a product specification, regulatory and societal viewpoint. Results: Students identified numerous reference materials that highlight the inequities/ disparities that U/U patients face. The American Heart Association, the CDC and NHLBI all maintain detailed epidemiological information related to the term project topic. There also exists a plethora of peer-reviewed published articles in this regard; a few examples being, Frech et al, “Ventricular Assist Device Technology and Black-White Disparities on the Heart Transplant Wait List.” Progress in Transplantation 2021; 31(1): 80-87; Nyak et al, “Understanding the Complexity of Heart Failure Risk and Treatment in Black Patients.” Circulation Heart Failure 2020; 13: 301-314. Among the student term project topics were, “Improving VAD Technology for the African American Community;” “Blood Pressure Measurement and Management in Black Ventricular Assist Device Users;” “A Multi-Factor Approach to Improve African American Heart Health;” “Innovative Design for Cystic Fibrosis in Women of Color.” Conclusions: Student response to the term project assignment was positive. Meetings with the students demonstrated the students’ interest in the topic. The ultimate success of this assignment will be achieved, if these students are sensitive to the special needs of U/U patients and incorporate these needs in their design of biomedical devices.

Turnaround time of tissue next generation sequencing in cancer patients at a safety net hospital: A quality improvement project.

e18707 Background: Tissue Next Generation Sequencing (NGS) testing can identify targetable genetic alterations, and has become an essential component of cancer care. Prolonged turnaround time (TAT) for tissue NGS testing (time from ordering until report arrival) can be a barrier for optimal treatment. Reducing the TAT can significantly aid the decision making process and improve the time to treatment initiation. As a quality improvement project, we explored a strategy to improve the TAT at our safety net hospital. Methods: We studied the TAT of commercial NGS test (FoundationOne CDx) ordered between January and December of 2022 by the oncology service at Bellevue Hospital Center (BHC), an affiliated public hospital. Tissue NGS testing was ordered for solid tumors regardless of tumor origin at the discretion of the provider. During the first half of the year, specimens prepared at BHC lab were manually dropped off at a central location for shipping to Foundation Medicine lab (testing site). The workflow was changed on 7/7/2022 to introduce a courier system to pick up the specimens from pathology lab at BHC. We analyzed the data reported by Foundation Medicine. Results: As of 2/1/2023, all specimens ordered between 1/1/2022 to 12/29/2022 were reported. There were 123 reports, of which 122 were used for the analysis. 1 report was excluded due to mislabelling of the specimen at the testing site that led to significant delay in TAT. The mean TAT of all reported specimens for the year 2022 was 23 days. Introduction of the courier reduced the mean TAT from 25 days to 20 days. After introducing the courier system, the mean time for specimen analysis was similar, but the mean time from specimen ordering to arrival at the testing site was reduced from 14 days to 10 days. Conclusions: The TAT for tissue NGS testing in underserved cancer patients at a safety net hospital can be successfully improved by implementing a courier system. The improvement was achieved regardless of tumor origin or patients’ financial ability to pay and can aid decision making for treatment initiation. This strategy can be adopted by community hospitals without access to on-site NGS testing. We are exploring additional interventions to optimize the workflow to further reduce the TAT. [Table: see text]

The implementation of a diverse clinical trial program at a rural community cancer center over a 10-year period.

e18559 Background: The state of Mississippi ranks last or close to last in the majority of health outcomes including cancer outcomes due to health disparities. Health disparities are worse for patients who face obstacles due to race, socio-economic status, or geographical exclusion. Improving minority enrollment in clinical trials allows for improved cancer care. We report the implementation of an inclusive and diverse clinical trial program at a community cancer center in rural Mississippi over a 10-year period of time. Forrest General Cancer Center cares for patients in a 19 county area with a diverse patient population consisting of 72.8% White, 25.8% Black, and 1.4% Other. Methods: We reviewed the patient database of our clinical trial program over a 10-year period of time from January 1, 2013, through December 31, 2022. All patients enrolled on clinical trials during this period were included in our analysis. We collected data regarding types of clinical trials, racial demographics, patient age, and primary sight of malignancy. A two proportion Z-test was utilized to analyze any differences between race in the study group and cancer population as a whole. A 2 tailed P value was then calculated. Results: During the 10-year period, 242 patients successfully enrolled onto clinical trials. 227 (93.80%) patients were enrolled on interventional treatment studies, and 15 (6.19%) patients were enrolled on registration studies. 233 (96.28%) patients were enrolled on pharmaceutical sponsored trials as opposed to 9 (3.72%) patients enrolled on cooperative group trials. 180 (74.4%) patients were White, 62 (25.6%) patients were identified as minority patients including 60 (24.8%) Black patients, and 2 (0.8%) Asian patients. There was no statistical difference between the racial diversity of study patients compared to baseline patient demographics (P = 0.608318). The average age of enrollment was 61.85 years (range 24-89). The average age for White patients was 63.06 years, and the average age for Black patients was 57.77 years. The most common malignancies included Breast 91 (37.6%), Lung 63 (26.0%), Hematolymphoid 41 (16.9%), Colon 16 (6.6%), Head and Neck 10 (4.1%), and Melanoma 5 (2.1%). Conclusions: We have demonstrated the successful implementation of an Oncology clinical trial program in rural Mississippi leading to advanced patient care with increased availability to high-quality clinical trials. During this time period, we have successfully enrolled a diverse population accurately representing our patients under our care. We feel that our model can successfully be implemented at cancer centers caring for underserved and minority patients throughout the United States leading to improved patient care in these populations. We have shown that it is possible to have a successful clinical trial program with primarily industry sponsored clinical trials in a community cancer center.

Clinical outcomes of patients with newly diagnosed acute myeloid leukemia receiving treatment in a safety-net hospital system.

7035 Background: Standard induction therapy for acute myeloid leukemia (AML) results in 5-year overall survival (OS) rates of 30-45%. Further risk stratification and novel therapeutic options have improved 5-year OS rates to 75-80% in certain subgroups. Despite these advances, many patients lack proper socioeconomic resources, leading to limited access to new therapies, clinical trials, and hematopoietic cell transplantation (HCT), which remains the only curative therapy for AML. Many of these patients receive care within publicly funded ‘safety net’ institutions designed to serve all individuals, regardless of ability to pay, insurance status, or immigration status. Outcomes of historically underserved patients with AML treated within a large county hospital system remain unknown. Methods: This retrospective cohort analysis evaluated clinical outcomes of patients aged ≥18 years with newly diagnosed AML (ND-AML) treated in the Harris County Health System, Houston, TX, from January 2005 to July 2019. Results: One hundred ninety-two patients with ND-AML were included (de novo AML [n = 140], secondary AML [n = 45], and therapy-related AML [n = 7]). Median age was 52 years (range: 19-89) at diagnosis, and 102 (52.3%) were male. Most patients identified as Hispanic (n = 110). Sixty-two percent of patients were uninsured or indigent, receiving care under the county’s financial assistance programs. The majority had adverse risk 84 (44.3%), while 78 were intermediate (40.6%), and 26 favorable risk (13.5%). Standard induction therapy was administered in 172 (88%) of patients. Cytarabine and Anthracycline (7+3) based regimens were most commonly used (72.8%). 192 patients were evaluable for response. 139 (72.4%) achieved composite complete remission (CR/CRi) and 41 (21.4%) had primary refractory disease. For those who achieved CR after first or second induction, the median duration of CR was 16.7 months. with 25 (13.7%) ultimately receiving HCT. Common reasons for not receiving HCT included lack of insurance (19.2%) or non-citizen status (23.1%) (n = 77). 82 of the 139 patients who achieved CR relapsed. After a median follow up of 16 months, five-year OS rates were 20.7% from our entire cohort of patients. Conclusions: In this cohort of patients treated in a safety-net system, median OS and EFS were lower than reported outcomes in the past decade. In addition, only 13.7% received HCT, one of the only curative treatments for AML. Some barriers included inability to refer to transplant center due to lack of financial support or immigration status. Addressing socioeconomic and institutional barriers to AML therapy and HCT will advance care for disadvantaged patients, thereby improving outcomes of patients AML.

Application of natural language processing to assess the performance status documentation quality metric in patients with non–small-cell lung cancer.

e13582 Background: Performance status (PS) assessment is important for cancer prognosis and treatment decision-making. PS documentation for stage IV non-small cell lung cancer (NSCLC) is a quality measure for ASCO Quality Oncology Practice Initiative (QOPI) Certification Program. We used natural language processing (NLP) to extract PS from unstructured electronic medical records (EMR) clinical notes in patients with metastatic NSCLC to verify PS documentation compliance and disparities in PS documentation. Methods: We identified oncology progress notes of stage IV NSCLC patients from Harris Health System, which is an integrated comprehensive healthcare system for underserved patients in Houston, TX. All progress notes within 60 days of first oncology clinic visit date were filtered based on visit type and author type. A rule-based regular expression NLP algorithm was used to extract the PS, namely Eastern Standard Cooperative Group (ECOG) or Karnofsky Performance Status (KPS). A total of 400 oncology notes were randomly selected from 400 distinct patients and annotated by two clinician reviewers, and discrepancies were resolved by a senior clinician judge. The annotated notes were split randomly into 200 training notes and 200 validation notes, and the performance of the NLP algorithm was evaluated. We used logistic regression to analyze potential disparity in PS documentation. Results: We identified 15,239 progress notes from 757 stage IV NSCLC patients. The NLP algorithm accuracy validated on the test data was 99% with corresponding positive predictive values (PPVs) for ECOG PS 0, 1, 2, 3, 4 of 100%, 100%, 95%, 94%, 94%, respectively. The NLP algorithm identified 85% of patients with appropriate PS documentation (sensitivity 100%, specificity 91%, positive predictive value 98%, negative predictive value 100%). Logistic regression analyses showed no significant differences in association between PS documentation and race/ethnicity, language, or sex. Conclusions: We developed an internally validated NLP algorithm to extract PS from EMR clinical notes. Despite potential language barriers, the QOPI metric of >75% PS documentation was met in our hospital system without significant disparities in clinical care. [Table: see text]

Diagnostic yield of pediatric and prenatal exome sequencing in a diverse population

The diagnostic yield of exome sequencing (ES) has primarily been evaluated in individuals of European ancestry, with less focus on underrepresented minority (URM) and underserved (US) patients. We evaluated the diagnostic yield of ES in a cohort of predominantly US and URM pediatric and prenatal patients suspected to have a genetic disorder. Eligible pediatric patients had multiple congenital anomalies and/or neurocognitive disabilities and prenatal patients had one or more structural anomalies, disorders of fetal growth, or fetal effusions. URM and US patients were prioritized for enrollment and underwent ES at a single academic center. We identified definitive positive or probable positive results in 201/845 (23.8%) patients, with a significantly higher diagnostic rate in pediatric (26.7%) compared to prenatal patients (19.0%) (P = 0.01). For both pediatric and prenatal patients, the diagnostic yield and frequency of inconclusive findings did not differ significantly between URM and non-URM patients or between patients with US status and those without US status. Our results demonstrate a similar diagnostic yield of ES between prenatal and pediatric URM/US patients and non-URM/US patients for positive and inconclusive results. These data support the use of ES to identify clinically relevant variants in patients from diverse populations.

Large-Scale, Primary Care-Based Hepatitis C Treatment in an Urban, Medically Underserved Patient Population.

Hepatitis C virus (HCV) infection is a critical public health concern in the United States. HCV is highly curable, but access to care is limited for many patients. Primary care models can expand access to HCV care. The Grady Liver Clinic (GLC) is a primary care-based HCV clinic founded in 2002. During 20 years, using a multidisciplinary team, the GLC expanded its operations in response to advances in HCV screening and treatment. We describe the clinic model, patient population, and treatment outcomes of the clinic from 2015 through 2019. During this period, 2689 patients were seen in the GLC, and 77% (n = 2083) initiated treatment. Eighty-five percent (1779 of 2083) of patients who started treatment completed treatment and were tested for cure, and 1723 (83% of the total treated cohort, 97% of those tested for cure) were cured. Building on a successful primary care-based treatment model, the GLC dynamically responded to the changes in HCV screening and treatment guidelines, continually increasing access to HCV care. The GLC serves as a model of primary care-based HCV care that aims to achieve HCV microelimination in a safety-net health system. Our findings support the notion that for the United States to achieve elimination of HCV by 2030, generalists can and should provide HCV care, particularly in medically underserved patient populations.

Benchmarking halcyon ring delivery system for hypofractionated breast radiotherapy: Validation and clinical implementation of the fast-forward trial.

The aim of this study was to demonstrate the feasibility and efficacy of an iterative CBCT-guided breast radiotherapy with Fast-Forward trial of 26Gy in five fractions on a Halcyon Linac. This study quantifies Halcyon plan quality, treatment delivery accuracy and efficacy by comparison with those of clinical TrueBeam plans. Ten accelerated partial breast irradiation (APBI) patients (four right, six left) who underwent Fast-Forward trial at our institute on TrueBeam (6MV beam) were re-planned on Halcyon (6MV-FFF). Three site-specific partial coplanar VMAT arcs and an Acuros-based dose engine were used. For benchmarking, PTV coverage, organs-at-risk (OAR) doses, beam-on time, and quality assurance (QA) results were compared for both plans. The average PTV was 806 cc. Compared to TrueBeam plans, Halcyon provided highly conformal and homogeneous plans with similar mean PTVD95 (25.72 vs. 25.73Gy), both global maximum hotspot<110% (p=0.954) and similar mean GTV dose (27.04 vs. 26.80Gy, p=0.093). Halcyon provided lower volume of ipsilateral lung receiving 8Gy (6.34%vs. 8.18%, p=0.021), similar heart V1.5Gy (16.75%vs. 16.92%, p=0.872), V7Gy (0%vs. 0%), mean heart dose (0.96 vs. 0.9Gy, p=0.228), lower maximum dose to contralateral breast (3.2 vs. 3.6Gy, p=0.174), and nipple (19.6 vs. 20.1Gy, p=0.363). Compared to TrueBeam, Halcyon plans provided similar patient-specific QA pass rates and independent in-house Monte Carlo second check results of 99.6%vs. 97.9% (3%/2mm gamma criteria) and 98.6%versus 99.2%, respectively, suggesting similar treatment delivery accuracy. Halcyon provided shorter beam-on time (1.49 vs. 1.68min, p=0.036). Compared to the SBRT-dedicated TrueBeam, Halcyon VMAT plans provided similar plan quality and treatment delivery accuracy, yet potentially faster treatment via one-step patient setup and verification with no patient collision issues. Rapid delivery of daily APBI on Fast-Forward trial on Halcyon with door-to-door patient time<10min, could reduce intrafraction motion errors, and improve patient comfort and compliance. We have started treating APBI on Halcyon. Clinical follow-up results are warranted. We recommend Halcyon users consider implementing the protocol to remote and underserved APBI patients in Halcyon-only clinics.

Consenting in the time of the COVID-19 pandemic

Consenting in the time of the COVID-19 pandemic