e18035 Background: The Oncology Care Model (OCM) captures patient costs in a 6-month episode triggered by administration of systemic therapy. Most breast cancer (BC) patients will receive systemic therapy, with variations depending on stage and hormonal status, which makes BC an ideal indication to study costs in the OCM. Practices earn performance based payments (PBP) if aggregate episodic expenditures are managed below set target prices. We investigated predictors for episodic expenditures exceeding target prices, thus reducing potential for PBP. Methods: We identified BC episodes with non-decedent beneficiaries attributable to our academic medical center from OCM Reconciliation Reports during 7/1/16-6/30/17. Cohorts were defined as episodes whose costs were above target (Cohort 1) and those below (Cohort 2). The Wilcoxon Rank-Sum test was used to compare actual and target episode expenditure between cohorts. Multivariable logistic regression models were used to assess association of maintaining costs below target due to various predictors. Results: A total of 369 episodes were included in the study, with 124 episodes in Cohort 1 and 245 in Cohort 2. Median actual and target episode expenditures were higher in Cohort 1 (actual: $23,466 vs. $2,691, p < 0.0001; target: $8,425 vs. $5,870, p < 0.0001). In multivariable logistic regression, episodes were more likely to be below target if novel therapies, Part B drugs, or inpatient admissions were not utilized, controlling for other predictors (see Table). Conclusions: Large disparities exist for both actual expenditures and target prices for BC episodes in the OCM. Novel therapies, Part B drugs, and inpatient admissions are negatively associated with maintaining episode expenditures below target. Risk-adjustments for these expenditures need to be overhauled in OCM to accurately capture costs associated with management of cancer patients, and provide practical target prices for institutions to continue delivery of value based care. [Table: see text]

109 Background: Oncology care, including drugs, represents a significant portion of US healthcare spending. Cost of Part B drugs has increased at a rate 5.7x that of overall Medicare spending (1997-2004). As a participant in the Oncology Care Model (OCM), we found drug costs represent a majority of our total costs. To reduce treatment (Tx) variability, our NCI-designated cancer center chose to implement pathways. Pathways are a clinical decision-support tool that use evidence-based care maps accounting for efficacy, toxicity and cost. At one institution, use of pathways contributed to $15k in savings for stage IV lung cancer Tx. We hypothesized pathway driven Tx standardization would favorably impact total chemotherapy (CTx) costs at the implementation site. Methods: In July 2018, we implemented pathways in Medical and Radiation Oncology for new starts or changes in Tx. Oncologists accessed the tool through our EMR, selected and placed orders for Tx. OCM quarterly data was used to compare 2 quarters immediately pre- and post-pathway implementation. The cancer-mix-adjusted Per-Member-Per-Month (PMPM) Allowed Amounts for CTx were compared between 3 groups; patients on-pathway, patients off-pathway and patients for which the pathways tool was not used (no utilization). PMPMs were evaluated pre- and post-implementation and an ANOVA test was used to evaluate significance of the difference between the two periods. Results: PMPM CTx costs decreased 4.6% between pre- and post-pathway implementation when oncologists followed pathways. By comparison, the off-pathway cohort and the no utilization groups had increases of 0.9% and 17.7% respectively. An evaluation of cost difference proved significant (p < .0001). Breast patients on-pathway had a cost decrease of 20%, compared to increases of 32% and 11% for off-pathway and no utilization groups, respectively. Conclusions: Pathway use reduced variation, a known contributor to healthcare costs, and therefore may be an effective cost control tool. Additional quarters of claims data is needed post-implementation to fully define the impact of pathways on total cost. [Table: see text]

Introduction: During the COVID-19 pandemic, telemedicine became an essential way to provide patient care. We assessed the feasibility, acceptability, and initial impact on health care disparities of using telemedicine in a large pediatric lipid clinic. Methods: Data were collected from a telemedicine (TM) cohort between 4/1/20-3/31/21 and compared to an in-person (IP) cohort between 1/1/19-3/13/20. Families participated in telemedicine visits through an online patient portal. Interpreter services were available. Family surveys were sent after each telemedicine visit; providers were surveyed monthly. We evaluated family and provider satisfaction, cancellation rates, travel-related time and costs, and visit duration (check-in to check-out). Data were analyzed using standard descriptive summaries and student’s t-test. Results: Five providers completed 784 lipid telemedicine visits, accounting for 79% of total visit volume between 4/1/20-3/31/21. Average age was 14 years (+/- 3.8 yrs) and 366 (46.7%) were male. Interpreters were used in 11 (1.4%) visits. Payor mix (Medicaid vs. commercial) was comparable between TM (19.9%) and IP cohorts (22.4%). Of 98 completed telemedicine family surveys, 18 (18.4%) noted technical difficulties; however, nearly all families were interested in future telemedicine visits. Sixteen monthly provider surveys were completed: 31% reported technical difficulties and rated telemedicine 9.1/10. The 48-hour cancellation rate decreased from 11% to 7% (IP and TM cohort, respectively, P < 0.018), with a decrease from 25% to 5% for Non-Hispanic Black patients ( P < .001). No show rates were consistent between cohorts at 2%. Total driving distance saved was 20,800 miles with an estimated travel-related savings of $11,440. Average visit time for IP cohort was 87.5 minutes compared to 39.5 minutes for the TM cohort. Conclusion: We found that using telemedicine in a pediatric lipid clinic is feasible, time efficient, and associated with travel-related time and cost savings. There was a significant decrease in 48-hour cancellation rates overall, most notably for Non-Hispanic Black patients. Future studies utilizing telemedicine in our clinic will evaluate the effect on long term cardiovascular risk and health care delivery.

6632 Background: The Oncology Care Model (OCM) is a 5-year demonstration project led by the Centers for Medicare and Medicaid Services (CMS) to create a framework for the future of oncology care in the United States. More than half way through the project, our large, urban NCI-designated cancer center chose to focus on and invest in resources and personnel for patient navigation and the development of clinical pathways. Although navigation has shown to reduce emergency department (ED) visits by as much as 6% per quarter compared to non-navigated patients, sustaining it is a challenge because it is a nonbillable service. Clinical pathways are a tool to reduce care variation by addressing drug expenditures, and represent an opportunity to reduce outpatient costs by as much as 35% when patients are treated on pathway.3 Many OCM practices made similar investments and all are facing the question: How will the infrastructure and efforts developed during OCM be sustainable after the demonstration project ends? Methods: An analysis of average ED cost and utilization as well as drug expenditures was conducted using OCM feedback data (Q1-Q8). Total utilization of ED visits and ED admits were used to determine a projected annualized cost which was compared to a budgeted navigation team. Similarly, projected annualized drug expenditures were compared to the annual cost of the pathways tool. Results: We found that ED visits and ED admits would need to be reduced by 11% to cover navigation costs. Similarly, a 0.7% reduction in total drug expenditures would cover the cost of clinical pathways. The OCM data represents a timeframe prior to implementation of these programs and an average increase of 1.6% per quarter for ED admits, a 0.6% decrease in ED visits and 2.7% increase in drug expenditures. This will serve as a baseline to measure progress towards our sustainability targets. Conclusions: Long term sustainability of the infrastructure developed during OCM to support cancer care transformation will be dependent on reducing high cost and highly utilized services. Aligning impact areas with resources/tools to ensure sustainability is an approach that can help define targets for OCM practices.

Objectives: Real-world melanoma clinical practice is changing rapidly with the introduction of new treatments. This analysis examined patient-reported treatment patterns for melanoma in UK real-world practice, including choice and duration of treatment, consultation frequency, and application of treatment guidelines, in close to real-time. Methods: Adults with any type or stage of melanoma were recruited to an observational study exploring the real-world impact of melanoma, conducted in collaboration with the patient advocacy group Melanoma UK. Data were collected via a bespoke 'bring your own device' app co-created with patients and Melanoma UK. Informed consent was obtained via the app, and ethical approval was obtained. Participants answered one-time questions relating to treatment patterns: current and previous treatments (drugs and surgery), where they lived/were treated, and frequency of consultations. Data were analyzed using R software; statistical analysis aimed to identify associations between treatment and other patient characteristics, including demographics and Eastern Cooperative Oncology Group (ECOG) performance status. Treatment patterns were compared with current guidelines. Results: Of the 219 (total registered: 396) patients who specified current disease stage, 45% were stage 3 or 4; 83% reported a latest ECOG performance status of unknown or 0. Current real-world treatment of melanoma in the UK was broadly consistent with published guidance; immunotherapies dominated treatment at stage 4 (ipilimumab 35% patients; nivolumab 23% patients). 178 patients provided their location, allowing regional differences in treatment patterns to be identified. Radiotherapy was more common in North-(16% of patients) and South-West (14% of patients) regions, but was not used to treat any patients in the East Midlands and South-Central regions. Centers treating the most patients included the Christie, Royal Marsden and Nottingham Hospitals NHS Trusts. Conclusions: Melanoma treatment in the UK real-world setting broadly follows published guidelines. Availability of close to real-time data offers the opportunity to quickly identify any inequalities, improving care.