The Centers for Medicare & Medicaid Services (CMS) announced its third annual evaluation report of the Oncology Care Model (OCM) in July 2020, covering performance periods 1-3 activities and results. In March 2019, this author and colleagues offered observations in this journal1 based on the previous second annual report on OCM, announced by CMS in December 2018. The current article builds upon those observations in light of data from the third annual report, focusing on potential opportunities for improvement in utilization and other performance metrics including end-of-life (EOL) care, depression and pain/oncology care integration, and application of continuous quality improvement (CQI) methods. OCM practices are continuing to evolve and their value transformation efforts take on a new sense of urgency in an era of pandemic care.
Legislation under the Medicare Access and CHIP Reauthorization Act of 2015 (MACRA) established a trailblazing approach to value-based reimbursement with the Quality Payment Program (QPP). Under the QPP, two pathways to payment were established: the Merit-based Incentive Payment System (MIPS) and the Advanced Alternative Payment Models (Advanced APMs).2 The two-sided OCM is an example of an Advanced APM, and the 5-year model (2016-2021) was designed to promote high-quality and high-value cancer care through a value-based reimbursement strategy incentivizing the transformation of cancer care delivery.3,4 Due to COVID-19, the model has been extended for a sixth year, through June 2022.5
In July 2020, CMS released the “Evaluation of the Oncology Care Model: Performance Periods 1-3” report (OCM Report) summarizing results for those periods and describing the continuing efforts and results of OCM participants to transform their practices to deliver better value and population health for cancer patients.6 This latest OCM Report addresses performance through the third performance period (including episodes beginning July 1, 2016 and January 1, 2018, all of which have ended by June 30, 2018). In many instances, data from OCM participant practices were compared with data collected from a group of non-OCM practices that the evaluators identified as similar to the OCM practices before the model was initiated. The comparison offered an opportunity to identify the impact of the OCM over time.6
In this most recent report,6 the evaluators noted that there was no impact of OCM on Total Episode Payments (TEPs) (without Monthly Enhanced Oncology Service [MEOS] payments). However, OCM led to a relative TEP decrease (without MEOS) for higher-risk episodes, while seeing a relative TEP increase in lower-risk episodes (without MEOS). The evaluators commented on rising chemotherapy drug costs and noted that few practices reported specific efforts to reduce drug spending, other than adoption of treatment pathway programs, with 60% of OCM-surveyed oncologists reporting use of treatment pathways.6 The evaluators also noted that, during the first 2 years, OCM practices’ primary foci were on efforts to better manage the care of patients at higher risk of adverse events, aiming to minimize avoidable emergency department (ED) and hospital utilization.6
This article will therefore review updates to performance reported on metrics such as ED visits, hospital utilization, advance care planning, management of depression and pain, and application of CQI methods; it also suggests specific paths forward, expanding upon current accomplishments. These factors may be of interest to both OCM and non-OCM oncology practices seeking to transform their practices to offer better value for patients and payers. Non-OCM practices may be participating in other value-based reimbursement models, such as the MIPS arm of the QPP, in which their Medicare reimbursement is impacted by performance in four areas, including quality and cost.7 They may also be considering participation in the still-evolving Oncology Care First model delayed by CMS in part due to COVID-19,8 as CMS gives further consideration to evaluation results and stakeholder comments.
Ed and Hospital Utilization: Disappointing Results
The OCM assesses practices’ performance on an array of metrics, including ED and hospital utilization. Prevention of avoidable ED visits and hospitalizations have been two of the primary emphases of the OCM. Although the second annual OCM Report noted a greater decline in all hospital utilization measures among the OCM practices, compared with the non-OCM comparison group, during the early OCM phase,9 these declines do not appear to have been sustained. During performance periods 1-3, the report noted “no impact of OCM on ED visits or hospitalizations at acute care hospitals, or on ED visits and hospitalizations due to chemotherapy toxicity.”6 Although OCM participants reported new or enhanced care processes relative to the availability of same-day appointments, and evaluators reported OCM practices are expanding urgent care access,6 relatively few reported the availability of evening or weekend appointments for patients with urgent needs.
For example, a survey of oncologists (N=399), conducted in August 2018-October 2018,10 found that only 16.5% of larger practices, with theoretically greater capacities to offer such services, offered access to evening/weekend appointments for patients with urgent needs; another 12.5% reported offering this level of access as a new or enhanced offering after OCM began. Access to evening/weekend appointments was even lower for the smaller oncology practices. These data suggest that there has been, at best, minimal progress in offering weekend/evening access for OCM patients with urgent needs, since OCM practice 2017 responses11 (N=183) found that only slightly more than one-third of practices reported offering extended evening or weekend hours for patients.
Although data sources differ (ie, the current report focused on oncologists surveyed in 2018; the earlier report focused on OCM Practice Transformation Plans in 2017), both reports suggest that access to weekend/evening appointments for oncology patients with urgent needs were not yet prioritized by most OCM practices during those performance periods. Thus, this gap in urgent care availability may be a key factor in the failure to reduce ED and acute care hospitalizations and is now of even greater concern due to COVID-19 and associated resource scarcity as well as infection vulnerability.
The Path Forward
Of note, oncology practices instituting urgent care services have reported a decline in avoidable ED visits and hospitalizations. For example, Yale’s Smilow Cancer Hospital and two nearby integrated community practices reported a decrease in the ED visit rate after the opening of an Oncology Urgent Care Center, concluding that the model may serve as a more efficient way for cancer centers to manage urgent oncology patient care.12 In another example, as a quality improvement intervention to reduce oncology patient acute care and cost, Baylor University Medical Center (BUMC) developed an Oncology Evaluation and Treatment Center providing after-hours acute care for oncology patients receiving outpatient care at BUMC. The authors reported that the intervention was successful, yielding decreases in oncology-related unplanned hospitalizations and related costs.13 Further observations on the topic of initiatives to mitigate avoidable ED visits and hospitalizations will be discussed during this article’s section on application of CQI methods.
Care at End of Life: Mixed Results
Performance on OCM-3, measuring the proportion of patients who died who were admitted to hospice for 3 days or more,14 does not appear to have significantly improved for years 2014 to 2018, based on data provided in the most recent evaluation report. Differences between the OCM and comparison group were not statistically significant, and the OCM group baseline mean was 58.4%, increasing to only 59.7% over time.6 In addition, OCM Patient and Caregiver Surveys for the period April 2016 to June 2018 found no meaningful changes over time in the Shared Decision Making Composite Quality Measure.10
However, evaluators noted that OCM practices appear to be encouraging increased advance care planning and enhancing access to palliative care.6 Further, the OCM was found to lead to statistically fewer inpatient admissions at EOL, with an intervention mean of 52.4% compared with the comparison mean of 53.5% (P<.05).6 The evaluators also found that TEP (without MEOS) and Part A payments decreased for dying OCM patients’ last episodes, relative to comparisons (P<.05).6
According to evaluators, 82% of surveyed caregivers of deceased patients indicated that a care team member had discussed hospice with the patient, and this did not change significantly over time. Family members indicated that their deceased family member died in their preferred setting most of the time (75%) with no consistent or significant change over time.6 These data suggest that use of advance care planning, palliative care, and discussions regarding hospice have resulted in less hospitalization at EOL, without an appreciable decline in meeting the wishes of dying patients or their families.
The Path Forward
Qualitative information gleaned from field conversations with clinicians suggested that sometimes patients continue with chemotherapy due to reluctance to “let their oncologist down.”6 Among surveyed oncologists, about one-third (32%) used standards or guidelines to trigger hospice discussions and felt they improved the quality of care for their dying patients.6 Clinicians wishing to further expand their efforts to facilitate dying patients’ care at EOL and expand hospice use may wish to consider use of standards or guidelines to trigger hospice discussions.
Shared decision-making training and tools could also advance earlier hospice discussions. A cluster randomized clinical trial15 in outpatient oncology conducted at the Dana-Farber Cancer Institute evaluated a communication quality improvement intervention and found it resulted in more, earlier, better, and more electronic medical record-accessible serious illness conversations about patients’ values, goals, and informed preferences (consistent with shared decision-making) through the use of clinical tools, clinician training, and systemic changes.
Management of Depression and Pain: Mixed Results
As the authors noted in their 2019 article,1 depression contributes to higher rates of suicide among patients with cancer, and pain is a common cause of ED visits and hospitalizations among oncology patients.
Although a vast majority of oncologists surveyed (OCM Clinician Survey, August-October, 2018, N=399 oncologists) report now screening for depression (95%) and for psychosocial distress (85%),6 and evaluators observed that OCM practices are enhancing access to palliative care,6 it is unclear if patients are experiencing benefits from these interventions. In a patient-reported survey about symptom management, including management of pain and emotional problems,10 a slight but statistically significant (P<.05) decline in symptom management was observed from baseline through the third performance period. More specifically, a statistically significant decline (74.8% at 2016 baseline vs 72.8% for January 2018-June 2018; P<.05) was noted on the OCM Patient and Caregiver Survey statement “received help with pain,” and less than half of patients at the 2016 baseline reported receiving help with emotional problems (44.3%) with no statistically significant improvement over time, culminating with only 44.9% receiving help with emotional problems in 2018 (January 2018-June 2018). The report noted that clinical experts did not view the slight decrease in symptom management as clinically meaningful, and the evaluators plan additional investigation, comparing changes over time among both OCM and comparison groups on an array of patient/caregiver experience measures, including symptom management.6
Notwithstanding the comments offered by the clinical experts, it should be noted that over a 2-year period, approximately one-quarter of OCM respondents reported not receiving help with pain, and less than half reported receiving help with emotional problems.10
The Path Forward
Additional investigation is required to better understand why a significant number of OCM patients do not report receiving help with pain and emotional problems, despite clinician reported initiatives to screen for depression and psychosocial distress and offer enhanced access to palliative care. Integration of behavioral health care with care for complex conditions such as cancer has been a long-standing challenge, and Milliman’s Stephen Melek and colleagues have reported that integration of behavioral health with physical health care can yield 5% to 10% reductions in total health care costs over 2 to 4 years for patients receiving collaborative care.16
It is possible that, while screening and referrals for management of pain and depression may be occurring, there is a lack of integration and coordination of care for pain and depression. This may result in suboptimal outcomes, including patient perceptions of not being helped and avoidable ED and hospital utilization. Of additional and considerable concern is that behavioral health conditions may be further exacerbated by the COVID-19 pandemic. Identification of and outreach to oncology patients with comorbid depression and/or pain could potentially yield greater opportunities to provide support, connect patients with timely care, and enhance patients’ perceptions that they are receiving help. Surveyed OCM oncologists (40%) indicated that routine proactive outreach calls are made to some or all high-risk OCM patients,6 and practices should consider the potential benefits of targeted outreach to patients experiencing pain or depression. Finally, behavioral health clinicians with expertise treating oncology patients experiencing conditions such as depression and pain should be integrated within practices, through co-location, telemental health services, and/or formalized agreements, including robust protocols for routine, urgent, and emergent referrals and appointments, with bi-directional clinical data exchange.1
Application of Continuous Quality Improvement Methods: Mixed Results
For oncology practices to be successful in a value-based ecosystem, it is imperative that they commit resources to the practice of CQI, with consistent application of a framework such as the Model for Improvement or the Framework for Cancer Quality Improvement.1 Despite the availability of quality improvement models, analyses of the 2017 Practice Transformation Plans (PTPs) in the OCM Report found that only 62.3% of the OCM practices (N=183) indicated they were employing a formal model of quality improvement.1 Data from the recent report did not indicate the percentage of practices using a formal quality improvement model, but the evaluators indicated that practices are using CMS’ Feedback Reports, claims data, and EMR data to identify opportunities for improvement, and the majority of practices are using dashboards to track performance.6
However, when reviewing 2018 survey data on the CQI performance metrics shared by OCM practices with oncologists, approximately two-thirds (67%) of the OCM oncologists indicated they routinely receive performance metrics, and slightly more than one-third (37%) reported receiving data on patient ED visits and hospitalization. Only about half receive survey data about patient experiences with cancer care. Yet, most surveyed oncologists indicated that comparative performance data was important to them and wanted more information about their performance.6
The Path Forward
With approximately 1 of 3 OCM oncologists not routinely receiving performance data, 2 of 3 not receiving data on ED and hospital utilization, and half not receiving patient experience data, there are opportunities to close gaps in physician feedback, better address oncologists’ interests in having more information about their performance, and more effectively address improvement opportunities. Further, routine receipt of performance data on metrics of interest to multiple stakeholders, such as ED and hospital use, could elucidate team members’ identification of opportunities for addressing performance on these metrics. Regular receipt of patient experience data could help teams better understand, for example, why many patients perceive they are not receiving help for pain and emotional problems, despite OCM practices’ increased efforts.
Conclusion
CMS has posted its evaluation of OCM performance through the third of what are now projected to be 11 performance periods. Value transformation will continue, and it requires practices to accelerate data review and practice modifications, especially in response to the rapidly changing COVID-19 environment. The COVID-19 pandemic has generated delays in care, potentially impeding access throughout the care continuum, spanning diagnosis, clinical trials, treatment, survivorship planning, and EOL care. Now more than ever, novel solutions and process redesign are needed to bridge gaps in care and generate better clinical and financial outcomes.
Oncology practices continue to evolve, now within an era of pandemic care. Opportunities exist for stakeholders collaborating with OCM practices (eg, health systems and plans, specialty pharmacies, pharmacy benefit managers, and pharmaceutical manufacturers) to support OCM practices’ quality improvement initiatives with training, tools, analytics, and consultation.
References
1. Hennessey M, Bolland L, Blandford L. Value transformation: observations on the second annual evaluation report of the Oncology Care Model. J Clin Pathways. 2019;5(2):31-34. doi:10.25270/jcp.2019.03.00067
2. Centers for Medicare & Medicaid Services. Department of Health & Human Resources. Quality payment program. Updated March 23, 2020. Accessed August 26, 2020. https://www.cms.gov/Medicare/Quality-Payment-Program/Quality-Payment-Program
3. Centers for Medicare & Medicaid Services. Department of Health & Human Resources. Quality payment program: APMs Overview. Accessed August 26, 2020. https://qpp.cms.gov/apms/overview
4. Centers for Medicare & Medicaid Services. Department of Health & Human Resources. Quality payment program: advanced alternative payment models. Accessed August 26, 2020. https://qpp.cms.gov/apms/advanced-apms
5. Centers for Medicare & Medicaid Services. Department of Health & Human Resources. CMS Innovation Center Models COVID-19 Related Adjustments. June 3, 2020. Accessed August 26, 2020. https://www.cms.gov/files/document/covid-innovation-model-flexibilities.pdf
6. Centers for Medicare & Medicaid Services. Evaluation of the oncology care model: performance periods 1-3. May 2020. Accessed August 26, 2020. https://innovation.cms.gov/data-and-reports/2020/ocm-evaluation-annual-report-2
7. Centers for Medicare & Medicaid Services. Department of Health & Human Resources. Quality payment program: MIPs overview. Accessed August 26, 2020. https://qpp.cms.gov/mips/overview
8. LegisLink. Value-based reimbursement: alternative payment models, Oncology Care Model/Oncology Care First Model. Accessed August 26, 2020. https://legislink.com/value-based-reimbursement/
9. Centers for Medicare & Medicaid Services. Evaluation of the Oncology Care Model: Performance Period One. December 2018. Accessed August 26, 2020. https://innovation.cms.gov/files/reports/ocm-secondannualeval-pp1.pdf
10. Centers for Medicare & Medicaid Services. Evaluation of the oncology care model: performance periods 1-3-appendices. July 2020. Accessed August 26, 2020. https://innovation.cms.gov/data-and-reports/2020/ocm-evaluation-annual-report-2-appendices
11. Centers for Medicare & Medicaid Services. Evaluation of the oncology care model: performance period one—appendix. December 2018. Accessed August 26, 2020. https://innovation.cms.gov/Files/reports/ocm-secondannualeval-pp1-appendix.pdf
12. Sedghi T, Canavan M, Gross CP, et al. Impact of an oncology urgent care clinic on emergency department rates. J Clin Oncol. 2019;37(suppl 15):6615-6615. doi:10.1200/JCO.2019.37.15_suppl.6615
13. Coyle YM, Ogola GO, MacLachlan CR, Hinshelwood MM, Fleming NS. Acute care model that reduces oncology-related unplanned hospitalizations to promote quality of care and reduce cost. J Cancer Policy. 2019;21. doi:10.1016/j.jcpo.2019.100193
14. Centers for Medicare & Medicaid Services Innovation Center. Oncology Care Model Overview. February 2020. Accessed August 26, 2020. https://innovation.cms.gov/files/slides/ocm-overview-slides.pdf
15. Paladino J, Bernacki R, Neville BA. Evaluating an intervention to improve communications between oncology clinicians and patients with life-limiting cancer. JAMA Oncol. 2019;5(6):801-809. doi:10.1001/jamaoncol.2019.0292
16. Melek SP, Norris DT, Paulus J, Matthews K, Weaver A, Davenport S. Potential economic impact of integrated medical-behavioral healthcare. Milliman; January 2018. Accessed August 26, 2020. https://milliman-cdn.azureedge.net/-/media/milliman/importedfiles/uploadedfiles/insight/2018/potential-economic-impact-integrated-healthcare.ashx
