Currently Viewing:
Evidence-Based Oncology January/February

Lessons From an Oncology Medical Home Collaborative

John Fox, MD, MHA
While many pathways programs implemented by third parties are based on state or regional consensus, they remain unnecessarily broad and lack input from individual oncologists and practices. In the OMH program, the plan and the practices mutually agreed to resolve both of these issues by having oncology practices establish single “preferred regimens” for at least 4 highvolume conditions in the adjuvant and first- and second-line metastatic settings, where appropriate. The regimens must be consistent with the most recent version of the NCCN guidelines (

Incorporation of performance status, such as the ECOG performance status or Karnofsky score, serves as a critical data point for decision making. It is relevant both for selecting the therapeutic options appropriate for an individual patient and for assessing whether or not chemotherapy should be given on a given day. In fact, the first ASCO Choosing Wisely quality improvement measure proposes avoiding unnecessary anticancer therapy, including chemotherapy, in patients with advanced solid-tumor cancers and a low performance status (3 or 4), and instead focusing on palliative and supportive care or clinical trials (

Early results demonstrate that practices can readily meet these requirements. Further, their selections of preferred regimens are very similar (Table 1), which is remarkable because the practices developed these regimens independent of each other and without external input.

At present, there is no automated means of collecting tumor type, stage, biomarker status, or performance status. All results are based on self report. Most practices have agreed to use an independent third party to collect and report results. The results from the first 2 months of the program are shown in Table 2. Overall, 97% of adjuvant, first, or second-line therapies were consistent with the NCCN guidelines, while 86% of therapies were consistent with the practices’ self-selected preferred regimens.

Care Management

Proactive patient engagement and early intervention is critical to reducing potentially avoidable care and complications. Consultants in Medical Oncology and Hematology, PC, has demonstrated that effective triage and symptom management protocols can reduce ED visits per chemotherapy patient year by 65% and hospitalizations per chemotherapy patient year by 43%.6,7 Furthermore, the vast majority of clinical calls (approximately 75%) can be managed at home when patients call early in their symptom course.

The Oncology Medical Home care management program is intended to improve patient care and outcomes by having a consistent expectation of patients and practices alike. For patients, the plan and the practices mutually agreed that setting patient expectations to call the practices first was critical to reducing side effects, ED visits, and hospital admissions and to improving time on therapy. Likewise, we agreed that proactive care management required patient education protocols related to expected medication side effects, and phone triage protocols to ensure consistency in response by the clinician. At minimum, specific target areas include chemotherapyinduced nausea and vomiting, dehydration, constipation & diarrhea, fever, febrile neutropenia, depression, and fatigue.

Practices track their triage line call disposition with, at a minimum, the following categories: manage symptoms at home, office visit today, office visit tomorrow, referral to primary care or specialty care physician, referral to the ED, or direct admission to the hospital. Re-engineering practices to meet these requirements has taken more time and effort than anticipated. Many practices have adopted software applications such as Navigating Cancer for their patient education portal. Most practices are adapting the Oncology Nursing Society’s PEP program for incorporation in their EMRs.


Both the practice and the health plan are committed to promoting evidencebased initiatives supporting the goals of improved patient experience, improved health outcomes, and reduced per capita costs. Peer-reviewed evidence shows that patients who participate in an ACP program have less fear and anxiety, have a better ability to influence and direct their medical care, believe that their physicians have a better understanding of their wishes, and have a greater understanding and comfort level with their illness than they had before ACP. Moreover, for patients who participate in ACP discussions there are fewer associated resuscitations, fewer intubations, fewer ICU days, better quality of life, earlier referrals to hospice, and less depression.8 Among caregivers, there is a better quality of life, less depression, and less bereavement when end-of-life issues are discussed.9 Patients with advanced cancer who reported having EOL conversations with physicians had significantly lower healthcare costs in their final week of life. Higher costs were associated with worse quality of death.10

Both ASCO and the NCCN Palliative Care Guideline (V2.2012) integrate ACP. For example, the latter states that providers should “elicit personal values and preferences for end of life care and congruence with values and preferences of family and health care team… (and) encourage the patient to discuss goals and preferences with family, provide advanced directives, and designate healthcare proxy.” Health insurers are required to have shared decisionmaking programs in place under Section 936 of the Patient Protection and Affordable Care Act (PPACA). Shared decision making requires collaborative processes between clinicians and patients, caregivers, or authorized representatives. ACP is, by definition, a shared decision-making process that provides patients and families with information about trade-offs among treatment options and facilitates the incorporation of patient preferences and values into the medical care plan.

By mutual agreement between the physician practices and the plan, the former must establish an ACP program for patients receiving chemotherapy. At a minimum this must include a facilitated discussion with patients receiving treatment for metastatic disease within 60 days of initiation of chemotherapy and prior to additional treatment for disease progression. The facilitated discussion is typically with a trained staff member—not necessarily with the physician—and the patient and family members. When ACP documents are completed, the program requires insertion into the chart and filing with the patient’s preferred hospital.

The optimal timing for these discussions is not known. The timing requirement for this program is arbitrarily arrived at based on the assumptions that 1) the majority of people would expect at least 1 trial of therapy, 2) introduction at the initial visits would create confusion, and 3) most patients would not progress within the first 60 days. However, it is known that goals and preferences for care change with cancer patients, thus the need for followup discussions each time the patient’s disease progresses.

Each practice is allowed to design or select a program most suitable to the practice. One practice has chosen to implement the Respecting Choices program from Gunderson Lutheran Foundation ( while other practices have adopted the Five Wishes program (www.agingwithdignity. org). In either case, practices have had to design a replicable process for ensuring that patients can have a facilitated discussion at the relevant time. All practices have chosen to have a scripted introduction, indicating that these discussions occur with all patients in the practice who have incurable disease. Other logistical issues that need to be resolved are who will introduce the topic, how the appointment is to be scheduled, where consultations will occur (home or office), and how to bill for the services.

Most experience to date has been with the initial discussion. In fewer than 10% of cases do patients decline the discussion outright, while approximately 25% of patients indicate that they’ve already had a discussion or have an advance directive. In these circumstances, strategies need to be developed for engaging patients and their families in a review of their goals and preference for care to assist the practice in knowing and respecting their choices. For patients and families who have gone through the process, over 90% have welcomed the discussions and found them helpful in sharing beliefs and preferences. In fact, in one practice, the availability of a trained social worker and psychologist has led to more referrals of existing patients by oncology infusion nurses than of new patients.

Many challenges have surfaced. Typically these conversations take 1 to 2 hours, yet reimbursement is not available from most insurers. Priority Health has begun to reimburse any licensed healthcare provider for discussion facilitation using S0257, counseling and discussion regarding advance directives or end of life care planning and decisions, with patient and/or surrogate. Documentation for reimbursement minimally includes 1) the person designated to make decisions for the patient if the patient cannot speak for him- or herself, 2) the types of medical care preferred, and 3) the comfort level that is preferred. This code can also be used for POLST documentation.

At present, coverage is only for fully insured and self-insured patients, although the plan is actively exploring how to cover these services for Medicare and Medicaid recipients. Although the ACA only requires that the plan cover these services as preventive services when recommended by the US Preventive Services Task Force, they are covered by the plan at no cost to the patient based on national recommendations from the American College of Physicians and the American Medical Association, independent of age or health status.

Oncology Medical Home Performance Metrics

The third key element in the OMH initiative centers is codevelopment of performance metrics in 4 different domains. A fifth domain on survivorship will be added later. The initial metrics and the party responsible for the measurement are outlined in Table 3.

Of the 4 domains, the most challenging has been the ACP domain, primarily because getting accurate date-of-death data is a challenge. Measures of admission and chemotherapy during the last weeks of life as well as hospice in the last 3 days of life are all dependent on accurate date-of-death data. While practices could secure this information, the preferred alternative is to match the health plan cancer registry to the State of Michigan Vital Records death index. A probabilistic linkage to the national death index is also a viable albeit more expensive option.

Another significant measurement challenge has been tracking phone triage disposition to assess the impact of enhanced patient education and nursing triage protocols. Capturing triage disposition in most EMRs is difficult, requiring either inefficient manual tallies, electronic tools independent of the EMR, or expensive revisions to EMR software.

Implementation Challenges

Copyright AJMC 2006-2019 Clinical Care Targeted Communications Group, LLC. All Rights Reserved.
Welcome the the new and improved, the premier managed market network. Tell us about yourself so that we can serve you better.
Sign Up