“Centers of Excellence” (COEs) is not a new concept in healthcare. The underlying hypothesis is that providers who specialize in a particular procedure or service will produce superior, predictable outcomes. Payers have developed COE networks to manage cost and quality for complex medical conditions for more than 2 decades, steering volume to high-performing providers in exchange for discounted contractual rates. Under significant pressure to reduce the burden of cancer spending, payers are beginning to make bold network decisions, including narrowing networks, but they need precision tools to ensure that quality of care is uncompromised, and even improved, while reining in unsustainable cost trends.1-3
In 2008, the National Quality Forum published a complete review of the existing quality measure sets from leading cancer care organizations, including the National Comprehensive Cancer Network, National Cancer Institute, American College of Surgeons, Agency for Healthcare Research and Quality, Institute of Medicine, American Society of Clinical Oncology, and American Society of Hematology.4 The report also cited remaining gaps and limitations, including the availability of outcomes versus process metrics; the complications of collecting and managing data; and the types of cancer and the span of the cancer care delivery continuum.4
Much of the focus and resources to develop and execute measure sets have been limited to large facility-based settings and comprehensive academic programs. However, these cancer centers are not always the optimal choice for patients. Patient care is often more expensive at facility-based centers, which are best suited for rare cancers, complex cancers, and cancers that are nonresponsive to conventional treatment (approximately 20%-30% of cancer diagnoses).
However, most common cancers can be treated effectively in the community oncology setting, which can keep patients closer to home during often lengthy courses of treatment. Community oncology plays an important role in managing the ever-increasing cost of caring for and managing patients with cancer; however, standards of care, tools for quality management, and data for community oncology are lacking, leaving this space largely untouched by the COE movement.
In the healthcare reform environment moving from volume to value, payers seeking to evaluate the value of cancer care are quickly overwhelmed by the volume of existing standards and are uncertain how to make them actionable.
Cancer is not a single disease, and treatments are tolerated differently by different patients. The lack of meaningful comparisons on the relative cost of treatments between sites of service to the complexity, as well as the emotional sensitivity of the topic, make the evaluation of the “value” of cancer care a daunting task.
Oncology Resource Networks (ORN) has invested in market research and independent analysis to inform a simplified strategy that is designed to help payers evaluate their oncology network based on cost and quality of care for all sites of service. ORN is a unique and independent company positioned at the intersection of providers and payers, with a singular mission to promote value in cancer care delivery.
Given the level of variation that exists in the practice of oncology, payers and patients need standards to meaningfully and practically evaluate and compare providers in terms of cost and quality.
ORN conducted the Optum/Villages study in 2014 to analyze medical cost variations in a tricounty central Florida market using episodic treatment grouper software. The study was limited to outpatient chemotherapy and radiation therapy episodes, and factored in the total cost of treatment, which included chemotherapy, radiation therapy, inpatient and emergency department admissions, imaging, laboratory testing, physician services, and pharmacy. The total cost of care for the episode was attributed to the physician with the most visits for that member.
The top 3 oncology physician groups in the study represented nearly 60% of the total volume of services for the tricounty area. Cost comparisons across the provider groups showed the most costly group had 206% higher expenses than the least costly group. Drilling into specific physician groups, the highest-cost physician had 651% higher expenses than the lowest-cost physician within the same group.
Comparing physician groups within a specific episode (eg, breast cancer, active management, with surgery) demonstrated a 546% variation in cost. Although controlling for risk and excluding outliers provides a reasonable “apples-to-apples” comparison of cost within specific episodes, the study did not seek to understand the specific cause of the variation, but certainly suggests statistical and practical opportunities for efficiencies.
ORN conducted the Florida Health Care Plan study in 2014 that focused on east central Florida captured a typical site of service opportunity, where the cost of chemotherapy in the outpatient hospital setting was found to be triple the cost of chemotherapy in the physician office setting. Shifting care to the lowest-cost site of service represented a potential 20% savings opportunity on total chemotherapy spending for this payer.
Articles that cite variation in quality of cancer care services are abundant in the literature; however, these are retrospective studies that cannot be practically applied to real-time decisions regarding where to seek quality, affordable treatment.
This leaves primary care physicians and patients to choose an oncologist based on reputation or on a referral from a friend or a family member. This might have been appropriate in the past, when cancer treatments were less complex and less expensive. In today’s economy, where cancer spending in the United States is expected to reach $173 billion by 2020,5 a more data-driven approach to physician selection is required to ensure the best possible outcome, at an affordable price.
Industry thought leaders are talking about value-based care within and outside of the oncology arena, but there is no clear, consistent definition of what value means. ORN has established a definition of value as representing elements of cost and quality that span the entire continuum of cancer care. The methodology and rating system for oncologists (ie, medical oncologists and radiation therapists) were originally intended to evaluate a physician’s readiness to perform in a value-based contract; however, the model could readily be applied as COE criteria for a payer’s oncology network.
Using a combination of electronic data collection, site surveys, and on-site audits, ORN ranks providers on their ability and readiness to perform on 10 key performance indicators that represent value in oncology care, including:
Each measure is weighted to reflect the relative contribution to value, and the result is a score on a scale of 1 to 10, with 10 being the best.
ORN applied this methodology to its OMNI Healthcare study in 2015 in a pilot demonstration project. OMNI Healthcare is a multispecialty physician group in Brevard County, FL. During the 3-month study period (November 2014-January 2015), the oncology practice had 239 patients, 61 of whom were receiving chemotherapy, and 358 office visits and consultations. The practice was originally evaluated by ORN in November 2014 and received a baseline score of 2.91 on a scale of 1 to 10. Although the practice ranked well on measures related to patient access and coordination of care, its opportunities for improvement included compliance with evidence-based guidelines, consideration of cost, patient satisfaction, the completion of advanced directives, and end-of-life management.
ORN made a series of recommendations to OMNI Healthcare and assisted the practice with their implementation, including the use of an online clinical decision support tool and a patient satisfaction survey. Within 30 days, the practice improved its score by 27%. Within 90 days, its score jumped to 5.56, representing a 91% improvement. These initial efforts moved the practice out of the “needs improvement” range (scores, 1-4) and into the “meets expectations” range (scores, 5-8). There remains a need to fully transition the practice to a patient-centered model, which will raise the score to the “exceeds expectations” range (scores, 9-10).
Based on the OMNI Healthcare study, ORN monetized the value of a single shift in a practice’s value score so that each improvement equates to $1000 per 1 million members in monthly savings to a payer. OMNI Healthcare earned 23 shifts in its value score, representing a savings of $23,000 per 1 million members monthly in this market. Across an entire network of providers, the savings potential is compelling, and the contribution to patient care is significant. On a population of 50,000 health plan members, the total savings was estimated to be $3.5 million.
Rob Crosby, RN, the lead oncology nurse at OMNI Healthcare, said, “ORN has helped tremendously in our efforts to provide evidence-based, cost-effective treatment regimens to patients. ORN has also helped in ways of improving patient care and the ability to provide up-to-date and cutting-edge treatment and follow-up to treatment.”
The ORN business model differs from a traditional COE in that providers who do not meet their performance targets are not excluded from the network. Rather, the barriers to performance are assessed and remediated in terms of measuring, monitoring, improving, and repeating. Rather than limiting the network to high-performing providers, ORN raises the bar of performance across the network to drive value and to preserve access to care.
Although the 10 key performance indicators could be used in a more traditional COE or a narrow network model, ORN’s more inclusive approach offers an improved option, given the growing demand for cancer services, the projected workforce shortage of oncologists that will further limit access to care, and the cost pressures forcing the closure of smaller community-based oncology practices.6 The challenges to scale the ORN model include:
Provider engagement Practices are already overwhelmed with administrative burden driven from payer programs (eg, utilization management and prior authorization). The program must be minimally intrusive and produce meaningful, short-term value in the form of increased referral volume or shared-savings arrangements
The collection of unstructured data from practices Some practices are more sophisticated than others. Several technology-focused companies (eg, CliniCast, Flatiron Health) have taken on this challenge. However, a minimum technology threshold must be in place within a practice as a baseline, and some practices are still operating largely in a paper-based chart world
Obtaining comprehensive data on cost for individual practices The model is dependent on payer claims data, so to evaluate a practice, one would need claims data from all contracted payers
Obtaining enough data from a practice to make meaningful cost conclusions and comparisons Smaller practices will run into the problem of small numbers and are not in a position to control for the cost of ancillary services, putting them at an unfair disadvantage relative to larger practices with insourced ancillary services.
Although the ORN pilot was successful in creating value, the data collection and integration was manually intensive and was intended to prove initial concept points versus scale as a business model. Additional experience and market research will help to validate common themes that can be addressed through repeatable solutions with minimal customization, making the economics of the ORN business viable for the long-term.
The ORN business model offers promise to payers that are interested in improving care and driving value through a COE model for oncology. However, it remains to be seen whether the model can create meaningful, repeatable, and sustainable change in the cancer care delivery system. More proof points are needed, beyond this initial pilot, to better understand the barriers that providers face in moving from volume-based to value-based models, and if and how those barriers can be addressed on a repeatable, scalable basis.
The development of this white paper benefited tremendously from the experience and support of an outstanding group of advisors, representing the clinical and business sides of oncology. They provided valuable insights and served as a sounding board for this white paper, challenging and improving the paper through thoughtful feedback and suggestions. However, the views expressed in this white paper are exclusively those of the author and of ORN, and may not necessarily represent the views of the advisors. The advisors include Ron Barkley, JD, Cancer Center Business Summit; Linda Bosserman, MD, City of Hope Cancer Center; Jason Fisherman, MD, Synthesis Capital; and Ira Klein, MD, Aetna.
