Publication
Article
Evidence-Based Oncology
Author(s):
UPMC Oncologists: Pioneers in the Clinical Pathways Movement
With 180 oncologists practicing at more than 30 locations, UPMC Cancer Centers is one of the largest networks of cancer specialists in the country. In the early 2000s, the oncologists at UPMC recognized the importance of consistency and quality of care and the difficulties of ensuring a uniform level of care across their large distributed network. At the same time, they found themselves caught in the crosshairs of healthcare plans’ efforts to ratchet down the rising costs of cancer care. Together, these factors drove UPMC oncologists to develop clinical cancer pathways. Clinical pathways are treatment plans that lead physicians to 1 preferred treatment for a given state and stage of disease.1 Pathway programs are believed to be a cost-effective way to provide evidence-based patient care.2,3 Over the past several years, UPMC has continued to use, improve, and expand its pathway program. For UPMC, clinical pathways are a way to ensure that all patients going to any UPMC location get the same care.
Prompted by its success and a growing interest in pathways among the broader oncology community, UPMC commercialized its program. This move made its well-founded pathways, now known as Via Oncology Pathways, available to practices across the United States and internationally. To date, UPMC is the biggest customer of the commercial pathway service. The added contributions of external practices on the pathways committees are considered a valuable commodity, according to Kathy Lokay, president and chief executive officer of D3 Oncology Solutions, the UPMC affiliate that now owns and operates the Via Oncology Pathways. To date, practices in 11 states, India, and Ireland have purchased the Via Oncology Pathways, with many more in active discussions.
Peter Ellis, MD, director, Medical Oncology Network of UPMC Cancer Centers, explained that commercialization allows UPMC to leverage costs to expand the pathways to more cancer types, more modalities of care, and more phases of care. The pathway program, which initially focused on chemotherapy protocols for 3 or 4 diseases, has grown to include 95 percent of cancer incidences and now supports decisions surrounding prognostic testing, imaging, radiation therapy, and supportive care. In 2011, pathways were launched for gastric cancer, bladder cancer, and chronic myelogenous leukemia. The group plans to roll out a testicular cancer pathway this year.
Keys to Success
The success of UPMC’s clinical pathway program is attributed to several operational and philosophical aspects.
Decision Support
In 2006, a software application was developed to provide real-time decision support via a user-friendly Web-based interface, referred to as the pathway portal. In effect, the software does the work of sorting through all the clinical algorithms for the physicians. The software also provides back-end information that serves as a valuable reporting source of user data, pathway adherence rates, and clinical trial accrual information.
Today, the pathway program is viewed as a software venture as much as it is a clinical content initiative. Lokay commented, “You can have all the best clinical content in the world, but if it sits on a shelf or a static website, you really don’t have anything to show for it, nor are you really driving that decision support.
“With the growing complexity of cancer diagnosis, workup, and treatment in the age of personalized medicine, pathways are a critical tool to helping the typical community-based oncologist stay abreast of the data,” explained absence of good data, the goal is to drive standardization. UPMC is home to the University of Pittsburgh Cancer Institute, a nationally acclaimed and recognized research center. So, it is not surprising that clinical trials are held in high regard and positioned prominently in the Via Oncology Pathways. The decision support tool displays open clinical trials upstream of relevant treatment recommendations. Trial accruals are counted on pathway. If a physician does not accrue to the trial, they are asked to provide the reason for that decision from a structured list. While much attention has been paid to the idea that UPMC, and now Via Oncology, adhere to a single treatment pathway model, this is often incorrectly interpreted to mean there is only 1 option or pathway for each disease. The reality is much more complex. The lung cancer pathways, for example, include 18 main branches with an additional 41 suboptions branching off of those. The committees stratify each disease by a number of attributes such as stage, histology, molecular pathology, and others. They derive 1 single best treatment for the general presentation of each unique intersection of those attributes. This process is executed with the rec- Ellis. In anticipation of these impending challenges, Via Oncology Pathways recently updated its extensive IT platform. The transformed platform will allow them to manage and update thousands of branches.
Pathway Development and Review
Transparency throughout the pathway development process is maintained by (1) opening disease committees to all physicians within practices using the pathways, (2) providing evidence reviews that detail the data and rationale behind pathway decisions, and (3) adhering to a clear conflict of interest policy. “We take conflict of interest very seriously and we have positioned ourselves so that we have no dependency whatsoever on the pharmaceutical industry that would cause a conflict of interest, whether perceived or real,” emphasized Lokay.
Thirty-three well-established disease committees—16 medical oncology committees and 17 radiation oncology committees— meet quarterly to review the pathways. Each committee is chaired by 2 physicians, 1 academic and 1 community- based. Separate medical oncology and radiation oncology committees operate for each disease, with 1 representative from each serving on the other’s committee. While Via has not yet developed surgical pathways, surgeons serve on committees for several diseases, especially those for which surgery is integral to care.
The inclusion of academic doctors in the Via Oncology Pathway decisionmaking process is somewhat of a distinguishing practice. It is a long-standing philosophy of UPMC’s pathway program that the process is best served by allowing the physicians who live by the pathways to design the pathways.
The disease committees strive to make evidence-based decisions whenever possible, but they do not adhere to a prescribed grading system. According to Brian Crandell, pathway pharmacist at Via Oncology Pathways, “We strive for the Phase III randomized trials, but unfortunately a lot of oncology drugs have been developed without a randomized trial against what is currently the standard of care.” In these situations, the committee may turn to meta- analyses for guidance, combine trial evidence, or make consensus-based recommendations.
Decisions are based first on outcomes. If there is no clear winner in terms of outcomes, toxicity is evaluated. Cost is considered when there is no clear winner in terms of efficacy and toxicity. There are not many cases where it comes down to cost. “Usually you get to a point where you go, wow, okay, this regimen is much more toxic than the other one,” said Lokay. In the ognition that the single best treatment may work for approximately 60 percent of patients. absence of good data, the goal is to drive standardization.
UPMC is home to the University of Pittsburgh Cancer Institute, a nationally acclaimed and recognized research center. So, it is not surprising that clinical trials are held in high regard and positioned prominently in the Via Oncology Pathways. The decision support tool displays open clinical trials upstream of relevant treatment recommendations. Trial accruals are counted on pathway. If a physician does not accrue to the trial, they are asked to provide the reason for that decision from a structured list.
While much attention has been paid to the idea that UPMC, and now Via Oncology, adhere to a single treatment pathway model, this is often incorrectly interpreted to mean there is only 1 option or pathway for each disease. The reality is much more complex. The lung cancer pathways, for example, include 18 main branches with an additional 41 suboptions branching off of those.
The committees stratify each disease by a number of attributes such as stage, histology, molecular pathology, and others. They derive 1 single best treatment for the general presentation of each unique intersection of those attributes. This process is executed with the recognition that the single best treatment may work for approximately 60 percent of patients.
Next, they consider whether there are other subgroups of patients for whom that therapy is not going to be a good fit, for example, patients who have poor performance status, can’t travel for weekly therapy, or have a contraindication to the primary treatment option. A subpathway is created for more commonly occurring profiles. Lokay clarified, “So, at the end of the day, for that node of the pathway, we may have 5 options, but 1 of those is considered the best for the more generalized population and the other 4 are suboptions for subpresentations. So, no, we don’t have just 1, but we do try to drive to 1 single best as a first choice.”
This approach allows solutions to be teased out when, from time to time, physicians call for more than 1 pathway option. Via believes that these requests are essentially the oncologists’ way of saying they have different kinds of patients that they see and they need more than 1 option. Crandell explained, “There is a logic process the committee goes through…and it is just a matter of us figuring out how to explain that in the software, through questions or getting the committee members to define more specifically when they would use each drug.”
In the 2011 calendar year, UPMC surpassed its on-pathway adherence goal of 80 percent. On-pathway decisions are considered to be: (1) treatment decisions made according to pathway recommendation and (2) appropriate decisions to take the patient off pathway, for example, upon reaching the last line of treatment.
Why Pathways?
When asked about the benefits of pathways, Ellis answered, “Our message to oncologists is straightforward. Payers are demanding a solution to the rising costs of cancer. While technology is the primary driver, not physician fees, the physicians are caught in the middle.” He went on to state, “It is imperative that physicians be the drivers of quality and cost containment for cancer, and if given the right tools for decision support and measurement, they can affect improvement far better than third-party intermediaries.”
A recently published study, led by UPMC Cancer Centers investigators and Via Oncology Pathways, reported that implementation and auditing of clinical pathways promoted uniformity of care across both academic and community care centers.4 In addition to reducing variability in care, pathways indirectly drive down cost by promoting the least toxic treatment options, according to Lokay. Detailing the benefits of UPMC pathways, Ellis mentioned the findings of 2 studies with Highmark Blue Cross Blue Shield, and 1 with IntrinsiQ, which demonstrate “a bending of the cost curve where the total cost of care for patients seen in pathways practices grows at a slower rate than the same costs for patients seen at non-pathways practices” (Table).
More recently, UPMC/Via Oncology has partnered with Highmark Blue Cross and Blue Shield, the oldest and largest healthcare plan in New Jersey, to operate a pilot project that supports and evaluates Via Oncology Pathways in 2 oncology practices in New Jersey. Preliminary cost analyses for the 2 pilot practices indicate successful results in terms of total cost of lung and breast cancer care for the pathway practices relative to non-pathway practices. Results from the project with Horizon went on to state, “It is imperative that physicians be the drivers of quality and cost containment for cancer, and if given the right tools for decision support and measurement, they can affect improvement far better than third-party intermediaries.”
More recently, UPMC/Via Oncology has partnered with Highmark Blue Cross and Blue Shield, the oldest and largest healthcare plan in New Jersey, to operate a pilot project that supports and evaluates Via Oncology Pathways in 2 oncology practices in New Jersey. Preliminary cost analyses for the 2 pilot practices indicate successful results in terms of total cost of lung and breast cancer care for the pathway practices relative to non-pathway practices. Results from the project with Horizon BCBSNJ are expected to be published later this year.
Early on, pathway proponents envisioned a scenario where payers who were benefitting from pathway-generated savings in drug- and hospital-related expenses would be willing to proactively contract around and support pathways. The reality is that payers generally have not taken an interest in funding new initiatives. While this may have been disappointing to some, Lokay sees it as a good thing. “If the payers are calling the shots on which pathways to buy/hire, I don’t know how an oncology practice can cope in that world….I think we want practices picking the pathways they like and feel the most comfortable with.”
The natural evolution has been that pathways have become a way for oncologists to demonstrate, “This is the way we care for our patients, all of our patients.” Along those lines, practices are using pathways to forestall payers from introducing measures such as rate cuts, prior authorization, or specialty pharmacy.
Health reform vehicles such as medical homes and accountable care organizations are a big factor in why more practices have started looking at pathways programs. Practices are investing in and adopting pathways to better position themselves in their local market. To some extent, practices use pathways to demonstrate that their patients are receiving evidence-based and personalized care that takes into account things like toxicity and cost. “We have a number of practices that are jockeying to make sure that they either keep all their existing referral sources or maybe actually in the process grow their referral sources by attracting primary care medical home docs,” said Lokay.
What does the future hold for the pathway program at UPMC? Dr Ellis offered his insights: “We are already on track to achieve the vision of pathways that integrate seamlessly with the major electronic medical records and cover virtually all cancer types, phases of care, and modalities of care. We see the pathways being chosen by the practices, not the payers! We see the ability to measure outcomes that prove that standardization to the best evidencebased medicine improves patient outcomes and reduces costs.”Funding Source: None.
Author Disclosure: The author reports no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.
Authorship Information: Concept and design; drafting of the manuscript; and critical revision of the manuscript for important intellectual content.1. Holcombe D. Clinical pathways programs: confusing choices for payers and physicians: part1: selecting the appropriate pathways program. JOMCC. 2010;3(5):10-11.
2. Hoverman JR, Cartwright TH, Patt DA, et al. Pathways, outcomes, and costs in colon cancer: retrospective evaluations in two distinct databases. J Oncol Pract. 2011;7(suppl 3):52S-59S.
3. Neubauer MA, Hoverman JR, Kolodziej M, et al. Cost-effectiveness of evidence-based treatment guidelines for the treatment of non-small-cell lung cancer in the community setting. J Oncol Pract. 2010:6(1):12-18.
4. Beriwal S, Rajagopalan MS, Flickinger JC, Rakfal SM, Rodgers E, Heron DE. How effective are clinical pathways with and without online peer-review? an analysis of bone metastases pathway in a large, integrated national cancer institute-designated comprehensive cancer center network [published online ahead of print January 13, 2012]. Int J Radiat Oncol Biol Phys. doi:10.1016/ijrobp.2011.09.056.
5. Ellis PG. All pathways are not created equal. Oncology Times. 2010;32(10):46-48. doi:10.1097/01.COT.0000381228.24720.6b.
Real-World Treatment Sequences and Cost Analysis of cBTKis in CLL