Skip to main content

Advertisement

Advertisement

ADVERTISEMENT

Clinical Pathways Forum

Dana-Farber Pathways: The Purpose, the Journey, the Solution

Founded in 2016, the Clinical Pathways Forum is a community of pathways professionals­—now totaling over 12 institutions from across the United States and Canada—who are utilizing clinical pathways in their practices and institutions to improve cancer care. Forum leader Mishellene McKinney, MHA, RN, OCN, organizes quarterly conference calls with Forum members to facilitate discussion of shared experiences and lessons learned regarding pathway use as clinical pathways become more prevalent and evolve to meet the needs of value-based health care systems and reimbursement models.


The mission of the Clinical Pathways Forum is to facilitate a knowledge exchange for overcoming the challenges of developing, implementing, and measuring clinical pathways to demonstrate the value of standardizing clinical care. The main activity of the Forum is to schedule time quarterly for conference calls to share experiences and lessons learned using clinical pathways. Health care professionals from across the continuum of care are encouraged to join in these collaborative discussions. Forum contact information can be found in the online version of this article.

The Forum publishes highlights from each of the Forum conference calls that occur throughout the year in the Journal of Clinical Pathways. This fourth installment summarizes the speakers and discussion from the August 2019 call, in which Carole Tremonti, RN, MBA, senior director of pathways operations at Dana Farber Cancer Institute (DFCI) spoke about their new clinical pathways platform, Dana-Farber and Phillips IntelliSpace Precision Medicine.

The Dana-Farber Pathways Journey

DFCI entered the pathways space in 2012 led by Joseph Jacobsen, MD, who is currently the Chief Quality Officer at DFCI. At the time, Dr Jacobsen was a general oncologist, and he was struck by the lack of standards that existed beyond broad guidelines in medical oncology. With physicians rotating between different campuses and seeing patients in different environments, DFCI wanted to ensure that the treatment and guidance patients were getting would be consistent no matter where they received care. Pathways help to limit unwarranted variation, but more broadly, the organization wanted to be able to share expertise and move individual physician experiences beyond the borders of their clinical office.  

DFCI held a retreat with operational and clinical leadership where the clinicians voted to establish practice standards through the use of clinical pathways. DFCI chose Via Oncology (then owned by the University of Pittsburgh Medical Center and purchased by Elsevier in 2018). Ms Tremonti described Via Oncology as “a wonderful business partner for years.” Initially, DFCI used the Via platform with semi-customized content that relied on the Via branching structure but gave DFCI the ability to change the treatment plan endpoints. After about a year of this hybrid approach, DFCI determined that they needed more autonomy over the branching logic. Via supported them in creating their own pathway algorithms but using the Via platform as a delivery tool. Over an 18-month period, DFCI built out fully customized medical oncology pathways that were implemented in 2017. They then repeated that process across radiation oncology, all the while focusing on how they could enhance analytics.  

In 2017, DFCI faculty Jackman et al published “Cost and Survival Analysis Before and After Implementation of Dana-Farber Clinical Pathways for Patients With Stage IV Non-Small-Cell Lung Cancer.” The study demonstrated a reduced cost for metastatic non-small cell lung cancer patients on pathway, with no difference in survival.1 Ms Tremonti acknowledged that there were limitations in the study as they did not have a way to compare themselves with other organizations, which is the question that payers want to understand. To address this need, DFCI recognized that they had to find a way to be able to answer these questions effectively.

As DFCI began to move into the radiation oncology space, Ms Tremonti noted that both DFCI and Via knew that “at some point we might outgrow each other.” DFCI’s vision required a system that could power their analytics capacity and be able to spread DFCI content nationally and internationally. They started looking for an organization that was interested in being a development partner, and Philips emerged as that solution. Philips is an informatics and technical leader within health care. As DFCI was seeking a global technical and commercial partner, Philips was also seeking an oncology partner to power its informatics portfolio and complement its strategy in precision diagnosis, subsequently aligning the goals both organizations sought to achieve. In May 2018, DFCI and Philips began collaborating in the design, development, and eventually implementation of an entirely new informatically and analytics-driven pathway platform over the next 13.5 months.    

The Business Case: Data to Drive Oncology Care

Ms Tremonti discussed how data-driven care is a foundational principle at DFCI: “The reason I moved from research into quality was because research was too slow. What’s the most important component of research and quality? It’s data. You can only manage what you can measure.” There were three guiding principles that drove the development of the product: (1) support and influence; (2) capture and curate; and (3) analyze and assess (Figure 1). To support and influence, data need to be informed, timely, and transparent. To achieve the goal of capturing and curating, a critical use case was formed: if the data was not already structured, the plan was to structure it; if it did exist in a structured format, somewhere, it should interface to the platform and then auto-populate into the pathways system for the providers. In describing the new platform to the Forum, Ms Tremonti explained that it allows for decision support and can make recommendations, learn responses to therapy, learn how people use the platform, and eventually will be able to analyze outcomes.

f1

Aggregating the clinical data is important. Ms Tremonti gave us her opinion on the current artificial intelligence (AI) and machine learning revolution in health care: 

"I don’t believe AI exists in the space of health care. It exists in a lot of other spaces, like finance, where there is a lot of structured, defined data that has a very binary, yes-or-no definition. But health care data is dirty. It’s messy, doesn’t exist, or it’s in a narrative note or entered in a way that we can’t access it. Right now, in this interim space, what I believe in is adaptive intelligence: AI combined with human intellect."

Ms Tremonti advocated for adjudicated scientific content balanced with human intelligence. She argued that clinical care requires the intervention of a clinician and that systems need to be adaptive because most of our electronic medical records (EMRs) do not support the data requirements to be able to move nimbly. 

A Platform for Physicians, by Physicians

The new platform represents a product that was built by clinicians for clinicians. Much of the feedback the pathways team received about the legacy system included comments many health care professionals are familiar with: “There are too many clicks; I want to see how I got to this decision; I don’t think this way about patient care.” DFCI wanted to be mindful of these types of concerns and ensure the needs of the users were represented from the very beginning of development. To accomplish this, they formed a panel of 25 physicians that met monthly for one year to design the product. 

One use case that was critically important to the physicians was visibility of the decision logic—they wanted to see where they came from and how they got there. The navigation needed to be easy and fast, but it also had to be fully integrated with any EMR and/or any data warehouse. The platform needed to be able to pull nimbly from different sources. This was achieved through the use of ontologies to standardize and manage oncology data across multiple cancer diagnoses, the establishment and availability of structured data, as well as an investment in Natural Language Processing to create structured data where it is otherwise missing. Ms Tremonti explained that, once the data is discreet and available, the data can be automatically pulled into the system, ie, as the provider is navigating that branching logic, if that data point is known, it is already populated.  

Key components from the old system that users liked were kept, such as the much-loved and appreciated consent form and treatment plan summary document. The new pathways program represents over 1000 different treatment plans, and for every different treatment there are treatment-specific side effect profiles that are curated by a physician and pharmacist and put into sixth-grade reading level language. This content is dropped into the consent form and the treatment plan summary. The treatment plan summary is a document that supports the prior authorization process, provides transparency for the nurses, pharmacists, and the patient education process.

“This looks how I think” 

The first nugget of feedback Ms Tremonti’s team received after implementation on July 29, 2019, was This looks how I think. She proudly stated, “To me, this was the biggest win. That’s really what the 25 physicians were trying to go for; they wanted the system to think like they did. They thought it was easy to use; it was intuitive; they liked that they could see the branching logic.” Ms Tremonti acknowledges that the product “wasn’t perfect” but emphasized that it is still a beta platform. For example, the providers were very specific about wanting to have patient context in the system, so that when a user is in a patient record, the pathway knows the patient and knows the user. That change meant moving away from having a list of patients in the old system. Ms Tremonti laughed and said, “Yesterday they let us know, ‘we really miss having our list of patients for that day that we can choose from.’” Ideas about how a program should work are sometimes good in theory but may be faulty in practice. The development team is eager to hear more user feedback to iron out the small but important details like this, which will be critical as DFCI and Phillips refine the product for commercial use. 

Bridge to Tomorrow

The pathways development team at DFCI sought to build a pathways platform that would take the organization into the future. Now, said Ms Tremonti, the focus is to move beyond the treatment decision and create a symptom management interface and develop an analytic platform with patient-reported outcomes that can feed back into the platform to drive decision making.  

Ms Tremonti closed her discussion by recognizing that the success of the implementation was facilitated by the vast number of people that contributed to the design and development of the product, including Philips, the physicians, the pathways team, population health, pharmacy, information science & informatics, and executive leadership.

Reference

1. Jackman DM, Zhang J, Dalby C, et al. Cost and survival analysis before and after implementation of Dana-Farber clinical pathways for patients with stage IV non–small-cell lung cancer. J Oncol Pract. 2017;13(4):e346-e352. doi:10.1200/JOP.2017.021741

Advertisement

Advertisement

Advertisement