Facilitators and Barriers to Implementing a Patient-Centered Oncology Care Model

Abstract

PURPOSE:

Oncology practices often serve as the “medical home” for patients but may not have systems to support all aspects of patient-centered care. We piloted a new set of oncology medical home standards that call for accessible, continuous, coordinated, and team-based care. We examined how adoption of the standards varies across a variety of practices and compared practice self-report with external evaluation of implementation.

METHODS:

Five medical oncology practices in southeastern Pennsylvania implemented the standards from 2014 into 2016. Implementation support included training webinars and technical assistance. External reviewers evaluated practices’ implementation of the standards. We conducted site visits to interview providers and patients.

RESULTS:

Between baseline and follow-up, practice self-assessments and independent audits showed practices increased implementation of the patient-centered oncology standards. The largest improvement was seen in continuous quality improvement (QI). Practices were less successful in implementing care coordination: achievement on two standards (access and evidence-based decision support) declined from baseline to follow-up. Qualitative analyses revealed that practices focused QI in five areas: goals of care, engaging patients in QI, financial counseling, symptom management, and care coordination. Interviewees talked about facilitators, such as leadership support and physician buy-in, and barriers to transformation, including inadequate resources and staffing. Health information technology both supported and limited implementation.

CONCLUSION:

Oncology practices showed some progress in their implementation of patient-centered care processes over the course of the pilot program. Systems for tracking and documenting improvement, training for staff and clinicians, leadership support, and alignment of financial incentives are critical to transformation.

The authors have declared that no competing interests exist.

INTRODUCTION

Despite advances in cancer treatments, there are gaps in quality of care that reflect nonadherence to evidence-based treatment, poor communication, inattention to care planning, inadequate symptom management, and poor coordination throughout the care continuum, including at the end of life.¹ The National Committee for Quality Assurance developed a set of standards for patient-centered oncology care² to articulate the capabilities oncologists should demonstrate, such as referral coordination, care coordination, and quality improvement (QI), and that emphasize key aspects of cancer care, such as comprehensive assessment and shared decision-making. These standards build on the framework of the patient-centered medical home³ for primary care and patient-centered specialty practice⁴ standards relevant to all specialists.

The patient-centered oncology medical home is designed to address problems in communication that often contribute to poor outcomes.^5-10 Studies show that clinicians ask for patient preferences in medical decisions only about half the time.^11,12 Because of poor or ineffective communication about goals of treatment, many patients undergoing treatment of advanced cancer do not understand that treatment is not aimed at curing the disease but rather at prolonging life or addressing symptoms.^13-15 The medical home also addresses lack of coordination between oncologists and other providers. Although medical oncologists are often considered the “captain” of the cancer treatment team, course of treatment may be decided during consultations with a primary care provider and surgeon, without the opportunity for a fully informed decision process and without development of a comprehensive treatment plan.^11,16 The lack of clarity in provider roles continues through treatment and survivorship.¹⁷

These medical home principles are also encouraged in the Centers for Medicare & Medicaid Services’ oncology value-based payment models, including the Oncology Care Model⁵ and upcoming Oncology Care First¹⁸ model, through the practice transformation requirements. Whereas the Oncology Care Model asks providers to attest that they have the structures and processes in place, the patient-centered oncology practice standards require providers to show documentation to demonstrate they have met the standards. This report describes the implementation of the patient-centered oncology model by five oncology practices and perspectives from the practice clinicians and staff on barriers and facilitators to implementation.

METHODS

We conducted a review of self-assessments on the standards at two points in time to assess transformation and site visits to learn about barriers and facilitators. We describe how patient-centered oncology care adoption varied across the pilot practices and which factors affected adoption. This project was reviewed and approved by the Chesapeake Research Review Institutional Review Board.

Participating Oncology Groups

Five practices in southeastern Pennsylvania volunteered to participate in the pilot program. The practices have between four and 41 full-time physicians; two are privately owned, two are at academic cancer centers, and one operates within a hospital-based outpatient department. Four practices have various electronic health record (EHR) systems; one has a paper-based system.

Patient-Centered Oncology Standards

The standards are described in detail in our prior publication.¹⁹ Participating sites began implementing the patient-centered oncology practice standards in January 2014. The implementation period, for purposes of the study, concluded in July 2016. During the 2.5-year implementation period, practices received support to implement the standards. We conducted monthly webinars during which we disseminated standardized content on implementation of the standards and encouraged practices to share successes and barriers. Topics were prioritized by the study team and pilot participants; often barriers brought up in one month’s webinar were explored in-depth during the following month’s webinar. We also provided technical assistance to each practice on the basis of their individual needs and barriers during biweekly QI check-in calls to troubleshoot how they could modify processes and reporting to meet the standards and brainstorm about how to take quality data and develop action plans to improve performance.

Data Collection

Self-assessments of implementing the standards.

Practices reported their self-assessment of implementation of the patient-centered oncology standards during the project baseline period from June to September 2014 and the follow-up period from September 2016 to January 2017.

Standards audit.

Independent reviewers audited the self-assessment and provided feedback to the practices. For the on-site audit at baseline, the reviewers determined whether practices met the criteria on the basis of whether they could show how their processes were implemented. The follow-up audit was conducted virtually; the auditor determined whether practices met criteria on the basis of documents or reports demonstrating implementation of the standards shown using a screen-sharing program. We focused our final analysis on nine elements identified as priorities for transformation by our multistakeholder advisory panel. The methodology was changed for follow-up to focus on the key elements critical for transformation identified during the course of the project and to lower the burden on practice staff to accommodate the study team on-site on multiple occasions during the follow-up data collection period.

Research site visits, patient case histories, and key informant interviews.

Two members of the research team conducted site visits (approximately a full working day, or 8 hours) at each pilot practice during baseline and follow-up. The start-up site visits included observation of at least three patient-physician interactions and six to 12 semistructured interviews (each lasting 30 to 60 minutes) with key informants at each site (n = 24), including physicians, nurses, administrators, and medical directors. These interviews focused on how the practice conducted QI activities, what processes were in place, what needed to be implemented or improved, and facilitators and barriers to implementing the standards. The number of interviews we conducted varied at each site on the basis of the size of the practice and availability of staff to speak with us during the site visit.

After the implementation period, we also conducted key informant interviews (lasting 30 to 60 minutes) by telephone with physicians, nurses, administrators, medical directors, and patient advisors involved with the QI efforts. These interviews focused on understanding practices’ experience with implementing the standards and perceptions of impact and future sustainability (n = 21). During the follow-up site visits, we conducted in-depth interviews (lasting 60 minutes) with five patients and members of their care team at each practice. Through these interviews, we sought to understand the course of treatment from diagnosis through care planning, active treatment, posttreatment, and survivorship; and how patients interacted with their care team, particularly in terms of decision-making, care planning, and symptom management, as well as if they had noticed any changes in their practice over the course of seeking care (n = 26 interviews). With the patients’ permission, we also spoke to members of their care team to ask about their medical journey in the practice and the types of information that was documented as a part of their care process (n = 18 interviews; some clinicians cared for multiple patients).

Analysis

We recorded all interviews with permission and obtained professional transcriptions. All interview transcripts were analyzed using NVivo, version 11.0 (QSR International, Melbourne, VIC, Australia), for themes related to transformation and processes required to meet the standards. Notes from the site visits and results of the practice standards audits were compiled and analyzed.

RESULTS

Implementation of Patient-Centered Oncology Standards

Between baseline and follow-up, practice self-assessments (Fig 1A) and independent audits (Fig 1B) generally showed practices increased implementation of the patient-centered oncology standards. Patient-Centered Oncology Standards are listed in Figure 1. Based on data from the audit, practices fully implemented functions related to communicating the role of the oncology practice (specialty practice responsibilities [2C]); assessing patient needs and preferences (comprehensive health assessment [3C]); defining team member roles, communication, and training (practice team [2E]); and patient-centered care planning (care planning and support self-care [4A]). The largest improvement was seen in continuous QI (6C): 22% of factors were met at baseline versus 89% at follow-up.

FIG 1.

(A) Self-assessment results for implementation of Patient-Centered Oncology Standards. (B) Audit results for implementation of Patient-Centered Oncology Standards. For the baseline audit, the practices were asked to show how their processes were implemented but were not asked to provide documents or reports. During the follow-up audit, practices had to show documents or reports that satisfied criteria for each element.

The audit suggested that practices were less successful in implementing care coordination systems (47% factors were met at baseline v 57% at follow-up). Achievement on two standards declined from baseline to follow-up: access (2A; 85% at baseline v 75% at follow-up) and evidence-based decision support (3E; 75% at baseline v 65% at follow up). Within the access standard, four of the practices were unable to demonstrate provision of nonvisit consultations with referring clinicians. Two practices were unable to demonstrate provision of timely clinical advice to patients who contact them when the office is closed. Within the evidence-based decision support standard, none of the practices could demonstrate that they adopt and update shared decision-making aids.

At follow-up, the practices reported they had more factors implemented than the auditors reported, particularly for elements on access (2A; 95% by self-assessment v 75% by audit), evidence-based decision support (3E; 85% by self-assessment v 65% by audit), medication management (4B; 97% by self-assessment v 90% by audit), and coordination of care transitions (5C; 93% by self-assessment v 67% by audit).

Transformation Across Practices

During our site visits, we identified five major areas that were the focus of practice QI efforts (Table 1): (1) goals of care and patient preferences; (2) QI and engaging patients; (3) financial counseling; (4) symptom assessment and management; and (5) care coordination. The first three topics correspond to areas of the patient-centered standards where both the practice self-assessments and auditor assessments demonstrated improvement in systems.

TABLE 1.

Summary of Domains of Quality Improvement, Barriers and Facilitators from Site Visits and Key Informant Interviews

Goals of care and patient preferences.

At baseline, none of the practices had systematic or documented discussions about patient preferences and goals of care during treatment. By the time of the follow-up interviews, all the practices reported, and auditors confirmed, that they had a process to collect information from patients about their preferences for quality-of-life issues during treatment and beyond, and that they document this information and use the documented discussion as a tool for following up on whether patient goals were being met. Staff at one practice said implementing this process had improved patient-clinician communication and relationships. The goals-of-care discussion was also meaningful to patients: “It gave me the ability to say ‘where am I spiritually? What do I need to do here?’ And also to tell my family, ‘I’m at the end of the line.’”

QI and engaging patients in the process.

At baseline, only one of the five practices had a systematic QI process or team to implement and test changes in the practice. By the end of the project, all the practices had formed QI teams and were using systematic methods to pilot, test, and implement processes. Despite having formal patient-family advisory councils (PFACs), two of the practices did not have systems for getting patient and family feedback before implementing new or changed workflows. During the pilot, these two practices started engaging the PFAC regularly, and the other practices identified individual patient advisors to provide feedback on workflows and test new forms. Practice staff, clinicians, and administrators found patient feedback useful for making changes.

Financial counseling.

At baseline, the provision of financial counseling services varied across the practices. Some practices identified patients with financial needs before the start of treatment to provide assistance, whereas other practices addressed issues only after patients had trouble paying their bills. By the end of the pilot program, all the practices had processes to connect patients with a billing representative, social worker, or financial counselor to discuss their health insurance coverage and potential out-of-pocket costs proactively during treatment planning. Staff at all the practices said they directed patients who need assistance to pharmaceutical companies’ copay assistance programs. At one practice, a social worker connected patients with an organization that helps with household bills. Patients reported positive experiences with the financial counseling assistance

Symptom assessment and management.

At baseline, symptom assessment and management varied across practices as well as among providers within practices. Some practices had standardized symptom-management protocols for triage nurses to follow when patients called; however, not all were comprehensive. Calls were documented in the patient record, but none of the practices summarized the nature or disposition of the call (eg, symptoms, disease site, referral to emergency department). At the end of the pilot program, all the practices reported that they had systematic patient education about managing symptoms as well as specified triage protocols (however, not all these systems were demonstrated to the external reviewers). One administrator said their philosophy of practice had changed to be more patient-centered in that they had improved and systematized the practice’s triage process to try and keep patients out of the emergency department despite not having any incentives.

Care coordination.

In our prior work,²⁰ we identified care coordination as an area of opportunity for structural improvement because transitions and flow of information between providers were not always systematically documented. All the practices actively worked to improve workflows, communication, and care transitions by implementing systems to send information to primary care physicians, other specialists in the patient’s care team, and other referring providers. Several patients noted positive experiences: “My oncologist and my endocrinologist, they send electronic stuff to each other so you don’t have to go and repeat your story.”

Overall impact of the pilot program.

By the end of the project, many of the staff, clinicians, and administrators, and sometimes patients, felt the pilot program resulted in a shift to patient-centered care, better information for patients, and better care coordination.

Facilitators and Barriers to Transformation

During the interviews we conducted at follow-up, key informants, including administrators, staff, and clinicians, talked about facilitators, such as leadership, and barriers to transformation, including inadequate resources and staffing. Health information technology both supported and limited implementation (Table 2). Informants mentioned that leadership support and physician and staff buy-in were important factors in implementation success.

TABLE 2.

Summary of Facilitators, Barriers, and Impact on Transformation at Baseline and Follow-Up

Practices cited lack of resources in terms of time, staff, and money as an issue for both making changes and providing better care. At follow-up, practices reported they often did not have adequate staff to meet these needs. In addition, the practices did not receive differential payment or other incentives for participating in the pilot program. Therefore, the additional coordination and navigation became additional responsibilities for existing staff, which was a particular challenge for smaller practices.

All practices reported that implementing a QI process was difficult. Some administrators had problems getting the right personnel to implement QI. There was sometimes confusion over how to reimagine current staff roles, create job descriptions, and implement new roles for existing staff. Another challenge was that not all staff and physicians knew about the new processes. Nurses across most of the practices were often burdened with the increased need for documentation of processes; practice administrators at two sites said these changes contributed to staff turnover.

Staff, clinicians, and administrators also acknowledged that developing, implementing, and getting used to new processes takes time. One practice reported that transition of ownership of the practice also led to staff and physician turnover that may have affected patient care and continuity. One administrator said keeping up with rapid advances in oncology in terms of new drugs and regimens and having the ability to incorporate these into the practice were challenges.

Practice staff and clinicians reported that EHRs and patient portals both supported and inhibited implementation of patient-centered care. Three of the practices had implemented patient portals; however, the portals were limited in the information they provided to patients. The existing patient portals included basic information, such as laboratory results, appointments, diagnoses, medications, and other basic medical history information. Lack of integration with the EHR limited patients’ ability to access relevant resources in a central place. The large academic centers could not make changes to their EHR to document some of the new processes they had implemented. Several of the practices found it difficult to collect data and give feedback to physicians, given their current health information technology. It was also a challenge to navigate documentation and data reporting through multiple clinical information systems (eg, scheduling, navigation, outpatient, ordering chemotherapy). One practice was still using paper charts. However, having an EHR to document processes systematically and report quality metrics was also a critical driver of success in implementation of the patient-centered oncology standards. For those practices able to configure their systems and implement the standards, EHR technology was a big advantage for reporting.

DISCUSSION

Over the course of this demonstration, oncology practice self-assessments and independent audits generally showed increased implementation of the patient-centered oncology standards. Clinicians, staff, and administrators identified specific changes in care they believed better addressed patient needs and improved care, including efforts to track patient goals of care, engage patients in QI, offer financial counseling, manage symptoms more actively, and coordinate care more effectively. The largest improvement was seen in continuous QI (6C), largely due to the establishment of QI teams and processes to implement changes and review quality metrics, as well as provide feedback to physicians. Achievement on two standards (access and evidence-based support) declined from baseline to follow-up, likely due to constraints on resources during the project. Practices were not able to increase staff and resources to meet the standards and the needs of patients.

Transformation results reveal a critical gap between the audit and self-assessment on standards: practices underestimated (at baseline) or overestimated (at follow-up) their capabilities. Value-based payment models such as the Center for Medicare and Medicaid Innovation Oncology Care Model rely on attestation to measure patient-centered care principles. Our results indicate we should be cautious when interpreting practice-reported results without consideration of other data sources.

These results are consistent with early studies of the patient-centered medical home that showed financial incentives are needed to drive improvement in patient experiences, and reductions in use and time are required for outcomes to improve.¹² A higher level of care coordination requires more staff and resources, and neither this pilot project nor payers offered pilot practices increased reimbursement fees for implementing the patient-centered care model. Recent studies have shown that where there has been greater investment in systems for population management or alternative payment strategies, there have been decreases in costs and use and increases in quality of care and patient satisfaction.^6-10,20-22 Early results from the Oncology Care Model demonstration have also shown that practices report developing care plans, coordinating care, improving survivorship planning, and improving end-of-life care, among other efforts to deliver more patient-centered care for their patients. However, overall cost and savings findings do not yet show meaningful impacts.²² Our findings²³ help interpret the quantitative analysis of the impact of the Patient-Centered Oncology Standards demonstration (data not shown). Overall, we found no statistically significant differences in hospitalizations, emergency department use or primary care visits between the pilot program and comparison group over time, despite improvement in performance on selected quality measures and the implementation of the standards reported here.

There are some important limitations to this work. We did not have information about oncology practices in the region that were not involved with our pilot program. The five practices that volunteered to participate in this project may have been more prepared to implement patient-centered processes than practices that did not participate. Last, at baseline, we conducted an on-site audit and at follow-up we conducted a virtual audit of a limited set of standards that were designated as high priority. The change in data collection methodology may have affected our results.

In conclusion, the five practices in this pilot program showed progress in their implementation of patient-centered care processes. Practices reported specific efforts to elicit and document patient goals, address financial needs, improve care coordination, and engage patients in QI. Key informants at practices felt these efforts improved their overall commitment to patient-centered care and improvement in specific aspects of care. Our findings suggest that patient-centered oncology standards can support practice improvement efforts. However, barriers such as limitation of electronic data systems and lack of resources limited progress. Payment models are needed to support these transformation efforts.

AUTHOR CONTRIBUTIONS

Conception and design: All authors

Collection and assembly of data: Manasi A. Tirodkar, Lindsey Roth, Sarah H. Scholle

Data analysis and interpretation: All authors

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Facilitators and Barriers to Implementing a Patient-Centered Oncology Care Model

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/op/authors/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).

Mark W. Friedberg

Other Relationship: Blue Cross Blue Shield of Massachusetts

Sarah H. Scholle

Employment: Rally Health/United Health Group (I)

Stock and Other Ownership Interests: United Health Group (I)

No other potential conflicts of interest were reported.