By Gary C. Bird, Ph.D., and Amy Mullins, MD, CPE, FAAFP, American Academy of Family Physicians
PerformanceNavigator was developed by the American Academy of Family Physicians (AAFP) as a novel multimodal CE activity that was national in scope, and designed to improve family physician learner performance, quality of care, and patient outcomes for one or more of the following chronic cardiometabolic conditions: diabetes; hypertension; or dyslipidemia. In total, the activity consisted of three separate online modules — one for each disease state, along with a single unifying live event and additional online learning. This activity represents the first of its kind and a breakthrough in harmonizing learning in a blended format.
The cardiometabolic triad of diabetes, hypertension and dyslipidemia together form a significant threat to long-term health. Over 100 million people in the U.S. have or are at risk for one or more of these diagnoses. This number is set to dramatically increase over the next decade. Despite the high expenditure associated with these diseases and the increased risk for a spectrum of causally related severe long-term adverse effects, only a fraction of patients are meeting recommended health goals.1
Effective primary care can be achieved when physicians are part of a collaborative practice level effort (including staff, educational materials and available electronic infrastructure resources) that embeds a patient-centric approach into the workflow.2 Such practices focus on health outcomes by coordinating care with other specialty providers and are proactive in reaching out to contact patients to monitor their progress or to remind them of the need for follow-up care. Furthermore, during an office visit, they maximize impact by following evidence-based guidelines, efficiently structuring the visit, and building strong relationships between the care team and the patient. Thus, improving understanding and implementation of the patient-centered concept presents an opportunity for changing cardiometabolic condition outcomes.3-5
The overarching educational goal of the activity was to provide a novel framework for simultaneously improving clinical and systems-based aspects of family physician practices regarding management of patients with cardiometabolic conditions, while facilitating a patient-centered approach to the care strategy. Outcome measures related to this goal included changes in the utilization of systems and teams, inclusion and improvement of the patients’ perspective on the care they received, and increases in learner clinical knowledge, competence, and performance.
PerformanceNavigator’s Blended Learning Approach
The design of this activity utilized an innovative hybrid (blended) learning format of both asynchronous/enduring online and synchronous live educational modalities that combined performance improvement (PI) and quality improvement (QI) methodology with live peer-to-peer practice deficit focused education. Using an online platform, learners completed case vignettes, abstracted patient and practice data, viewed enduring evidence-based presentations, and implemented an action plan with their care team. The measurement tools scored entered data using Likert-type ordinal scales, yes/no nominal questions, or for measurement of clinical performance/procedures as the percentage of patients who were appropriately managed. The self-guided activities were coupled with a live on-site workshop that taught how to interpret the practice data generated in the initial asynchronous phase and how to utilize this data to formulate appropriate individualized solutions to identified practice gaps.
The primary audience was family physicians; the secondary audience was members of their care team. Registrants for the activity came from across the U.S.
Dates of Course and Credit
Online module registration began in June 2015 with the live workshop held in Carlsbad, California, Nov. 4-6, 2015. The activity was accredited for AAFP Prescribed Credit and AMA PRA Category 1 Credit™ for a period of one year. Each disease state module was also approved by the American Board of Family Medicine (ABFM) as a Maintenance of Certification for Family Physicians (MC-FP) Part IV, Performance in Practice (PIP) Module.
PerformanceNavigator is unique in that it represents an evolution from the three-stage framework for CME coupled with PI and QI. The three stages of the activity, with additional education, were as follows:
Stage A—Assess Current Knowledge, Competence and Performance
In Stage A, learner “baseline” knowledge, competence and performance, were measured via required learner/patient entered quality data into four different tools in the online learning platform. These were:
- Clinical metrics based on National Quality Forum measures from de-identified patient charts (20 charts per learner) related to the study conditions;
- De-identified patient surveys (20 surveys per learner) assessing the experience and perspective of patients during the office visit;
- Learner self-assessment survey of their practice structure; and
- Structured case vignettes to assess learner knowledge/competence of evidence-based medicine around the study conditions.
Once learners had submitted baseline data, an auto-generated report was created with individual and aggregate peer data allowing learner review, benchmarking and reflection. This information provided the basis for self-identification of practice gaps at the clinical and systems levels and specific educational needs around these deficits.
Stage B—Engage in Evidence-Based Education Driven by the Results of Stage A
In this stage, the format of the course utilized both live and online components. Learners attended a live conference workshop that combined traditional didactic presentations with small group peer-to-peer problem solving. In this environment, every learner in the room was considered an expert on their own practice deficits, and each had something to bring to a discussion to help their colleagues who faced similar issues. Learners were trained on the basics of practice redesign and transformation, root cause analysis of barriers, and received education on evidence-based practice guidelines and their application into practice. They were then challenged to apply this information to the data generated in Stage A to identify an area that needed to change in their practice. Finally, learners formulated a personalized action plan based on the didactic education and peer-to-peer interactions. Learners were able to select from a list of prewritten interventions or create a new intervention specific to their identified practice deficit and upload the plan into the online module.
Additionally, during the workshop, learners were able to reinforce their knowledge/competence skills by completing up to two ABFM approved MC-FP Part II, Self-assessment Module (SAM) working group sessions on diabetes and hypertension. These group sessions presented learners with the opportunity to work through 60 case-based questions on aspects of care for the two disease states and to evaluate answers with evidence-based rationales and appropriate remediation.
Stage C—Re-Assessment of Practice Following an Implementation Period
Following an implementation period of at least 30 days, practices re-measured their performance using the same tools and questions as per Stage A. Automated immediate reporting from this stage of the activity allowed learners to access their own performance/competence change as well as the aggregate change of their peers. Access to these data sets provided both an ongoing motivation for the user to improve practice and also to utilize PerformanceNavigator as the first step in a continuous quality improvement (CQI) cycle.
The Online Educational Platform
The PerformanceNavigator platform provided a convenient online experience and environment to engage learners in the PI/QI process. The educational platform was designed by the AAFP and CECity, and maintained by CECity (CECity Inc., Homestead, Pennsylvania). Each of the three disease states was packaged as a separate optional module, with distinct measurement requirements, CME accreditation, and MC-FP credit.
The platform was arranged such that it took a learner sequentially through all of the required items in the activity, including the live workshop, and indicated when each step had been completed. Between Stages B and C (implementation of the action plan) the platform allowed learners to view their baseline data and action plan but not to move forward until at least the minimum required time of 30 days had passed. In order to facilitate peer-to-peer interaction before, during, and after the workshop, the PerformanceNavigator platform also gave learners the opportunity to participate in an online social learning forum, to share experiences or to address practice transformation related questions.
Preliminary outcomes data shown below were current as of June 2016. Included are data for the diabetes PerformanceNavigator module from 29 physicians who had full scope practices and continuous patient panels. Over 2,000 patients were evaluated during this study. All learner data was paired (i.e., Stage A and C data were compared for each individual).
Self-Assessment Survey Data
Data from learners demonstrated comparative improvements between Stages A and C across of a range of systems-based metrics. Improvement was evaluated by comparing the percentage of learners who had scored in the top two Likert Scale levels for both stages (Figure 1). Using this rationale, improvements were observed in team-based care, care coordination strategy, and integration of specific targets and goals into the patients’ management plan.
Patient Chart Data
Chart data (Figure 2) was evaluated by comparing the percentage of patients who had appropriately received a procedure between Stages A and C. Data from the 29 learners and 1,236 patient charts indicated improvement in the procedural performance measures for foot examinations (statistically significant), retinopathy screening, and nephropathy screening (statistically significant). There was also improvement in the patient health outcome measure of flu vaccinations given between the stages. However, so far, little difference has been observed regarding improvements in patient education.
Data from patient surveys (Figure 3) was evaluated by comparing the top two levels of Likert Scale questions or the “Yes” answers of Yes/No type questions between Stages A and C. Data from the 29 learners and 1,178 patient surveys indicated performance improvement in access to care measures (appointment making and waiting times), cultural competency (inclusion of personal and family beliefs), patient inclusion into their own management plan (asked about health preferences and goals), and coordination of care (assistance with making appointments with other health care providers).
Strengths of Our Approach
The greatest strength of the activity was its ability to merge assessment of clinical skill, patient perspective and resource utilization together to give learners a snapshot of their current environment and provide motivation for participation in directed education to bridge individualized gaps. To date, one of the biggest failures of the “traditional” PI/QI methodology has been the lack of creating this connection for learners. PerformanceNavigator facilitated this process using an innovative design by presenting individualized practice data to the learner through the online component and then through the workshop helped them understand their data and focus transformation efforts based on their findings. Although our approach involves a sophisticated online platform, the format can be scaled based on budget or educational goal — for example, by allowing learners to simply log their own data into a pre-built Excel template or by adjusting content in the live component to meet the needs of a specific type of healthcare professional.
As physician administrative work load is a concern, this activity provided a unique opportunity for learners to concentrate their efforts and complete several requirements simultaneously including MC-FP Parts II and IV, and CME — all centered on cardiometabolic conditions. Facilitating practice transformation while improving clinical expertise, particularly to chronic disease conditions, has direct implications for the future of medicine in the U.S., as it moves toward a value-based care payment structure. In addition, any course that allows physicians to consolidate certification and educational efforts will likely liberate time to focus on patient care while reducing stress and burnout.
Considerations for CE Providers
Firstly, an activity this large in scope and complexity will always cause confusion for some learners. Although we strived to provide clear documentation, for a few individuals, additional telephone coaching provided by AAFP staff was required. Secondly, it is likely that some learners will come to the live workshop without having completed the prerequisite data entry. This results in two sets of learners — at different stages in the process — which can be frustrating for both groups and faculty. We found that regular email reminders with active links embedded in the email text were helpful in prompting learners to complete data entry before the workshop. However, for the few who had not completed Stage A, CE professionals should consider the possible heterogeneity of the group, and plan accordingly so as to include them in the interactive group components of the course.
Next Steps/Changes in the Second Iteration
As part of the ongoing process to keep the material scientifically current and to improve the activity based on learner feedback, some minor changes have been made. These included adding enduring how-to guide videos for each stage, and altering the length of some of the small group sessions during the live workshop.
Higher level educational outcomes (performance and patient outcomes) generated from the PerformanceNavigator activity appear to demonstrate learner improvement across a range of measures. For many learners, clinical and patient perspective improvement was observed concomitantly with improved resource utilization. Increased use of team-based care, care coordination, planned visits, and integration of individualized targets and goals into patient management plans was also observed. Although preliminary, these data are consistent with an adoption of a more patient-centric approach into learner practices and relate back to the original goal of this activity. We believe that the educational format described in this article facilitates change, in part by intrinsically motivating learners to want to improve their practice. The long-term impact of practice transformation that highlights the patient as a stakeholder is likely to result in improved practice efficiency, and ultimately improvement in quality of care and patient level outcomes beyond this activity’s study conditions.
- CDC website disease statistics: Available at: http://www.cdc.gov/. Accessed 01/25/2017.
- AAFP, AAP, ACP, and AOA. 2007. “Joint Principles for Medical Education of Physicians as Preparation for Practice in the Patient-Centered Medical Home”. Available at http://www.aafp.org/dam/AAFP/documents/practice_management/pcmh/initiatives/PCMHJoint.pdf. Accessed 01/25/2017.
- Wang QC, Chawla R, Colombo CM, Snyder RL, Nigam S. Patient-centered medical home impact on health plan members with diabetes. J Public Health Manag Pract. 2014;20(5):E12-20.
- Pagán JA, Carlson EK. Assessing long-term health and cost outcomes of patient-centered medical homes serving adults with poor diabetes control. J Prim Care Community Health. 2013;4(4):281-285.
- Van Berckelaer A, DiRocco D, Ferguson M, Gray P, Marcus N, Day S. Building a patient-centered medical home: obtaining the patient's voice. J Am Board Fam Med. 2012;25(2):192-198.