Brief
Goals of care conversations are often initiated and documented by clinicians. Prompts to patients can encourage documentation of patient goals and values. A patient-initiated survey on goals in the electronic health record is feasible and can guide goal-concordant care.
The Problem
Only 25% of Duke Health patients near the end of life have had documented conversations with healthcare providers about their goals of care. Consequently, plans of care, including plans for end-of-life care, are being developed and implemented without recorded input from patients and their families. A large body of evidence has shown that goals of care conversations are essential in promoting care concordant with patients’ preferences. Communication improves patient and family satisfaction, reduces healthcare utilization, and lowers care costs. In response, Duke Health began a broad mission to promote goal-concordant care.
Our Solution
Duke Palliative Care, the Duke Institute for Health Innovation (DIHI), and Duke Population Health Sciences formed a team to oversee the implementation and measure the impact of direct-to-patient goal concordant care support. Whereas previous work has focused on clinician-initiated conversations, our project’s overarching aim was to implement, optimize, and evaluate the Patient Initiated Note about Goals (PING), encouraging patients to share their goals with their care team via MyChart proactively. The PING was developed in late 2021-early 2022 with clinicians and patient stakeholder groups, including Duke Patient and Family Advisory Council, the Duke Patient Education Office, and underrepresented minorities in the community. It includes a brief three-minute survey on patient goals using multiple-choice and free-text options and is written at a fifth-grade reading level. Categories focused on important goals for patients, such as: “not being a burden on others” and “living as long as possible,” and answers included “not very important,” “somewhat important,” and “very important.” The project began with a go-live: Duke Palliative Care, Duke Population Health Sciences, and Duke Health Technology Solutions implemented the PING in the spring of 2022. A “silent BPA” in Maestro Care began identifying candidates to receive the PING based on their comorbidities and recent clinical events (e.g., a recent hospitalization). This instantiates a MyChart message to the patient, encouraging them to complete the PING. Our project team began analyzing response data. After a review of preliminary data, however, we observed that the direct-to-patient messaging was not having the scale of impact we had hoped, with minimal responses from direct-via-MyChart outreach. So, starting in January 2023, we partnered with the Duke Primary Care (DPC) Population Health Nurse (PHN) Team, led by Beth Soule and Sarah Tucker, to incorporate the PING into their workflow for assessing patients during their annual wellness check visits. Figure 1 shows the workflow used by the PHN Team.
Impact
We analyzed results from the PING responses since the 2022 go-live (04/01/2022 – 07/15/2023), wherein 678 unique patients responded to the PING via direct-to-patient MyChart message or the DPC PHN workflow. From the period directly after going live (i.e., the direct-to-patient only route via MyChart, 04/01/2022-01/30/23), 279 patients (27.9 per month) responded to the PING. After the partnership with the DPC PHN nurses began, 400 patients responded to the PING (71.4 per month, 02/01/2023-07/18/2023), more than doubling the direct-to-patient-only route. Our 678 PING respondents were, on average, 61.5 (SD 17.2) years old, predominantly female (67.6%), White (70.6%), and Non-Hispanic (92.8%). It was “very important” for respondents to “not be a burden on others” (85.5%) and “take care of myself than depending on others” (84.4%), and “not important at all” to “avoid travel to and from clinic appointments” (40.9%).
Next Steps (Dec 2023)
Our next steps are to improve our reach and understanding of Duke patients’ goals of care so that we can best provide concordant care. We will prioritize two objectives: (1) scale our inclusion of the PING across additional wellness visits and other appropriate ambulatory/outpatient settings and (2) refine our understanding of direct-to-patient outreach via MyChart and other platforms. Additionally, we plan to measure the impact of a completed PING on subsequent clinical outcome metrics: hospitalization and Emergency Department (ED) visit rates, palliative care utilization, and Intensive Care Unit (ICU) utilization. We hope to show the positive impact of goal-concordant care and the alignment of healthcare utilization with our patients’ goals.
