Development of Palliative Care Quality Measures for Outpatients in a Clinic-Based Setting

Palliative care has expanded rapidly in recent years, and a consensus has been growing within the provider community regarding the need for measuring the quality of palliative care. Yet little is known about the quality of palliative care delivered, particularly among patients who receive their palliative care early in their disease trajectory in the outpatient setting. The patterns of palliative care received in outpatient clinics differ substantially from palliative care received in other settings. Outpatient palliative care often supplements a primary treating service such as oncology. Patients may have several visits with different members of the palliative care team, or they may only have a single visit. This variability in the patient experience of palliative care raises important measurement challenges. The Centers for Medicare & Medicaid Services (CMS) provided funding to the American Academy of Hospice and Palliative Medicine (AAHPM) as part of the Medicare Access and CHIP Reauthorization Act of 2015 (MACRA) to develop patient-reported experience performance measures in the areas of pain and symptom management and communication for adult patients with serious illness, including those receiving palliative care. Between fall 2019 and summer 2021, AAHPM and their subcontractors—the RAND Corporation and the National Coalition for Hospice and Palliative Care—will develop, test, and implement two patient-reported experience measures for patients participating in outpatient, clinic-based, palliative care.

The two measures will need to be both broadly applicable to patients and families, and useful to clinicians and health systems in measuring and improving the quality of care that patients with serious illness receive. To meet these goals, this three-year project (October 2018–September 2021) includes a series of information-gathering, stakeholder engagement, and testing activities. The project is using an innovative approach to ensure broad stakeholder input; the Technical Expert Clinical User Patient Panel model elicits the perspectives of patients, caregivers, and family members, in addition to clinicians and researchers. Development of measures has been and will be informed by focus groups and interviews with providers, patients, and caregivers. Two stages of field testing will provide data on the feasibility and performance of the measures. After development and testing, the two measures will be submitted to the National Quality Forum for endorsement, with the ultimate goal of inclusion of the measures in CMS's Quality Payment Program, including the Merit-Based Incentive Payment System (MIPS) and Alternative Payment Models (APMs).

This study describes information-gathering activities completed as of September 2019. The purpose of this study is to summarize important background information that demonstrates the importance of and need for development of quality measures based on patient-reported experience. Components of this report include a review of the literature and measure repositories to identify measures already in use, potential competition among existing instruments, and work needed to facilitate harmonization among measures; a literature review to identify the most common symptoms experienced by patients with serious illness; a review of clinical practice guidelines to understand domains of clinical importance; a review of existing regulations to understand the parameters within which the proposed measures would be implemented; and patient or caregiver and provider interviews and focus groups to understand end-user perspectives, including what is most important to patients and what implementation of the proposed measures might look like in practice.

Our findings confirm that there is a need for quality measures focused on clinic-based palliative care, given the growth in demand for and provision of these services. Stakeholders previously convened consensus panels that included experts on serious illness care, researchers, payers, and patient advocates. These consensus panels generated measurement priorities that included the importance of assessing the quality of symptom management and communication, the areas of focus for measure development under this project. Core goals of high-quality palliative care are good communication, which can be measured by items that assess the extent to which a patient and/or caregiver felt heard and understood, and symptom management, which can be measured by items that assess the extent to which a patient's clinical needs have been met. In this work, we note the difference between measuring symptom management by level of symptoms reported (e.g., pain rated on a 1–10 scale), where higher levels of symptoms might be considered poor symptom management, and by asking patients the extent to which symptoms were managed adequately for their goals (e.g., getting as much help for pain as they wanted). Focusing on these two core areas—feeling heard and understood, and level of unmet needs for symptom management—will help assess the quality of clinical care delivery management of psychosocial needs, while accommodating the subjective experience of the patient.

To provide a broader context for the findings of our literature review and stakeholder engagement work, we also summarize pertinent clinical guidelines and regulations that influence palliative care quality measurement. Clinical guidelines for palliative care have been defined by the National Consensus Project's Clinical Practice Guidelines for Quality Palliative Care (NCP). The fourth revision of the NCP Guidelines, published in October 2018, set “expectations of excellence” across eight domains of care, ranging from structure and processes, cultural aspects, and legal aspects of care. It also includes a systematic review of the evidence base for palliative care. The levels of evidence—that is, the quality, validity, and applicability of the underlying studies—that inform care practices vary substantially, with low-quality evidence for several physical aspects of care, such as pharmacological interventions for common symptoms like pain, or for incorporating familial or cultural preferences into care delivery. Moderate quality evidence was found that a palliative care team can reduce symptom burden, and that care planning discussions can lead to preference-concordant care, yet the NCP notes little attention to patient goals or inclusion of the patient perspective in existing measures. Against this backdrop, we believe in developing quality measures that are based on patient-reported experience that is an important next step in generating evidence on and ensuring that providers prioritize the perspective of patients receiving palliative care.

In recent years, regulations and policy priorities have focused on ways to ensure high-quality care by linking payment incentives to performance on quality measures. The Hospice Quality Report Program, created by the Affordable Care Act in 2010, established quality reporting requirements for hospices, which currently include the Hospice CAHPS (Consumer Assessment of Healthcare Providers and Systems) survey, a survey of bereaved family members of patients who die in hospice care. More recently, in 2015, the passage of MACRA changed payment calculation and quality reporting in the Medicare program and set the stage for a broader focus on high-value care that is enabled by more robust quality measurement and reporting. CMS expanded the types of providers eligible to participate in MIPS in 2019, all of whom will be able to use the quality measures developed and implemented through this project when they provide palliative care to patients. The measures that are being developed by this project are aligned with current CMS quality reporting priorities in that they will provide more options for providers to measure and report on the quality of care they are providing in order to meet the requirements of MIPS.

We identified over 300 existing data elements in 31 patient survey or interview instruments that assess communication and unmet needs with regard to symptom management in various contexts. These included the CAHPS suite of surveys, instruments that assess the experiences of bereaved family members, and instruments that assess patient experience and quality of end-of-life care, including within the Veterans Administration. The data elements we found will inform the selection and refinement of the quality measures we are developing for this project. We did not find many data elements targeted to the palliative care population in widespread use, and those that were in widespread use were not tailored to a seriously ill population. However, we did identify several themes across data elements for both areas of focus. For communication, data elements tended to assess how well providers listen to patients and their concerns, explain clinical information in an easy-to-understand manner, demonstrate caring and respect for the patient, and explain the dying process. Most data elements on symptom management and/or unmet need for symptom management tended to assess satisfaction with providers' management of overall symptoms (though some asked about specific symptoms, such as pain and trouble breathing), as well as the speed with which that care was delivered.

We also conducted a search of symptoms likely to be addressed by measures being developed in this project. Some studies on the prevalence of symptoms among end-of-life populations examined symptoms within specific care settings; others looked at symptoms within specific time frames such as the last year or weeks of life. Both types of studies found that pain, fatigue or sleeping problems, and eating or digestive symptoms are likely to be common problems among patients who receive palliative care. Other problems such as respiratory issues, low mood, and anxiety were reported by fewer studies or at lower rates but may also be important to assess.

Considering this information, we assessed gaps in existing quality measures to understand how the measures under development for this project will fit within the wider quality measurement landscape. Our scan identified 13 measures related to communication and symptom management. A central gap is that existing measures were not designed for use in, and have not been tested among, the population that this project focuses on: patients with serious illness receiving outpatient palliative care services. We concluded that current measures of communication and symptom management would need to be tailored for our focus population.

Findings from our focus groups with providers and interviews with patients and caregivers provided feedback on possible quality measure concepts and ways to assess key concepts through survey questions. Participants elaborated on the types of unmet symptom management experienced by patients with serious illness who receive palliative care from an outpatient clinic and provided feedback on ways to talk about symptoms and unmet needs that will be explored further in alpha and beta testing. Providers, patients, and families also discussed considerations for implementation of patient-reported outcome measures around unmet symptom needs and communication that we will take into consideration during measure development.

Results of the information-gathering activities to date have demonstrated the gap in measurement around the concepts and population of interest, those in outpatient palliative care. Information on data elements and quality measures from the research literature, and input on data elements and data collection approaches from providers and patients, will be used to inform data element development, alpha and beta testing, and the final specifications of the quality measures.