Abstract
W
Community-based palliative care has many flavors and encompasses many care settings. The multidimensionality of community-based palliative care contributes to both its challenge and opportunity. Outpatient palliative care practices in an office setting may be standalone or may be embedded in specialty or primary care clinics. They may have their own interprofessional teams or may draw on certain disciplines from the clinics with which they are affiliated. Nursing homes, likewise, may have their own palliative care consult team or service or more frequently draw on palliative care services from hospital-based palliative care teams or from hospices. Palliative care practices in the home have the most variety of configurations—free-standing consultative programs, outgrowths of longitudinal home-based primary care programs, offshoots of hospice or home-health services, or part of special services offered by payers. The wide spectrum of community-based palliative care programs is essential to meeting patient need across the continuum of care, both where health care services are widely available and where they are not.
The challenge with so many palliative care arrangements in the community is that no clear path is followed by any program and the minimum requirements to appropriately define a program as “palliative care” are not widely accepted. In this issue, Ceronsky and colleagues report on a field test of quality measures they developed or adapted for rural, community-based palliative care programs. The rural palliative care programs with which they worked highlight the wide variability in community-based palliative care teams, from a service based in the hospital to a service based in home care. Many programs call themselves palliative care programs even when a core team is not present or available and the members of the team do not have evidence of palliative care expertise through certification or rigorous training. In the programs described by Ceronsky et al., the only team member referenced was a registered nurse or a social worker. The presence of chaplains, pharmacists, and physicians was not mentioned—nor was the palliative care training or certification.
With the palliative care practice path so poorly delineated in the community, our opportunity is to think through which practice models are the most effective and what competencies are critical in each and across settings. Practice models in the long-term care or office setting may well be able to leverage clinical and/or specialist palliative care resources already present in that setting. In the home, the family caregiver often takes on a more central role in daily care. Certain practice standards particularly relevant in one setting may be less important in another.
With the January 26, 2015 watershed statement by the Department of Health and Human Services that by 2016 85% of Medicare fee-for-service reimbursement payments will be tied to quality and value, clarifications around what constitutes “quality and value” in community-based palliative care is now a particularly urgent issue. Medicare, to date, has utilized measures vetted through the National Quality Forum (NQF) and available through the Physician Quality Reporting System (PQRS) library as a starting point for defining quality for patient care. More recently, quality improvement efforts through Qualified Clinical Data Registries and through practice certifications (such as that delineated through the NCQA Healthcare Effectiveness Data and Information Set [HEDIS] measures) serve as supplementary approaches to measuring value that is recognized by the Center for Medicare and Medicaid Services. 3
Unfortunately most National Quality Forum endorsed measures and HEDIS measures are not oriented towards persons with serious illness, who often have limited life expectancies and high illness and treatment burden. Many 2015 HEDIS measures are prevention based (e.g., colorectal cancer screening, treatment of high blood pressure). Few to none focus on assessment of symptoms, communication about and prioritization of treatments, or management of caregiver distress.
As important as measuring quality is, it is a difficult task and complicated by many issues inherent to the field of palliative care—the subjective and transient nature of patient preferences and priorities in the setting of progressive illness, the variability of palliative services across the continuum of care, and the complexities of interdisciplinary teams. Many quality measures focus on processes of care, with fewer assessing outcomes of care. As a field, we have an idea about what constitutes quality of care in the hospital or hospice setting; we have much larger gaps in our knowledge of what constitutes quality of care in community-based palliative care. The field has little data by which to benchmark either processes or outcomes, especially in community-based settings.
Additionally, new and developing community-based palliative care programs may have difficulty collecting sufficient, reliable data to be able to assess quality across multiple domains of interest. Several initiatives have contributed positively to our knowledge and understanding of the potential of capturing quality in palliative care practices: the American Academy of Hospice and Palliative Medicine Measuring What Matters project, 4 the National Consensus Project Guidelines, 5 the Global Palliative Care Quality Alliance, 6 the Home-based Primary Care and Palliative Care Network, 7 and the Palliative Care Quality Network. 8 These quality of care initiatives have focused on an array of palliative care settings but have not systematically demonstrated appropriateness of specific measures in each setting, and few have put the measures used through rigorous testing for importance, feasibility, usability and impact.
The study by Ceronsky et al. demonstrated both the promise and challenges of measurement in community-based palliative care. The researchers worked to develop measures to assess clinical outcomes, patient and family experiences, and health care utilization in palliative care programs in rural Minnesota. Given the absence of standard assessments in the rural setting, measures were developed using the existing literature and expert opinion. The project demonstrated that the operational measures were feasible to measure by the five programs evaluated, but the response rate by patients and families was, not surprisingly, low.
Whether the findings are generalizable are also unclear given the small sample size and the likely diversity of rural programs nationally. The measures developed and assessed by Ceronsky et al. included a few key process measures, but other key indicators of quality suggested by the National Consensus Project (such as the use of an interdisciplinary team and advance care planning) were not assessed. These measures were developed in other settings; the working definition of quality clinical care and the measurement of quantifiable clinical outcomes remains unexplored in this care setting. The ability of the measures to reliably assist with program comparison and improvement was suggested by the data presented, but these findings must be validated. This work is a good start, but other gaps remain as well, including measures to assess psychosocial outcomes and those for assessing community collaboration.
Our opportunity as a palliative care community is to dig deeply into what measures best characterize quality for community-based palliative care programs. We need to hear from patients and caregivers about what high-quality palliative care looks like in the office, in the nursing home, and at home. We need to develop measures that fill current measure gaps identified by patients, caregivers, and other stakeholders. We need to use existing quality initiatives to develop standards for community-based palliative care in the areas where measures are robust and tested and develop crosswalks across currently accepted quality domains. In those areas where we have measurement tools, we should develop shared data elements that allow for risk adjustment and benchmarking across practices and settings. Finally, we need to aggressively develop measures where there are none and do the hard work of testing them not in the hospital, but in the community, where the majority of our patients reside.