A presentation at the National Hospice and Palliative Care Organization's conference in Washington, DC, in April explored the relationship between hospice and palliative care providers and the growing number of hospital quality reporting initiatives that incorporate risk-adjusted mortality rates as a marker of the hospital's quality. J. Brian Cassel of Virginia Commonwealth University (VCU) and Amber Jones of the Center to Advance Palliative Care explained that these initiatives use complex methodologies, including varied data sources, different included diagnoses, different methods of adjusting for severity of illness and balancing mortality with other metrics of quality -- and different criteria for whether to include or exclude patients receiving hospice or palliative care.
Examples discussed by Cassel and Jones include the federal government's Hospital Compare, which publicly reports data on patient satisfaction, hospital outcomes and mortality; HealthGrades, a company that ranks hospitals and other providers; and U.S. News & World Report's "Best Hospitals" ratings. These quality initiatives suggest that deaths from any cause, either during the hospitalization in question or within 30 days after admission, for more-or-less equally sick patients are on average a fair reflection of the quality of care provided by the hospital. They have an underlying premise that death is a medical failure, Cassell said, while the underlying premise of hospice and palliative care is that death generally is a normal event in the human life cycle.
But what if the patient is known to be terminally ill, is enrolled in hospice care, or was referred to a palliative care consultation service to explore treatment options, goals of care and a possible transition to comfort-oriented care? Can some deaths be labeled unavoidable, reasonably predictable or even appropriate outcomes, given the reality of the illness?
And should such patients be counted out of the hospital's mortality ratings? What if it is a hospice patient receiving terminal care who comes to the hospital for in-patient hospice care to manage an acute symptom crisis? What if the hospital refers patients to hospice care shortly before their deaths -- in order to enhance its mortality ratings? What if the palliative care service's involvement allows a seriously ill patient to choose the comfort care option? Hospital rating services address questions like these in somewhat different ways.
Hospitals are naturally concerned about mortality rates because they want to be seen as quality providers. Some hospice and palliative care advocates want to use mortality rates as a conversation starter with hospital administrators. Cassel said that palliative care teams need to know which quality measures their hospital pays attention to, and what these look for. Palliative care advocates can also learn more about hospital billing codes and opportunities to flag palliative care involvement and more fully report symptoms and co-morbidities, given their potential effects on the rates.
Cassel said he started studying the issue because of concerns that an acute palliative care unit at VCU might be hurting the hospital's mortality scores, although this proved not to be the case. (See the Journal of Palliative Medicine, April 2010, 13: 371-74.) To take another example, Bon Secours, a 14-hospital health system based in Marriottsville, MD, in 2006 launched a systemwide zero potentially preventable deaths quality initiative inspired by the Institute for Healthcare Improvement. Its palliative care leaders have been involved in helping to define where palliative care fits in this initiative and how to recognize and prevent suffering when death is appropriate and anticipated. (See the Journal of Palliative Medicine, April 2010, 13: 367-70.)
Eugene Kroch of Premier, Inc., and co-authors discuss this issue in a recent American Journal of Medical Quality article (published online January 1, 2010), evaluating the benefits and caveats of trying to incorporate care-limiting medical measures such as do-not rescusitate orders and palliative care consultations into a model of hospital mortality risk. Palliative care, they say, "is unequivocally valuable in accounting for patient risk." However, Kroch told me that the national focus on hospital quality and adjusted mortality rates may not reflect what hospice and palliative care professionals would recognize as a natural and appropriate end-of-life care pathway. The numbers can be misleading, and other observers have also raised methodological concerns about hospital mortality rates. "Can we at least understand the expectation of the patient at the time of hospitalization?" Kroch posed.
Brian Cassel gets the final word. Regardless of what scores come into play and where hospice and palliative care fit into hospital mortality rates, the number one responsibility for these providers continues to be providing the best possible care to patients and families.