This article is more than 5 years old. Information may no longer be current.
Is hospitalization within the final 30 days of life appropriate for patients with hematologic malignancies?
Click Here to Manage Email Alerts
Click here to read the Cover Story, “Hem/oncs toe ‘line in the sand’ between quest for cure, overly aggressive end-of-life care’.”
Yes.
As both a practicing blood cancer specialist and a board-certified palliative medicine physician, I know how valuable palliative care services are in ensuring high-quality care at the end of life. Indeed, palliative care is not only about end-of-life care. My colleague, Areej R. El-Jawahri, MD, demonstrated its enormous benefit in a randomized controlled trial of concurrent palliative care among patients receiving curative-intent stem cell transplantation.
I wish every patient with blood cancer had access to specialist palliative care from the time of diagnosis. However, patients with blood cancers often have different needs than those with solid organ tumors, and these needs sometimes mean that hospitalization is not only appropriate but necessary, even at the end of life. One reason is because patients’ unique needs sometimes cannot be met via existing models of home-based palliative care — namely hospice — which typically require patients to forego any disease-directed therapies or costly supportive treatments like transfusions, growth factors and antimicrobials.
Consider a patient with relapsed, refractory acute leukemia. Such a patient may still derive significant palliative benefits from disease-directed therapies — such as hydroxyurea — or supportive treatments, like blood and platelet transfusions, and may require transfusion support two to three times per week or more to feel reasonably well and prevent serious bleeding. Most hospice agencies cannot provide such care, leaving this patient to fend for themselves at home, while their clinical care team tries to piece together a plan for extra support from caregivers and home health services. Sometimes it is not possible, and a patient like this may, therefore, end up in the hospital due to significant care needs that cannot be met at home.
If this is the kind of care a patient wants to receive, and supportive measures like transfusions and antibiotics are helping them feel better and live a bit longer to achieve important goals, who are we to object?
This brings me to a more fundamental objection about blanket end-of-life quality measures — namely, such measures fail to capture the patient’s actual preferences. In fact, some patients with blood cancers prefer to spend their last days in the hospital. Those who have undergone intensive therapies like stem cell transplantation or acute leukemia inductions already have spent a lot of time on our inpatient wards and may feel like the inpatient staff are a part of their family. They also may feel it is not possible to manage all their needs at home, and there can be enormous fear and anxiety about going home in such a situation. Rather than bemoaning such a hospital stay, we should measure patients’ goals and preferences, and set our sights on “goal-concordant care” as the important outcome of interest, rather than a blanket prohibition against what might actually be a very helpful hospitalization that helps the patient meet their goals near the end of life. We should set our sights on ensuring that patients get the type of care they want, when and where they want it.
Unfortunately, there are no validated or widely available methods for reliably measuring goal concordance. Rather than debate the merits of hospitalization near end of life, we should address this gap first. Let us focus our energies on devising and standardizing measures of goal concordance, and elevate the state of the science of quality measurement in oncology.
References:
El-Jawahri A, et al. JAMA. 2016;doi:10.1001/jama.2016.16786.
Howell DA, et al. BMJ Support Palliat Care. 2015;doi:10.1136/bmjspcare-2014-000793.
Thomas Leblanc, MD, MA, FAAHPM, is assistant professor of medicine at Duke Cancer Institute. He can be reached at thomas.leblanc@duke.edu. Disclosure: Leblanc reports no relevant financial disclosures.
We can only answer the question once we talk openly about prognosis and better define patient goals.
To answer this question in a fair manner, one needs to discuss two aspects of clinical care with which we continue to struggle: prognostic accuracy and person-centered care. The question presupposes we know when the last 30 days of someone’s life will be, and that clinicians have a good understanding of what goals drive the patient and their family. We are poor in our ability to prognosticate and confident in understanding goals of care for patients. I propose we need to have more confidence and transparency in prognostication, and discover more humility in thinking we know what matters to patients. Then we can determine what care is most appropriate.
Quality measures and cost analysis of end-of-life care start at the date of death and march confidently backward to inspect the clinical decisions and care delivery, all while presuming the clinicians and the patient knew they were nearing the end of life. People with hematologic malignancies can shift from stable to critical without much warning, making accurate prognostication very challenging. Data from ICUs show that the mortality of people with hematologic malignancies and stem cell transplants actually has been improving.
If patients decline rapidly, and a small minority may benefit from aggressive measures — such as ICU and mechanical ventilation — clinicians will be hesitant to discuss prognosis openly. Yet, in my practice as a palliative care physician who works closely upstream with our hematologists and stem cell transplant physicians, discussing prognostic estimates with my colleagues leads to better care for patients. More open discussion addresses the concern of hematologists about inaccurate prognostic disclosure from palliative care teams. When we have a wide mismatch in our estimates, then we review medical factors and treatment plans, which usually leads to a revision of prognosis. Once we agree on a more unified prognosis, patients and surrogate decision-makers can be better informed, instead of hearing mixed messages. With a prognostic map openly shared, physicians and patients can readdress goals as new information shifts the prognosis for better or worse, which can help address the inherent uncertainty in predicting the future. Transparent prognostic discussion and documentation also will help future research improve our accuracy.
Once we have a more credible prognosis and patients are open to hearing it, they can more readily adapt their goals and decisions. Without hearing our shared prognosis, they are left to read the tea leaves of our vague words and concerned faces, which is likely less accurate than anything we may communicate directly. Many people with hematologic malignancies are used to being aggressive in their goals and have been exposed to a high intensity of medical treatments. This often leads to presuming goals are aggressive, even in the face of someone being critically ill. The fighting language that can help get through tough times often leads to fear of “giving up” or “quitting” once “very sick” becomes “actively dying.” Exploring these ideas takes time and active listening, which can be hard to do as a single clinician in a busy clinic or an active inpatient service. The collaboration of hematology and palliative care allows patients and their families give voice to their worries and concerns, which can help them shift goals of care as an illness progresses.
References:
LeBlanc TW, et al. J Oncol Pract. 2015;doi:10.1200/JOP.2014.001859.
Oliver C, et al. Blood. 2014;124:6017.
Christian T. Sinclair, MD, FAAHPM, is assistant professor of medicine at University of Kansas Health System. He can be reached at csinclair@kumc.edu. Disclosure: Sinclair reports no relevant financial disclosures.