Thursday, August 19, 2010

Palliative Care Prolongs Life

Thanks to Wikimedia commons for picture,
A puffer fish kissing the glass.
No intentional relationship to the post,
Just loved the picture.
A study published today in the New England Journal of Medicine by Jennifer Temel and colleagues at the Massachusetts General Hospital (MGH) showed that patients with advanced lung cancer treated with early outpatient palliative care lived longer than patients randomized to usual care. 

What?

A randomized controlled trial of outpatient palliative care showed a survival benefit.

Palliative care prolongs life?  Really?  REALLY???

Here is my take on the study, and I'm interested in yours.  151 patients with metastatic non-small cell lung cancer were randomized within 8 weeks of diagnosis to either monthly visits to an outpatient palliative care clinic plus usual oncology care or usual oncology care alone.  Patients in the usual oncology care group could attend the outpatient palliative care clinic if requested, but only 14% did.  The characteristics of the two groups looked similar before randomization for all of the things you would care about, like functional status, presence of brain mets, and baseline measures of quality of life, symptoms, and mood.  The outpatient palliative care clinic was staffed by great docs and NPs, like Vicki Jackson, MD, and Connie Dahlin, NP. They used the National Consensus Project for Palliative Care guidelines.  The main focus of the study was improvement in quality of life, symptoms, mood, and health services use.  After 3 months, here's what they found:
  • Higher quality of life ratings in the palliative care group (score of 98 on Functional Assessment of Cancer Therapy-Lung scale) versus usual care (score of 92)
  • Better symptom control in the palliative care group (score of 21 on the Lung Cancer Subscale) versus usual care (score of 19).  A composite score showed similar improvements.
  • Fewer subjects meeting criteria for major depressive disorder in the palliative care group (17% vs 4%).
By the end of the study follow up period (about 30 months)
  • 70% of subjects had died
  • 53% of subjects in the palliative care arm had documented resuscitation preferences compared to 28% in the usual care arm.
  • Fewer hospitalizations and emergency department visits among the palliative care group
  • Trend toward greater use of hospice in the palliative care group (69% vs 60%)
  • Median survival in the palliative care group was 11.6 months from enrollment vs. 8.9 months in the usual care group
So how can we account for these findings?  What mysterious things are they doing in that palliative care clinic at MGH?  They must be so much better than anyone else, at Man's Best Hospital, how could we possibly achieve outcomes like this?

If you read the fine print, I don't think it's rocket science.  I worked in that clinic during fellowship, and I don't think it's all that different from what outpatient palliative care providers are doing in other parts of the country.

Here's what I think happened:
  • The guidelines for the outpatient palliative care group state they should attend to pain, cough, dyspnea, and depression. Those outcomes they assessed using the scores above with the fancy names...guess what they measure?  Pain, cough, dyspnea, and depression.
  • The guidelines say the palliative care providers should inquire about prognostic understanding and clarify treatment goals.  The study investigators wisely assessed documented rates of advance care planning.
  • Why did they live longer?  The study authors would like us to think it's all of the great things palliative care does to improve patients symptoms and quality of life.  Sean Morrison agrees, "When you control pain and other symptoms, people not only feel better, they live longer."  Probably a good part of it.  The other part may be avoiding hospitalization, and to an even greater extent chemotherapy at the end of life (33% of subjects in the palliative care group received chemotherpay within 30 days of death, compared with 42% in the usual care group). 
To what extent these survival differences are due to avoidance of "aggressive" care like chemotherapy and hospitalization, and to what extent differences are due to improved symptom care and mood remain to be studied.  Why is this a key point?  Because outpatient palliative care providers are trying to make a name for themselves as providers of outstanding symptom management for patients with advanced illness.  If the survival benefit was due to avoidance of chemo, maybe the palliative care team isn't so necessary.  On the other hand, you could argue it was the quality of the communication and clarification of goals in the palliative care group that led to less chemotherapy.  In any case, the prolonged survival findings add convincing evidence of a causal relationship hinted at in this retrospective study of hospice use, and this prospective trial of a nursing-led telephone outpatient palliative care intervention.

What can you tell your patients?
  • Early outpatient palliative care in addition to usual oncology care for metastatic non-small cell lung cancer improves physical symptoms, depression, sense of well-being, and not only doesn't shorten life, it may prolong life.
What do you tell your administration when you want to fund an outpatient palliative care clinic?
  • Early outpatient palliative care leads to reduced use of high intensity high cost care, such as chemotherapy, emergency department visits, and hospitalization at the end-of-life. 
For further thoughts on this topic, please see the write up on Pallimed by Lyle Fettig and the editorial by Amy Kelley and Diane Meier in NEJM.  This study did not address costs directly, a topic we discussed on a previous GeriPal post, called "What if palliative care doesn't save money?"

What are your thoughts?  Please add a comment.

by: Alex Smith

11 comments:

Christian Sinclair, MD said...

Pufferfish analogy - we shouldn't let our heads get too big about this?

I like you r analysis of why this was an effective study. We need to look at the parts of palliative care and the whole of palliative care. I am cautious about looking at one part alone though because it may not be reproducible just to look at avoidance of chemo as the main reason for the survival improvement unless you look at the avoidance of chemo in the patient who it was likely to hasten death in. And you only can know that with good goals of care discussions and fair assessments of likely prognosis. (does that make sense?)

Interesting title for your post Alex. Good one for SEO optimization! Hopefully that pulls up high in searches on Google if people are looking for Palliative Care. I did not think about that when Lyle wrote up his post.

Alex Smith said...

Thanks Christian. Certainly the survial benefit was due to a combination of factors, including less depression, less pain, less hospitalization, and less chemotherapy. The question is, was one of these a dominant factor? Won't be clear until further research is conducted.

Glad you like the title. Other titles I considered:
1. "Without palliative care, patients with lung cancer die sooner."
2. "Lack of palliative care leads to early death."

These are tongue in cheek, but there is a point to be made here. For those patients whose goal is life prolongation, NOT having palliative care is not in their best interests.

How generalizable is this finding beyond stage 3B and IV non-small cell lung cancer? Can't wait to hear the results of those studies, my money is on outpatient palliative care prolonging life across cancers and terminal conditions.

I would hazard a guess that those who would have the greatest survival benefit are geriatric patients with multiple co-morbid illnesses and no dominant terminal condition. Nothing like tailoring of medication, attention to physical, psychological, social, and spiritual concerns for frail elders. I think that's why so many patients with "general debility" and "failure to thrive" are discharged from hospice alive, only to fall apart again after discharge.

Ori Tzvieli said...

"On the other hand, you could argue it was the quality of the communication and clarification of goals in the palliative care group that led to less chemotherapy."

Isn't it sad that modern oncology, with it's twisted profit motive (more chemo = more profits) needs another specialist to come and bring a little common sense into the picture? When chemo takes the place of the difficult discussions about death patients clearly lose.

Next agenda item for palliative care docs to tell patients: "4 PET scans a year never cured anybody"

Dan Matlock said...

This trial really makes you wonder what prior researchers meant when they studied a new chemotherapeutic regimen against "best supportive care."

What, in fact, was the best supportive care used in those trials? This trial provides a clear, evidence based definition of "best supportive care." All future chemo trials have a new standard.

Gail Austin Cooney MD said...

I disagree with your conclusions, Alex. To me, it only makes sense that good symptom control and good communication lead to better outcomes. In 2008, I was diagnosed with ovarian cancer and had palliative care along with my intraperitoneal chemotherapy. I would not have tolerated the toxic treatments without the focus on symptom management and psychosocial support.

Alex Smith said...

Good point, Gail. I have no doubt that excellent symptom and psychosocial managment helped some patients in the palliative care group endure some of the initial chemotherapy. Still, I would speculate that the greater effect of the palliative care service was to reduce toxic chemotherapy for patients who were too more likely to be harmed than benefit.

All of this is purely speculative, of course. We have no evidence from this trial as to the source of the mortality benefit in the palliative care arm.

The biggest point of this trail, that I regret was likely lost in my enthusiasm for the survival story, is that the primary outcomes related to symptom control and well-being were improved with pallaitive care. While those findings are not particularly suprising or new, I actually think they're potentially more important than the survival story.

I say potentially because I'm not exactly sure how much difference in quality of life there was in this study. It isn't easy to grasp the quality of life outcomes as - sadly - the endpoints in this story were scales with names and measures that just aren't familiar to most people. When research gets to the point where the primary endpoint is the "TOI," a composite of the "FACT-L" and the "LCS"...the face validity and accessability of the results get a little lost.

What does this 2.3 point increase in TOI in the pallaitive care group compared to a 2.3 point decrease in the usual care group over 12 weeks actually mean? In simple language? This is a statistically significant difference, but is it clinically meaningful? What ever happened to rating quality of life with "excellent" "very good" "good" "fair" or "poor". That's a scale most people can understand.

ken covinsky said...

With all the excitement over this study, we are have neglected to note another important study in this week's NEJM:

A very important study showed that ipilmumab, a novel new type of chemotherapy, improved median survival in patients with metastatic melanoma. This study will offer some hope of life prolonging therapy in patients with this awful disease. As a result of this study, ipilmumab will probably become standard therapy. And because of this study, insurance companies will be expected to pay for this agent, at a cost of tens of millions of dollars. As they should.

But wait a second. The improvement in survival with ipilmumab for metastatic melanoma (3.7 months) is similar to the improvement in survival with palliative care for metastatic lung cancer (2.7 months). So, isn't it logical that palliative care be viewed as standard of care for patients with metastatic lung cancer?

It is interesting that the life prolongation benefit of palliative care for metastatic cancer seems to be quite similar to the benefit achieved with benefical chemotherapy regimens for advanced metastatic malignancies.

But there is a key difference. Palliative care also seemed to decrease the time spent in the hospital, markedly reduced rates of depression, and improved well-being. It is not common for chemotherapy studies to show these benefits.

Certainly, there is a need to replicate this study in other settings. And since this intervention is probably very dependent on the skill of the palliative medicine providers, we need to better understand how the process of palliative care delivery improves outcomes.

But, if this intervention were a drug, there is no question that this level of evidence would be sufficient to incorporate its use into routine care, and mandate payment for its use.

The evidence clearly supports the incorporation of palliative care into the care of patients with metastatic lung cancer, irrespective of whether or not they choose chemotherapy.

Tim Cousounis said...

Aggressive care (as defined in this study) does not necessarily prolong life. And it often diminishes quality of life.

Yet, until there are better delivery models for palliative care, the dissemination and adoption of palliative care as best practices will likely be a slow and painstaking process.

Helen Chen said...

I overheard the following while getting a haircut this past weekend:

"My brother had a terrible time. The cancer was already in his lungs and he decided on more chemo even though the doctors told him there was only a 5% chance. He really suffered in his last months. I don't think it was worth it...he should've gone for palliative care. They've shown that people with his cancer do better if they choose that instead of more chemo."

I don't know if the speaker was in health care, or if she read it in the NY Times, but I was glad to hear the article being discussed outside the geripal world. Perhaps it is a small sign that the tide is slowly turning?

Celeste Wilcox said...

In medical oncology training, I was taught that treatment of metastatic lung cancer (and most solid malignancies) includes an assessment of performance status. Patient who have a poor PS in most studies have no benefit from chemo.

In response to the statement about chemotherapy vs. best supportive care: if you look at the trials examining "new and phenomenally improved combination of old drugs" or "new phenomenal paradigm altering drug" these are run by companies (for the most part) hoping to have a new indication for a drug. Many of these trials allow only patients with a PS of 0 or 1 to participate. Many times, the studies are very specific about co-morbidities that are allowed. That is NOT my patient population. My office participates in clinical trials, but the vast majority of patients fail to qualify -- often because of co-morbidities, yet they are still treated (and successfully treated).

As a community oncologist, I do believe that chemo given when a patient is too debilitated shortens life. Recognizing that tipping point is difficult. Convincing patients and their families that we are to that point is even more difficult.

clay m. anderson, md said...

late chemotherapy or other systemic therapy in cancer patients is only ONE of MANY minimally beneficial and usually burdensome, toxic, risky, and costly disease-directed treatments and procedures that ill persons with metastatic cancer and other late life-limiting illnesses MAY choose to avoid or truncate when good patient-centered palliative care teams are helping care for these persons. the VALUE of the hospice and PC interventions derive from three areas: helping the patients achieve their goals and have their wishes met reflecting the accurate assessment of their condition and prognosis; excellent holistic symptom management; and avoiding undesired, harmful, and burdensome interventions. my guess is the SURVIVAL benefits are mainly in the one area of avoiding harms: treatments, procedures, ER visits, hospitalizations, etc. really a small part of what we do. most of the POTENCY of these components occur with dying people, near the end, yet it may take weeks to months to set the stage to allow PC teams to help dying persons in this way. great discussion all!
clay anderson
north kansas city, mo