Skip to content
Deer Valley, Utah, from our lodge at the NPCRC retreat

I’m at the National Palliative Care Research Center’s Annual Foley Retreat in Deer Valley, Utah – rough life of the researcher, I know – and all the buzz is about the study published today in the New England Journal of Medicine.

In this study, Jane Weeks and colleagues surveyed 1,193 patients with incurable stage IV lung or colon cancer receiving chemotherapy.  They asked patients if they thought the chemotherapy might cure them.  They then tried to figure out if certain groups of patients were more likely to think that chemotherapy might cure them.  The remarkable findings:

  • 69% of patients with metastatic lung cancer and 81% of patients with metastatic colorectal cancer thought the chemotherapy might cure them
  • Non-white patients were more likely to believe chemotherapy might cure them than white patients
  • Patients who rated communication with their physician highly were more likely to believe chemotherapy was curative

That last sentence is not a typo!  Phrased differently, patients who didn’t believe chemotherapy would cure their disease rated their physicians as worse communicators.


The study had no hidden camera in the room, so they can’t tease out which factors led to these communication ratings.  Plausibly, it could be that patients rate optimistic (but false) physicians communication highly, or they rate pessimistic (but true) physicians poorly, or both.

So what should we take away from these findings?  I’m interested in your take (in the comments); here’s mine.

  • Denial is not just a river in Egypt.  It’s an incredibly powerful coping strategy, as articulated eloquently in the accompanying editorial by Tom Smith and Dan Longo
  • However, patients often make different decisions if they understand their prognosis.  They’re more likely to forgo chemotherapy if it has no chance to cure them.  We have an obligation to offer patients accurate honest prognostic information, and realistic assessments about the possible outcomes of treatments.  Saying the disease is incurable and immediately moving on to discuss chemotherapy regimens is poor care.
  • Tying physician reimbursement to patient satisfaction may create perverse incentives for physicians to give patients a false sense of optimism that treatment may be curable, so they’ll receive higher ratings and more $$$.  
  • Palliative care clinicians can teach oncologists how to have honest conversations with patients about prognosis and the outcomes of treatment, in a way that enhances trust and esteem. This is palliative care 101: Breaking Bad News.  Palliative care can and should play a major role in training physicians to have these discussions (see this terrific interview with the American Cancer Society’s Rebecca Kirch urging a greater role for palliative care based on this study).  And ideally, a patient should be referred to a palliative care clinician or hospice soon after diagnosis with these advanced cancers.

No matter what we do, some patients will continue to believe chemotherapy will cure them of their incurable cancer.  That’s not the point.  

This is about having the opportunity for informed decision making.  Every patient should have the opportunity to receive accurate information about their prognosis and outcomes of treatment, delivered in a compassionate manner.  

That’s the point.

Addendum: I just talked to Tony Back about these results, and with his permission would like to paraphrase his thoughts about the physician communication rating.

Essentially, he said the satisfaction with communication questions were asking the wrong question at the wrong time.  If your doctor tells you the bad news that your cancer can’t be cured with chemotherapy, what do you expect if you ask them about satisfaction after the encounter?

But if later, when they’ve had time to look back, if you asked them, “Did you doctor help you prepare?” you would probably find that those who had openly and honest conversations would rate their doctors favorably.  And that is the more important metric, asked at a more appropriate time.

by: Alex Smith @alexsmithMD

Back To Top