Monday, February 5, 2007

ICU and bereavement in the NEJM


Probably one of the most important palliative care articles to come around in a long time was published last week in the New England Journal of Medicine. It's from a project to improve communication with families of dying ICU patients (article here; supporting editorial here). It's important in a variety of ways, but first--a summary.

The project was a French, multicenter, randomized trial of a communication intervention for families of dying ICU patients. 126 patient/family units were enrolled. Basically the only enrollment criteria was that the ICU attending believed death was inevitable (only 1 patient, in fact, survived) and that some form of limitation of life-prolonging treatment was indicated. At that time the families either received a routine end-of-life conference (held the way it happened to normally be held at that institution) or the family was randomized to the intervention. Their goal was to see if the intervention decreased bereaved family members psychologic symptoms (including PTSD symptoms, depression, and anxiety). The intervention involved giving the family a brochure on death/bereavement as well as having the intervention family meeting. The family meeting intervention is loosely described in a supplement available online (you need access to full-text to see it), was multidisciplinary, and is based on the VALUE framework developed by JR Curtis (Value and appreciate things family said, Acknowledge emotions, Listen, ask questions that allow you to Understand who the patient is as a person, and Elicit questions from the family). You may say that this sounds a little vague and in addition to that you may be skeptical that the docs had a hard time going back and forth from a routine family meeting to a VALUE one. You may be right about this but the researches can say that despite these limitations they found real differences in how the family meetings went, and about the outcomes...

...Including: Intervention group meetings took 50% longer than control meetings (average of 30 vs. 20 mins); families spoke almost 3 times as long in the intervention group meetings ( 13.5 vs. 5 mins). The care patients received after the meetings seems to be pretty similar--median survival was 1-2 days--the intervention group had slightly fewer life-prolonging interventions continued for what's that worth--this is important because it seems that the intervention group family members had better outcomes not because the patients received different care but because of the communication differences. Interestingly both groups felt like they spent sufficient in the meetings and that they had clear information. As for the study's primary outcomes: the intervention group family members did significantly better symptom-wise 90 days after their loved-one's death (lower score on a screening scale for PTSD, and significantly lower scores on the HADS depression and anxiety scales). These improvements were modest, but likely real, particularly given the relatively low N of this study.

It's important because:

1) I'll go ahead and say it: because the NEJM published it--an ICU outcomes trial in which the outcome was the emotional well-being of bereaved family members! This underlies the reality that medical culture is changing into one which more or less supports the integration of palliative care/palliative-care-type approaches to severely ill people. And, like it or not, the NEJM has a certain normative effect--like the NY Times--in defining what is important.

2) While most of us think what we do--you know, talking to patients/families, hearing their stories, supporting them in their grief--actually helps people in real ways, it is nevertheless extremely important to prove that what we do really does help people. Or else why would institutions, payers, policy-makers, and other physicians change or support us?

3) Caveats (below) aside, what was studied here was well-accepted principles of medical communication that are the foundation of all the expert recommendations on communicating difficult information with patients/families, running family meetings, etc: find out what people understand and find important, value emotions/emotional reactions, let people speak. Wisdom and experience have shown that this is a much more humane and effective approach to communication but as per #2 above it is extremely important to empirically support this.

4) Carrying #3 further the trial is important because it's a sort of proof of principle--that you can do this sort of investigation and find meaningful, and important, results.

Caveats, because there are always caveats:

1) There are certainly cultural differences between France and, say, North America and I'm not sure how to interpret this.

2) I don't want to oversell the findings as these are the first of their kind (as far as I know--if I'm wrong PLEASE leave a comment).

3) Beware that people may interpret this as something like "we should spend 30 minutes in family meetings and families should speak for 13.5 of those minutes" (the editorial does promote the idea of increased time). One, the results don't support this, although they do suggest the more time the better. This is my own personal gloss on this but the message from this trial should be the promotion of the family/emotions/values-centered family meeting and not any pre-conception about time or minutes spent doing X. It's how you spend that time--and I guess who does the talking--that matters. At least I'd like to think so...because...

4) It's possible that all the talking did nothing and that it was the brochure that had the beneficial effects. No way of knowing this from the trial. They should have given the brochure to everyone. Giving two separate interventions to the same group makes it impossible to figure out which one was the effective one....

Pallimed | Blogger Template adapted from Mash2 by Bloggermint