Thursday, September 11, 2008

White lies?

Suppose a psychiatrist is giving you a prescription for an antidepressant. And suppose the antidepressant in question has this strange-sounding yet very typical property:

If the doctor tells you it has a 10% chance of working, it will have a 35% chance of working.
If the doctor tells you it has a 50% chance of working, it will have a 50% chance of working.
If the doctor tells you it has a 90% chance of working, it will have a 65% chance of working.

What do you want the doctor to tell you? This happens every day.

[*Addendum after reading various comments: Okay, so the percentages/statistics where merely for illustrative purposes. I don't actually tell anyone a percentage for efficacy. But there is a substantial placebo effect for many medications, antidepressants for sure-- and as the doc you have the option to play this up, or play it down. And what you say can influence whether there is placebo benefit or not. I have a colleague who, I think, gives her patients more encouragement than I do that the medications will work. She might say, for example, "I really think this medication is going to help you a lot". Whereas I might say, "This one has a pretty good chance of helping you, but it's hard to predict, and we might need to try several before we hit on one that works well." I suspect, overall, that her patients do better than mine-- they may even have to try fewer medications to find one that works. On the other hand, based on scientific data, she might be exaggerating a bit. But is that the right thing to do? That's the question...]


Blogger Johanna said...

Do you have to tell me a percentage? I'd just as soon happily take the prescription and *assume* that it will work for me!

9/11/08, 1:39 PM  
Blogger jcat said...

I'd like him to say pretty much what my pdoc says....
"this might be THE one that does it"
Of course, we both know that if aspirin had the same success rate as most of the A/Ds, it wouldn't even be on the market anymore.

When one works it is so damn awesome that it does kind of make up for the dozens of no-shows before that...

9/11/08, 1:45 PM  
Blogger Rach said...

Ditto what Johanna said. I don't care what the percentage is. I want to know approximately how long it's going to take til it kicks in - how I'm going to know that it's working, and what the side effects are.
Screw the percentages.

9/11/08, 7:59 PM  
Blogger Turboglacier said...

Rach, Johanna-- Okay, I understand you wouldn't want the percentages. But assume you're the shrink, and someone's asking. Because people do all the time. And, in fact, to some degree I am obligated to tell people this even if they don't ask-- "informed consent", "risks and benefits", so forth.

9/11/08, 8:21 PM  
Blogger marcia said...

Um... give me the 90% answer and don't mention anything about the placebo effect.

9/11/08, 9:19 PM  
Blogger Rach said...

But why does informed consent have to include a percentage?... This is what frustrates me? Are you telling me because it's your duty to tell me how likely it is based on your experience with the drug, or just to appease me and lower your chances of getting sued or otherwise screwed over?

Turbo - you're a shrink! Tell the patient that every patient has different physiology, slightly different body chemistry - therefore every medication effects a person differently.
And then see what jcat said.

There's no need to bring a percentage into the mix - because that leads to false hope sometimes.

(I'll get off the soap box now... But it frustrates me when someone says - yup this is the one! but i'm only X% sure... I've given up on percentages and have started taken a wait and see approach. I seem to have better mental health as a result)

9/12/08, 5:49 AM  
Blogger Johanna said...

Well, in that case, (b) or(c) depending on how many times we've tried various drugs. If this is my first go-round, (c) - subsequent ones if the first drug didn't work, (b). The way I'm wired, if we'd tried two or three or four drugs with what I think is a 90% success rate, I'd - I'm not kidding - feel like it's my fault they're not working and probably convince myself/you that they are just to stop feeling like the freak who doesn't respond to so many drugs with a 90% success rate. (Never (a), because if I think it's only got a 10% chance of working, I won't be conscientious about taking it).

Still surprised that people ask/expect that sort of information. Seems a bit like asking "how long do I have" of your oncologist and expecting him to rhyme off "8 months, three weeks" or something.

9/12/08, 8:26 AM  
Blogger jcat said...

so I understand why you are asking, TG, given that you get asked.

But from the patient perspective, as johanna says, maybe it also does depend on the # of drugs you've been trying and how far down the road you are with a TRD.

After the first couple of no responses, it wasn't so much as not believing in the drugs as not believing that there was gonna be a cure-all for everything.

For any A/D to make it to market, it has to have some real effect on a significant number of people. And yeah, the results can include some placebo effect, but I'd like to believe that that is fairly small in the total picture. So, every A/D available has worked on SOME people.

What are the odds that maybe I will be one of the lucky ones with the particular med we are trying now? Realistically, I know not great - but if it worked on some people, why not on me? Maybe?

Don't give the worst of the odds. As the prescribing shrink, maybe give the serious S/Es to look out for, give the standard 'not all A/Ds work for all people' bit

But at least keep the possibility of THIS one working going...

9/12/08, 2:35 PM  
Blogger hilllady said...

Two things. First, if an exaggeration of effectiveness actually does boost effectiveness, then the exaggeration is part of the treatment plan. At the same time, I'd want to be told, "I think this will work for you. If it doesn't, we're going to do whatever it takes to find what does work. You aren't alone, because I'm with you."

9/12/08, 5:25 PM  
Blogger Richard said...

What do you want the doctor to tell you?

I think the shrink had better opt for the truth, because when I get home I'm going to research the med online, and if I discover my shrink is playing reindeer games with me, I'd become one pissed off SOB.

9/13/08, 12:20 AM  
Blogger Novalis said...

I agree with several of the comments to the effect that most people want, not statistics, but reassurance and confidence so long as they are not frankly misled.

So as a psychiatrist starting an antidepressant, I usually convey disclaimers like "No medicine works for everyone" or "I cannot promise this will work," but also that "I am confident this will help." One wants to be honest while avoiding the "nocebo" effect, which tends to predispose people to do poorly with a treatment.

9/13/08, 7:56 AM  
Anonymous Anonymous said...

I would want the doctor to tell me the freaking truth. This is why I don't go to doctors - they can never seem to tell you the truth, the WHOLE truth, and nothing but the truth.

9/13/08, 1:36 PM  
Blogger Bill said...

(a) and (c) are not true statements are unethical. You should never have to say "give it to me straight, Doc,"- your doctor should always be honest.

Of course, doctors are like everyone else, and dislike having to give bad news. They also don't like acknowledging that they don't know, and as a result frequently hedge their answers. On the stand this often comes down to saying something like, "There is still more art than science to a great deal of medicine." Doctors love to say this, because it means they get to be artists and scientists. This is much more modest than calling themselves gods, so it is a win all around for the practitioners of the healing arts.

A statement of potential efficacy expressed as a percentage to an individual is fallacious, but most patients aren't going to know that. Give them the drugs, and a shot of optimism, but don't try to spin it like that.

9/13/08, 2:17 PM  
Blogger brushfiremedia said...

I disagree that c) is unethical. If the effectiveness of the medication is due in part to a placebo effect, then maximizing that effect is a correct part of the treatment.

Saying that it is 50% effective when it will be 50% effective instead of saying it will be 90% effective in order to make it 65% effective would be like a doctor not proscribing a full dosage of medication.

So much of the work of a shrink is mental anyway.

And if I go home and do the research, presumably what I'll learn is that the drug is strongly influenced by the placebo effect, and so in all honesty, there is no "truth" that can actually be applied here.

Do what's best for the patient's success, I think.

9/14/08, 8:35 AM  
Blogger Turboglacier said...

"they can never seem to tell you the truth, the WHOLE truth, and nothing but the truth."

Anon.: This is point I'm making-- that there is no clear truth. The WHOLE truth would be, "This medication would work better than it really does if I told you it works better than it really does and didn't tell you what I just told you."

9/14/08, 8:01 PM  
Anonymous Anonymous said...

Statistics apply to groups, but don't identify which individuals belong to which group. Years ago when I asked the fertility doc my chances of getting pregnant on a particular drug regime, he gave me a range but then added something I have always remembered, and used with clients 'If it works for you, it's 100%." So with a particular drug, giving a range, and reminding folks that if they get a good response then their success is 100% is a good way to go,along with reminding the patient that you will keep working with them til you are both satisfied that you have found a drug or combination of drugs that gives effective symptom relief.....that's what i need to hear when i am trying to sort out my antidepressant meds....and my PTSD symptoms....same thing works for side effects...."it's fairly rare, I'm sorry it happened to you..." and then focus on treating the side effect if possible, and changing the med,etc....

9/18/08, 8:52 PM  

Post a Comment

<< Home