Rationality and Cancer

post by SwingDancerMike · 2012-07-11T06:48:18.332Z · LW · GW · Legacy · 9 comments

Today, my dentist found a possible oral cancer.

I'm 31, a non-smoker, in good health. I know the research showing that doctors ignore base rates and overestimate your chances of cancer. (I asked the doctor the base rate, he didn't know.) I know that we grossly overprescribe biopsies and surgeries, when it would be better to just wait and see. But I'm having it removed and biopsied on Friday, even though I don't have dental insurance and it's costing me $1,000 of my own money.

Why?

I thought this would be an interesting case study: Introspectively, what's going on to make me ignore my rationalist training, ignore the external data, and choose what I know is probably the less optimal path?

My first thought is embarrassment: If I do nothing, and it turns out I have cancer, will my support network roll their eyes and blame me for not being more aggressive? My feeling is yes, that even though they wouldn't do it to my face, they would secretly blame me, and become less available.

My second thought is fear of the unknown: I roughly know what the biopsy will entail. It's a light anesthesia, a few stitches, and 1-2 days of recovery. No big deal. And $1,000 isn't tiny, but it's not a big deal for me, either. In contrast, what happens if I don't do the biopsy? Huge, scary unknown. And, even if I know that I only have a 0.01% chance of having cancer (to guess a number), I also know my emotional mind is bad at math, and I'll have great difficulty controlling its worry. And so, it's rational to buy some level of anti-worry insurance -- I don't know what the rational value of that anti-worry insurance is, and I don't know if that value exceeds $1,000, but clearly, anti-worry insurance has some positive value, and probably a fairly high value.

There are other considerations. I need some wisdom teeth removed, and we're doing them at the same time, so adding the biopsy doesn't affect the recovery time. And I have a 6-week trip to Australia coming up, and I'd hate to have problems while I'm traveling. But mostly, I think it's embarrassment and worry.

By the way, this is a personal matter than I'm choosing to share with you. Honest advice about how to handle it, particularly from people who've faced similar decisions, is welcome and appreciated. Flames about how I'm ignoring research are not. Thanks for understanding.

Update: I talked with a friend who does research in medical decision making. She explained that, for the specific population a doctor serves, he's usually fairly accurate in his estimates of how prevalent a disease is. She also encouraged me that I'm right to get the biopsy. I feel much more relaxed, and I realized that I was feeling guilty about being irrational. I'm sure there's some meta-lesson in there that I'll figure out someday.

9 comments

Comments sorted by top scores.

comment by falenas108 · 2012-07-11T13:18:39.408Z · LW(p) · GW(p)

There's a simple utility calculation going on here. I'd say the chances of having cancer given your dentist says you might have it is much higher than .01%. Without doing any research, I think it's safe to assume it is at least 10%, probably more.

So, you have a 90% of wasting $1000 and mildly inconveniencing yourself for a few days, vs. a 10% chance of having major oral problems in the future because you didn't get this treatment. Plus all the social stigma you mentioned earlier. With this analysis, it seems perfectly reasonable to go through with the biopsy.

Replies from: SwingDancerMike
↑ comment by SwingDancerMike · 2012-07-11T13:27:13.348Z · LW(p) · GW(p)

Thanks. I'm not entirely sure about the 10%, but you're right: When we run the math on breast cancer given a positive mammogram, it's typically a few percent, not 0.01%.

By the way, lest people worry about me too much, my dentist thinks it's most likely pre-cancerous, and that if we remove it now, there's nothing to worry about. Or rather, that it's most likely nothing, maybe pre-cancerous, but probably not actual cancer yet.

Replies from: Decius
↑ comment by Decius · 2012-07-12T06:20:37.321Z · LW(p) · GW(p)

Consider the entire probability space: The odds of a major surgical accident, multiplied by the cost of a major surgical accident, against the odds of this untreated anomaly becoming cancerous, multiplied by the cost of that.

Go down the list towards more probable things, and I think you'll find that the median outcome is losing $1k, but the expected outcome is saving 2% of your jaw and some embarrassment. (Or some other percentage).

comment by MileyCyrus · 2012-07-11T16:22:14.687Z · LW(p) · GW(p)

It's wise to be aggressive when you think you might have cancer.

Another tip: if you think you might have serious medical condition, get a diagnosis immediately. If necessary, pay out-of-pocket. Once you have a diagnosis, you can begin treatment. This is important because early treatment usually means a better survival chance. But if you don't have a diagnosis, your insurance company will deliberately waste your time, claiming that they aren't really sure if you have a medical condition and that you need more tests.

comment by shokwave · 2012-07-12T20:30:50.759Z · LW(p) · GW(p)

I think focusing on the chance of cancer is a red herring. Human brains being what they are, even if you notice you're privileging the cancer hypothesis, you're still going to worry. Sure, do the math, but save that result for another calculation.

How much utility are you likely to derive from $1000? How badly does suspecting cancer disturb your peace of mind? And how much is your ability to derive utility going to be reduced when you are without peace of mind?

We could do the math, but it seems clear to me you know something about the shape of these equations: you opted for the biopsy.

comment by ChristianKl · 2012-07-12T19:31:23.656Z · LW(p) · GW(p)

How about getting a second opinion from a second dentist?

comment by shokwave · 2012-07-12T20:39:51.199Z · LW(p) · GW(p)

I have a 6-week trip to Australia coming up

Are you in Melbourne for some of that time, by any chance? It was great fun at the May minicamp, catching up would be great. There's also a LessWrong community that are mostly friendly - although I can't promise you any of them know how to dance.

comment by Kingoftheinternet · 2012-07-11T16:22:56.837Z · LW(p) · GW(p)

Robin Hanson has something to say on this issue, of course. For breast, colorectal, prostate, and lung cancers, at the very least, screening seems to have no expected impact on your overall chances of dying.

comment by billswift · 2012-07-11T10:21:19.391Z · LW(p) · GW(p)

I talked with a friend who does research in medical decision making. She explained that, for the specific population a doctor serves, he's usually fairly accurate in his estimates of how prevalent a disease is.

Any evidence that she is any more likely to be correct than your doctor? Isn't this another case of confirmation bias? She told you what you wanted to hear, so you believed it without much, if any, support. Remember, even most published medical research turns out to be wrong.