Hacker News Reader

Comment by gumby | Hacker News Reader

It's hard for people to evaluate risk. This morning my gf said she was "living in dread" of an upcoming procedure: 80% chance of success, almost 20% chance of doing nothing, <1% chance of things going very wrong. She's in a lot of pain so elected to do it. Her worry is affecting her sleep.

This is despite the fact that her PhD was extremely statistics-heavy, and she still does statistical analysis in her job, not every day, but more than once a week. Admittedly she's not a medical statistician, but I don't think that should matter.

View on HN

A big part of the problem is that as patients, we observe single discrete outcomes.

For doctors an aggregate 1% chance is low, but the realization of the outcome in a single person is not a percentage, its binary bad side effects or not.

Admittedly, doctors could usually do better in communicating risks (e.g. see the famous example of 1/3 risk of impotence, which patients interpret either as 1 out of 3 people, 1 out of 3 times having sex, or 1/3 of the time of sex).

> For doctors an aggregate 1% chance is low

Anybody who has played much D&D knows that a 1% chance isn't as low as it sounds.

I was thinking XCOM. Rely on a 90% chance to hit a couple of times and you'll lose the mission.

unfortunately I agree that 1% is non-negligible.. I met a man who had Lasik eye procedure around 2000 and it went badly.. prism vision for life.

Many years ago, while developing data analysis software, my employer was able to receive a 'sample' dataset of patient outcomes from a prospective client. There were many fatalities as the patient outcome of elective surgeries.

It doesn't matter what the probabilities are when the stakes are high:

- 1% chance of living the rest of your life in mind numbing pain

- 50% chance of having to remove that pesky wisdom tooth

(numbers pulled out of thin air)

In the first case, I'd really really think about it hard and deep, and question myself if it's really that necessary; in the second, meh, worst case I lose a wisdom tooth.

It's not just the risk percent but also the scale of "very wrong" that matters.

Sounds like she is proceeding with the procedure despite this fear, which seems to indicate to me a fairly astute assessment of the risk.

However, I agree people are generally not very good at evaluating risk.

How valid are those statistics? Lots of medical procedures are just estimates.

And it’s not like the outcomes are random so that <1% bad outcome with unknown causes could be 100% for a gene she has.

Her fear isn’t completely irrational as perhaps her stats knowledge is useful to not assume perfect estimation of those odds.

I always found it interesting how people assume numbers are spot on when they are frequently just estimates or averages of multiple experts guessing.

Well, how much less than 1% are we talking? I’d be worried about something really bad that had a 1 in 200 chance of happening, for example. I would be worried about playing a single round of Russian Roulette with a 200-chamber revolver.

A one-in-five chance of the pain not getting any better is frequently worthy of dread on its own.

I mean how much < 1%? If it’s a 1 in 200 chance of things go south, I can sympathize with losing sleep over it. 1 in 100k, maybe not so much.

There’s a huge difference between rationally evaluating risk and how we feel about risk. Just because I rationally understand and choose a certain set of risks, doesn’t mean I can necessarily make myself feel good or even okay about them. Even a small or inconsequential amount of risk can be really scary if the potential bad outcome is really bad.

Yeah humans are bad at exponentials and volumes. What's the difference between one in a thousand or a million or a billion? It's quite hard to treat those chances differently.

Reply on news.ycombinator.com