Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Bias in the ER

nautil.us

Post reply on HN
Print view View original

11–20 of 84 posts

Re: Bias in the ER

#11

Post by Hasz » 

  They hadn’t bothered to consider statistically far more likely causes of an irregular heartbeat. In Redelmeier’s experience, doctors did not think statistically.
Statistics assumes independence, something that may be hard to come by in the ER. Even when you start factoring in the covariance between conditions, there's so many confounding factors it's probably difficult to tease out any sort of significant r^2 for an individual patient

Instead, I would advocate for a pattern recognition method. Given a database of x million patients, it's likely someone has had the same type of case before, if not very similar. A sample would provide much less variable results than the individual patient,making diagnosis more confident.

Such a system could be used to predict symptoms before they become symptoms, saving time and preventing pain while serving to reduce hospital workload. Figure out a way to standardize it in the intake, and let the system predict what's wrong. It'll solve most of the cases, and identify new trends, leaving human doctors to focus on the edge cases like the one in the article.

Re: Bias in the ER

#12

Post by AmVess »

fencepost wrote: post #5

As a completely trivial takeaway from this, now I want to start using tissues to press elevator buttons in hospitals, and possibly bring along surface disinfecting wipes to use on them as well.

While this has nothing directly to do with the OP, but there's an assumption of cleanliness amongst everyone I know. If it looks clean, it must be clean, right?

100% completely wrong. Simple casual observation of people exiting restrooms is proof that if anything is touched, it cannot be clean. I have seen far too many people not washing after going #2 that I simply do not trust a single damn thing that I touch in public. I'm not OCD, but the general piggishness of a large portion of the population is quite disgusting.

I operate on the assumption of filth, and it has kept me from getting sick for nearly 7 years. I avoid public restrooms unless nature is screaming at me, and I carry safety wipes and use them all the time. I use them the moment when I get into my car from visiting anything, anywhere.

I filled in for the building manager at a higher end office building, and the things I have seen were shocking. I have seen non-public bathrooms defiled in disgusting ways...all on touch surfaces. Feces, blood, ejaculate, and who knows what type of drippy bile. This in an upscale business building, mind you. Imagine what horrors are visited upon your common public restroom.

I even wash my soda cans. Probably a bit overkill, and my dad used to give me stick about it until he got sick from drinking from a can that had mouse pee on it.

The neat thing is that being fastidious doesn't take much time or effort. And, it's not like my immune system became Superman overnight 7 years ago, it's that I started being careful with what I did with my hands.

Re: Bias in the ER

#13

Post by eraboli »

moxious wrote: post #4

There's a book called "Thinking Fast and Slow" by Daniel Kahneman that talks about systematic bias in human heuristic thinking. > The more easily people can call a scenario to mind, the more probable they find it. This is practically a restatement of the "availability bias" from that book. It came to mind immediately when I saw the article's title "How can Medical Professionals Avoid Making Assumptions That Lead to M…

One thing to note is that the social priming studies in “Thinking Fast and Slow” are not very replicable [1]

1. https://replicationindex.wordpress.com/2017/02/02/reconstruc...

Re: Bias in the ER

#14

Post by chimeracoder »

spqr0a1 wrote: post #10
Baeocystin wrote: post #8

Earlier quoted context omitted.

I wonder if this is why I've seen metal buttons more often in hospital elevators than elsewhere. https://en.wikipedia.org/wiki/Oligodynamic_effect

Unfortunately, just being metal isn't enough. Most metal surfaces in hospitals are stainless steel, which can hold viable pathogens for a week or more. Silver plating would sterilize some pathogens in as little as 5 minutes.

> Silver plating would sterilize some pathogens in as little as 5 minutes

That would be very expensive for very little benefit - five minutes is more than long enough for multiple people to use the same buttons at a busy hospital during the day.

It's much better to promote behaviors like washing hands before touching food or mucous membranes (like rubbing your eyes) and using alcohol-based hand sanitizers when that isn't practical. That's why you see hand sanitizer dispensers every few feet at (good) hospitals, and why infection rates plummeted when hospitals introduced them.

In addition, never touch those buttons with the tips of your fingers. Use the knuckle of your pinky instead - it's much less common to accidentally use that to touch a mucous membrane later on, so it's slightly better.

Re: Bias in the ER

#15

Post by nerdponx »

Hasz wrote: post #11

They hadn’t bothered to consider statistically far more likely causes of an irregular heartbeat. In Redelmeier’s experience, doctors did not think statistically. Statistics assumes independence, something that may be hard to come by in the ER. Even when you start factoring in the covariance between conditions, there's so many confounding factors it's probably difficult to tease out any sort of significant r^2 for an…

"Statistics assumes independence"

What exactly do you mean by this?

Re: Bias in the ER

#16

Post by js2 »

AmVess wrote: post #12
fencepost wrote: post #5

As a completely trivial takeaway from this, now I want to start using tissues to press elevator buttons in hospitals, and possibly bring along surface disinfecting wipes to use on them as well.

While this has nothing directly to do with the OP, but there's an assumption of cleanliness amongst everyone I know. If it looks clean, it must be clean, right? 100% completely wrong. Simple casual observation of people exiting restrooms is proof that if anything is touched, it cannot be clean. I have seen far too many people not washing after going #2 that I simply do not trust a single damn thing that I touch in pu…

Counterpoint: I go through none of this trouble and rarely get sick.

Re: Bias in the ER

#17

Post by djsumdog »

Hasz wrote: post #11

They hadn’t bothered to consider statistically far more likely causes of an irregular heartbeat. In Redelmeier’s experience, doctors did not think statistically. Statistics assumes independence, something that may be hard to come by in the ER. Even when you start factoring in the covariance between conditions, there's so many confounding factors it's probably difficult to tease out any sort of significant r^2 for an…

I had a professor in undergrad who worked on order entry research. He talked about one of the problems facing Vanderbilt as far as scanning costs.

CT scans and MRI are crazy expensive. Trauma patient comes in, doc orders a CT scan, it comes back and he says, "shit, this doesn't tell me what I need to. I knew I should have ordered an MRI instead."

They used a decision tree learning algorithm and trained it using attributes of incoming patients and which type of scan would have been the most useful. Then they did a trail where ER docs would enter in patient information and what they would have picked, but then defer to the algorithm's judgement instead.

I can't remember the number, but they significantly reduced the number of unnecessary scans.

Re: Bias in the ER

#18

Post by djsumdog »

moxious wrote: post #4

There's a book called "Thinking Fast and Slow" by Daniel Kahneman that talks about systematic bias in human heuristic thinking. > The more easily people can call a scenario to mind, the more probable they find it. This is practically a restatement of the "availability bias" from that book. It came to mind immediately when I saw the article's title "How can Medical Professionals Avoid Making Assumptions That Lead to M…

It's a bit different because a doctor's mistake can lead to an irreversible outcome.

That's true of some engineers (those who design bridges or the software that goes in a pacemaker or car ECU), but not others (the engineer that designs alarm clocks .. or the ones that design half the shit Big Clive buys from China).

Sure mistakes in other fields can lead to loss of work or infrastructure changes/mistakes that lead to millions being lost in revenue, but rarely something as irreversible as death.

Re: Bias in the ER

#19

Post by spqr0a1 »

chimeracoder wrote: post #14
spqr0a1 wrote: post #10

Earlier quoted context omitted.

Unfortunately, just being metal isn't enough. Most metal surfaces in hospitals are stainless steel, which can hold viable pathogens for a week or more. Silver plating would sterilize some pathogens in as little as 5 minutes.

> Silver plating would sterilize some pathogens in as little as 5 minutes That would be very expensive for very little benefit - five minutes is more than long enough for multiple people to use the same buttons at a busy hospital during the day. It's much better to promote behaviors like washing hands before touching food or mucous membranes (like rubbing your eyes) and using alcohol-based hand sanitizers when that i…

The importance of handwashing can't be overstated, it is one of the keystones of sanitation and public health! Hospital transmitted infections would go way down if staff followed all checklists to a T.

It's a matter of defense-in-depth. When the traffic is heavy enough to have less than 5 minutes between contacts, how many more people would be infected after a week?

As for cost, it would be more expensive but less so than you might think. With how thin the plating layer is, even for heavy wear, the material cost is under $500 per square meter. For small objects like buttons, door knobs, and hand rails it's negligible compared to installation and other costs. By switching to a cheaper base material it could even be cheaper over all.

Re: Bias in the ER

#20

Post by snowpanda »

projektir wrote: post #2

I wonder how much this is influenced by medical staff being overworked.

I think there's more factors than that. Air traffic controllers are very overworked and they don't often cause collisions[1].

I think this part of the article is what the big problem is:

“Eighty percent of doctors don’t think probabilities apply to their patients,”

I've recently gone through 4 years of medical treatment, and it's still ongoing. Only to find out this week that I have Lyme disease. Given that the chances of getting Lyme where I live is relatively small, it was overlooked for many years. I even asked doctors to test for it, but they refused. Meaning I'm a direct example of doctors not considering non-regular cases. It cost me over $20,000 out of pocket. Not to mention the other things it cost me.

I don't know if its a form of arrogance or how they are trained or a combination of both. But can be life-destroying for patients to say the least. The leading cause of death in Lyme patients is suicide.

[1] http://www.businessinsider.com/air-traffic-controllers-are-d...

Post reply on HN