Hacker News Reader

Comment by jobu | Hacker News Reader

Often the medical errors happen on very common procedures like central-line catheters or IV insertions. Staff are overworked and tired, or they get distracted and miss one step and it results in a patient blood infection and often death.

John's Hopkins came up with a five-step checklist for central-lines that reduced overall hospital deaths by 10% http://www.hopkinsmedicine.org/news/media/releases/safety_ch...

View on HN

It always surprises me when I learn that more industries haven't adopted standardized approaches like this. Using a checklist when doing anything that may have the chance to seriously injure or kill, like for many things in medicine, just seems like a really good idea to me. Even if it takes a couple extra minutes to run.

I'm a pilot in my spare time; every single thing we do usually has a checklist associated with it. Preflight cockpit, external and cabin, pre-engine start, engine start, pre-taxi, engine runup, pre-takeoff, takeoff, after takeoff climb, cruise, descent, before landing, base and final, after landing, shutdown and park. Every phase of a flight has a checklist associated with it. Emergencies or unusual situations also have their own checklists. Hell, even walking into many FBOs, there are signs reminding you to turn off your master switches and close your flight plans.

In training, it is drilled into new pilots to always, always follow the checklist to the letter. Because when we don't, we stand a good chance of missing something. In extreme cases (just like in medicine), just one missed item on a checklist can get you killed [0].

I've found this obsession with checklists seeping into other areas of my life. I have a 15-item checklist for leaving the house when I go out of town, for instance. And in software development, I have a 10-15 item checklist I always go through before rolling code to master (this has saved me countless embarrassments on the low side and potentially bringing the site down on the high side).

TL;DR checklists are good, and more people should use them.

[0] http://www.rapp.org/archives/2015/12/normalization-of-devian...

A fundamental difference between medicine and flight is motivation. When pilots screw up they die, when doctors screw up, someone else dies. Medicine idolizes heroes, a single doctor doing everything and saving the day. In air travel the heroes tend to die. The origin of the airline checklist is when Major Ployer P. Hill, one of the best test pilots in the Army, forgot to release the elevator lock on a B-17 prototype before takeoff and died as a result.

I don't disagree. Aviation generally has a safety conscious culture, which is easy for pilots because we're putting our lives on the line each time we get in the cockpit. I like not dying, so I'm going to do all things I can to avoid screwing up. :)

Ironically, failing to release the gust lock was exactly what killed the pilot and passengers in the article I linked. Although like most crashes that was just the final result of a long string of failures.

When pilots screw up often dozens or hundreds of people die. Even the worst surgeon can't kill more than one patient per mistake.

If I didn't know better I'd think you were CGPGrey. You're right, though, checklists do make things so much better, as long as you're not tied to the list. I would imagine (please correct me if I'm wrong) that pilots' checklists are developed over time by veterans and changed infrequently. If you're developing your own checklists, be prepared to modify them. Often. This is still better than the alternative, because you are now aware of problems or steps you should have included.

Checklists are also good for helping you start things, if you have trouble getting started with stuff. It might be a little weird, but I keep a list of fun things I want to do. And I don't mean just big things ("visit London"), I mean smaller stuff. I find that after busy weeks my (old) brain is tired and doesn't want to start a new task, even if it's a fun task, so it tends to fall into patterns like "eat popcorn and watch Netflix." But if I have a list that reminds me that I could read up a bit about that microcontroller project I want to work on, or that topic I wanted to write something about, it helps me get started.

> I would imagine (please correct me if I'm wrong) that pilots' checklists are developed over time by veterans and changed infrequently.

Generally yes. For most general aviation aircraft, the checklists are part of the POH (Pilot's Operating Handbook, required to be with the airplane at all times). These are developed by the manufacturer during testing and are considered the standard and the minimum requirement for doing a check. They may be occasionally updated either by the manufacturer themselves or by Airworthiness Directives issued by the FAA (in the US). AD updates often mandate putting a placard on the instrument panel to remind pilots of some issue.

In addition, in the pilot community many veterans make extended lists with other "gotchas" to look out for or "shortcuts" to help you remember parts of the checklist (like "GUMPS" for landing - gas [fuel tank selected, pumps, etc], undercarriage (gear down), mixture set, prop set, seat belts and switches) or "Lights Camera Action" for takeoff (lights on, transponder set, radio set, GO!).

And if you're renting a plane, the owner may have additional instructions as well. So it ends up being a combination of all three.

> It might be a little weird, but I keep a list of fun things I want to do.

I DO THIS TOO! :)

Things is great for organizing this kind of thing. This way, if I ever find myself with some spare time rather than just veg or mindlessly watch TV, I can find something productive to do. Like wiring a new switch, or playing with the Raspberry Pi on my desk or something.

Related to GUMPS:

In driver's education, I learned DSMBV [doors (locked), seats (adjusted), mirrors (adjusted), belt (seatbelt, attached), ventilation (adjusted as needed)].

It's not needed in my own car, but it is still useful to me for rentals and borrowing other people's cars (or in driver's ed, where the driver changes between students), since it can be overwhelming figuring out where everything is and everything is out of adjustment for myself.

I quickly made my own version by adding one more letter to the list, by experience: DSMBVL... L for Lights!

Apparently, the acronym that more people actually use is DSSSM (aka cockpit drill), which includes steering but not ventilation or lights.

A key aspect of that checklist program was that nurses were empowered to make sure doctors followed the checklist. Doctors are resistant to adopting such programs.

https://en.wikipedia.org/wiki/Peter_Pronovost#Work

There’s a related bit regarding pilots and copilots. http://www.publicspeakingtoolkit.com/ethnic-theory-of-plane-...

(take w/ the usual Gladwell-grain-of-salt)

The story of this checklist has been covered in “The Checklist Manifesto” by Atul Gawande. It’s a great read on how diligence in work can literally save lives. We can advance medicine by breakthrough discoveries that push our knowledge further, but improving how we apply the existing knowledge is just as (or even more) important.

Gawande's essay (and book) on the subject came up just recently in discussion on Hacker News. What it all boils down to is that the medical profession in hierarchical and conservative: doctors don't want to change and don't want anybody telling them what to do.

So was the piloting profession (in part because many pilots were ex-military), and yet somehow, that culture was changed to one of standard procedures, checklists, and Crew Resource Management. I'm still hopeful that the same can happen in medicine.

The relationship between the military and piloting actually may explain why standard procedures, checklists, and Crew Resource Management were more natural than in medicine, despite both being hierarchical and conservative.

I mean, the military is itself a hierarchical and conservative organization in which standard procedures, checklists, etc. are not exactly out-of-norm.

Fair statement but the military also has medical professionals. A wild speculative guess from my perspective is that airlines view it as an objective profession, you do these things to handle the machine and that is end of the story. Medical fields are somewhat subjective since we don't really understand the human body like we do airplanes - we didn't build humans from the ground up. We need to make the medical industry absolutely objective and quantitative in order for procedures and checklists to be effective. It still weirds me out how you need second "opinions" on a diagnosis, it just highlights our inability to understand ailments.

> Fair statement but the military also has medical professionals.

Sure, the military has medical professions, but the military is not as dominant a source of medical professionals as it has been for professional pilots. So, it seems (to me, at least) plausible that military culture might have a substantially stronger impact on the piloting profession than medicine.

>Staff are overworked and tired

Yes, I wonder if anyone has studies the medical error rate as a function of how many hours the provider has worked previously that day, and the time of day. One thing I found crazy was that nurses were working three 12-hours shifts per week, but then there was overtime. This might be okay for a 7AM-7PM shift, but what about the 7PM-7AM shift? What if they changed to five 8 hour days instead? How would that effect the error rate?

Also, if demand is so high that they need to work such hours why doesn't the nurses compensation reflect that? Shouldn't they all be making 300k a year?

Oh, they are still working 40 hour weeks, just with a crazy schedule.

Typically a nurse working 12 hour shifts will do 3 days one week and 4 days the next week.

The 7pm - 7am shift is actually EASIER then the day shift because a lot less happens. The patient is sleeping, the doctors are away, family is gone, and the test labs are closed. At that point the nurse has one job and one job only, keep the patient alive until morning (a hospitals real job is not to fix you, but to keep you alive. It's your primary's job to fix you).

Of course the ED plays by none of those rules :)

The Catch-22 there is that a majority of medical errors occur during shift changeover. (Google will reveal an abundance of commentary / articles). In my personal experience (Medical school) the shift change's amongst Doctors could occasionally be profoundly short, unprofessional, and unregulated (think Resident handing off to Intern, in a rush, late in the evening, using a some notes hastily written on a notepad).

Importantly, changeovers often happen AFTER a 12 hour shift - makes it hard for this to directly causal of the changeover vs the tiredness.

Maybe Google's new project in the NHS will attack this problem by giving smartphones/tablets to medical staff and arranging the data in such a way to make accurate and fast changeover possible.

Hopefully, some of the studies showing that work detriments after 6-8 hours will spillover from select tech companies and academia to the broader market and other industries at some point.

This is not a perfect study by any means but this was just done, it's known as the FIRST trial[1] and it looked at flexible vs standard surgical residency work hour restrictions. The conclusions were that work hour restrictions did not improve mortality or patient outcomes. There is a related study that is currently ongoing that looks at the same thing but in medical specialties.

1. http://www.nejm.org/doi/full/10.1056/NEJMoa1515724 2. http://www.nejm.org/doi/full/10.1056/NEJMe1516572

If I'm reading Table 1 of the FIRST study correctly, the comparison is between doctors working up to 28 continuous hours with those working more than that. The 28 hour mark seems so far beyond the "starting to get sloppy" threshold that I'm not surprised there weren't major differences.

By contrast, pilots are typically allowed to work for 9-13 hours in a row, and a bit longer with extra support/breaks.

When I had my ACL fixed, the nurse doing the IV asked if I was worried about the surgery. I told her the only thing I was worried about was catching some hospital bound staff infection or being given the wrong medicine while I was under. She seemed a bit taken back, since it was not about my knee (I had already done the "This leg!!!" in sharpie), or her currently sticking me with an IV.

She continued to chuckle a bit until she realized I was serious.

Currently my reason for procrastinating on getting my wisdom teeth pulled. I freaked myself out by reading many stories of things going wrong. Ugh.

You don't have to be put all the way out for a wisdom tooth extraction. They can do it on local. It's safer and cheaper although you might get grossed out by the sound of your tooth shattering when they have to break it to get it out--probably not though, the drugs are very good. You're lucid enough to feel them going for the wrong tooth or something.

Just FYI, many hospitals other than JH were doing this long before JH.

JH is actually on the traditional/conservative end of the healthcare spectrum when it comes to process/algo/checklist style care, for better or for worse.

Reply on news.ycombinator.com