Saving 28,000 Lives a Year

The New Yorker is running a piece by Atul Gawande that starts by describing the everyday miracles that can be achieved in a modern medical intensive care unit, and ends by making a case for a simple and inexpensive way to save 28,000 lives per year in US ICUs, at a one-time cost of a few million dollars. This medical miracle is the checklist. Gawande details how modern medicine has spiraled into complexity beyond any person's ability to track — and nowhere more so than in the ICU. "A decade ago, Israeli scientists published a study in which engineers observed patient care in ICUs for twenty-four-hour stretches. They found that the average patient required a hundred and seventy-eight individual actions per day, ranging from administering a drug to suctioning the lungs, and every one of them posed risks. Remarkably, the nurses and doctors were observed to make an error in just one per cent of these actions — but that still amounted to an average of two errors a day with every patient. Intensive care succeeds only when we hold the odds of doing harm low enough for the odds of doing good to prevail. This is hard." The article goes on to profile a doctor named Peter Pronovost, who has extensively studied the ability of the simplest of complexity tamers — the checklist — to save lives in the ICU setting. Pronovost oversaw the introduction of checklists in the ICUs in hospitals across Michigan, and the result was a thousand lives saved in a year. That would translate to 28,000 per year if scaled nationwide, and Pronovost estimates the cost of doing that at $3 million.

Comments

The Greek god Nike said:

[AndGodSed]

JUST DO IT!

Look at Airplanes

[corsec67]

They use checklists for everything, and flying a plane is much less dangerous than operating on someone in an ICU.

Don't they already have some kinds of checklists for "make sure we don't leave any sponges or scalpels in the patient."

Re:Look at Airplanes

[evilad]

Aviation checklist users suffer from a condition that I'll call "known-data blindness" for lack of a better term.

I've run the C-172M checklist several hundred times, and let me tell you, it's *very* easy to lose track of your place in the list, and forget whether your memory of having completed a given item is from this evening's flight, or from the one you did this morning.

This is almost never deadly in a beast as simple as a fixed-prop, fixed-gear Cessna 172. Come to think of it, I've *never* caught a condition with my checklist that would have killed me, had I missed that item.

My personal experience leaves me wondering if it's possible that checklists could cause obvious things to be *missed*.

The problem could be easily fixed with dynamically generated checklists that cannot be answered without having read and comprehended the question... but that would slow things down so much that I bet it would cost more lives than it would save.

Re:Look at Airplanes

[evanbd]

In my experience with rocket engine tests, both professionally and as a hobby, I've seen checklists be invaluable tools. I've seen them catch problems that were irrelevant, ones that would have resulted in loss of data, ones that would have resulted in incorrect operation, and ones that had direct safety impacts. However, the problem you describe is very, very common. The simplest solution is quite effective, and they discuss it in the article (but fail to mention how amazingly important it is). You need the person who is responsible for reading the list and making sure each item happens to *not* be the one doing it.

In the article, the nurses follow the checklist and stop the doctors if a step gets missed. At an XCOR Aerospace rocket test, at any given time there is someone whose sole responsibility is reading the checklist (who that is may change through the day, but there always is such a person, and who it is is always clearly defined). In both cases, the person with the checklist has the authority to stop whatever is happening and correct the situation. When I test my hobby rocket motors, the test crew is much more limited (usually two or three people, compared to at least six and often many more at XCOR). As a result, the person reading the checklist is usually also doing things on it. Mistakes are more common, and it's not uncommon to set down the checklist and just do things for a while.

That separation of roles is simple, yet highly effective. Obviously it's a bit hard in a single-pilot airplane. But, in a situation where it's at all possible, it's well worth doing. There are a number of reasons it helps, but one of the simplest is important: the reader can hold the checklist binder with their thumb pointing at the last step completed, since they don't have to use that hand to actually do anything. In the medical case, you're actually making checks on a piece of paper that goes into the file, but the idea is the same.

As an aside, having the checklist be unfamiliar is a bad thing -- mistakes and confusion are much more common after a checklist change. The fix lies in how you use the checklist, not what it says. The reaction to hearing the next step on the list read needs to be "yep, I've already got the tools in my hand" or "oh, right, nearly forgot that" -- not "wait, what was that? Oh, right I was already doing that." If you do that, people will be more inclined to ignore the checklists, because they interfere with operations.

Re:Look at Airplanes

[Chapter80]

Your comment reminds me of a programming anecdote that I have told here before. We are responsible for a software package that coincidentally has patient data in it (but this applies to all sorts of applications).

The medical staff was supposed to log all interactions, which range from medicines administered to having a conversation with the patient or parent/guardian. Everything was to be logged, so that nothing was forgotten. And nothing could ever be deleted, by design.

Well, people made mistakes (the nerve of them!), and sometimes a record would be entered on the wrong patient, and you'd really WANT to delete that misleading information. This spawned numerous debates as to whether the we should really remove the erroneous information, or mark it as bad information. For instance, if Note 5 was that a certain drug was administered, and a Doctor relied on Note 5's misinformation to do whatever was done in Note 6, by deleting Note 5, you remove the defense and rationale of the Doctor.

Likewise, if you allow temporary removal of a note, then you allow someone to "undelete", you could end up in a similarly indefensible position. Note 5 correctly says that full dosage was administered at 10PM. Note 5 gets inadvertently deleted (recycle bin). At 10:05, a nurse sees that no dosage has been administered, so administers another full dosage, and logs it as Note 6. Someone undeletes Note 5, and makes the nurse look incompetent. Patient dies. Nurse got framed. All bad.

After all these discussions, at the direction of the administration, we built a permanent delete function, so that these erroneous notes could be permanently removed. No "recycle bin". Heavy logging of what transpired and when. And an alert window warning the user that they are about to perform an irreversible action of delete.

... and the "known data blindness" (or something like it) caused people to click through the warnings. How many Windows Alert boxes do users get per day, where they just press OK. Well, we kept getting requests to "undelete something that I just deleted", even though we warned them with a Windows Alert box.

So we made the warning bigger and longer and wordier. And the rate of calls to undelete something went UP.

Finally we changed the alert box to prompt the user to do something different. In order to complete the Delete function, the user had to key in the word "irreversible" into the alert prompt.

Requests to undelete went down to near-zero.

Re:Importantly

[DigiShaman]

At least in America, people's ecconomic status in life is in constant change. Someone rich or middle class can be poor while someone poor instantly changes to middle class simply by being employeed with the right company.

Anyone who has been stuck eating Ramen for months at a time please raise your hand. God knows how many time's I've been in a funk. Today, I enjoy making around 60k a year.

The point I'm trying to make is this. We need to stop looking at people being rich or poor and see them for who they are. Human beings.

1% is actually quite awful

[syousef]

Imagine if the brakes on your car failed just 1% of the time. For every 100 times you brake 1 time you'd just keep going. How many times do you brake on an average 1 hour trip? Sometimes for mission critical systems even 99.999% isn't good enough. It's not just mission critical systems though. What about computers. If they made errors once in 10000, with several billion cycles per second, they'd be unusable.

Anyway if each patient requires 178 actions then 1% means every patient has between 1 and 2 mistakes made for them per day. I presume some of these actions are trivial otherwise I'd be amazed if anyone survived.

Re:1% is actually quite awful

[NIckGorton]

Imagine if the brakes on your car failed just 1% of the time. For every 100 times you brake 1 time you'd just keep going. How many times do you brake on an average 1 hour trip? Sometimes for mission critical systems even 99.999% isn't good enough. It's not just mission critical systems though. What about computers. If they made errors once in 10000, with several billion cycles per second, they'd be unusable.

You are comparing apples to... well not even oranges... to manhole covers. With a computer or a mechanical device it is possible to ensure that failures don't happen 99.999% of the time. With human beings taking actions that is much less reasonable.

Though if you think that is possible, go an entire day without making one single mistake. No misplacing your keys. No forgetting the milk at the store. No traffic tickets. No wrong turns while driving. No spelling mistakes while you are typing. No truthfulness when your girlfriend asks you if she looks fat in this dress. Not. One. Single. Mistake.

Of course one might argue that if something important like a life is on the line, people should be much more careful than they are while shopping or typing a reply on /. That is a reasonable question, but again as soon as there are no more motorcycle accidents, no more drunk drivers, and Vista is taken off the market we can then expect a human being to do any task with 99.9999% perfection.

Yes, and it's called LifeWings

[spineboy]

Pilots come and give talks in hospitals about how checklists significantly reduced air crashes in the USA, which it has.
I saw this presentation at my hospital, but I'm not exactly sure if it is applicable it medicine.

Re:Yes, and it's called LifeWings

[MichaelSmith]

A couple of days after our son was born I left my wife alone in her hospital room. She was sitting up to feed the baby but started slipping off the seat. She pressed the call button for a nurse but nobody came. Eventually she use the phone to call reception and they sent a security guard up to help her. The call light had been on outside her room but none of the nurses had decided to respond.

There is a hopeless lack of process in the medical industry. They need a good solid dose of ISO9001 or CMMI.

Re:Yes, and it's called LifeWings

[bickerdyke]

There is a hopeless lack of process in the medical industry. They need a good solid dose of ISO9001 or CMMI.

Or simply more nurses.

Re:Yes, and it's called LifeWings

[bickerdyke]

Nah. Mostly we could even do with worse ones. Many of the tasks of a nurse don't require special training. (Like handing that glas of water to the woman that can't reach it, making sure that guy doesnt faint on his way to the toilet and falls to the floor out of reach of the alarm button.) On the other hand, a ringing buzzer may also be a sign of an emergency. So every buzz (service and alarm are indistinguishable) has to be answered as soon as possible.

So for five simultanious alarms, you need five people, not a single better one.

Re:Yes, and it's called LifeWings

[devonbowen]

After becoming a pilot, I became a firm believer in checklists and brought them into my computer work. I make checklists for software delivery processes, framework installations, toner cartridge changes, etc. Then I ask someone else in the team to carry them out while I watch over their shoulder. And then I make improvements and put them in a well-known directory. My vacations are never interrupted anymore. ;-)

Devon

Re:Yes, and it's called LifeWings

[bickerdyke]

I'm sorry, you're right.

"worse" was the wrong word. "less qualified in the actual medical stuff"

Nurse support staff, as the other poster named ist.

Re:Yes, and it's called LifeWings

[Mr. Slippery]

Mostly we could even do with worse ones. Many of the tasks of a nurse don't require special training.

You apparently have no fscking idea what a nurse does.

One of best predictors of whether or not you survive your hospital stay is the quality of nursing care.

Nurses are responsible for infection control, for monitoring and record keeping of vital signs and other diagnostic data, and for administering medication. They are often the primary providers of patient education, and are often the ones who keep harried doctors from making stupid mistakes - in a "primary nursing" environment, nurses are the ones who are tracking and coordinating all the varied aspects of care, the ones who see the "big picture".

If you want to live, go to the hospital with the best nurses.

Re:Yes, and it's called LifeWings

[Mr. Slippery]

The real issue is that nurses are not nurses anymore. They are doctors.

Uh, no.

Doctors are less doctors and are more specialists.

It is true that more doctors are specializing rather than going into primary care. This has nothing to do with the roles of doctors and nurses.

When a person thinks "nurse", they think of a doctors assistant. Not someone that is directly responsible for giving medical care to a patient.

If a person thinks this, then that person has no idea what a nurse does. Nurses have been giving direct care to patients since the days of Florence Nightingale.

So, in a bizarre kind of way, we may be seeing a 1984ish situation where language is actually having a heavy influence on behavior.

No, we're seeing a situation in which massive ignorance of a vitally important medical profession leads to that profession being undervalued as mere doctor's assistants.

Forgive me if I am unsubtle here but my mother and both grandmothers were nurses - it hits home for me.

Personal anecdote - it works!

[bertok]

Checklists certainly aren't just for medicine, they work great in IT too. As a system integrator / contractor, I've found that lots of IT people have a somewhat.. haphazard approach to day-to-day tasks like building servers or provisioning users. This inevitably leads to mistakes and forgotten steps, which then results in angry users and system failures. Of course, the same IT people then apply the fix with the same lax methodology, and the result is yet another failure, and an even angrier user. I've seen this cycle repeat as often as 3 or 4 times, until managers get involved, and fingers are pointed. It's very unpleasant for everyone after that.

My method is trivially simple, and required nothing other than a text editor, even Notepad works fine, but a more robust editor is even better.

I call it: "Pete's Patented TODO List System". (Patent not pending).

Simply open a new file, and create a line for every single step of the task you're about to do, even if it's blindingly obvious. Prefix each step with a pair of square braces as a placeholder for a check box. E.g.:

=== BUILD SERVER ===
[o] Back up existing data from drives
[.] Patch BIOS to latest version
[ ] Reset BIOS settings to platform defaults
[ ] Boot installer from \\foo\bar\...
[ ] Configure drives
          [ ] Clean existing RAID
          [ ] Create mirror pair
          [ ] Configure block size to 64KB ...

Once you have the file, as you build the server, you tick steps off by filling in the checkboxes. E.g.:

[ ] Not done
[.] Started / partially done
[o] Done.
[!] Issue / problem
[?] May not need doing / optional / ask
[-] Cancelled / no need.

So you ask.. why is "Pete's Patented TODO List System" so awesome? Because it works, it's free, and it's flexible. I found my error rate plummeted, and I could then email the list to someone else, and they could reproduce a successful procedure flawlessly by simply following the steps. There's actually a whole range of reasons why a text-based TODO list system is the best for IT:

- It can be cut & pasted back and forth between local and remote systems.
- After a task is complete, you can email it to managers or coworkers as a "record of activity".
- Others can read your list without requiring a client such as Microsoft Project.
- Any idiot can use a text editor without special training. Quickly reorganising the structure and order of a complex multi-step task in dedicated Project management software takes effort, and may cumbersome.
- It's easy to cut & paste parameters, values, scripts and command in and out of the TODO file.
- You can reset a file to "blank" by doing this Regex search and replace: \[.\] => [ ]
- Did I mention it's free? (I do accept PayPal donations, however)

In general, TODO lists rock, especially in environments where scripting everything is not practical. For example, if you build multiple servers in parallel, simply open two Notepad windows, and track the progress of each server. If you're interrupted by a phone call or a user, you WILL forget what exactly you have or haven't done, and that's when fuckups occur.

Not to mention that most IT people just don't do "all the steps", which is one of the main points in the article. For example, do you, or the people working with you, do ALL of the following when configuring a server:

- Wipe all of the existing configuration, including BIOS settings?
- Run a memory check?
- Set the date and time in the BIOS?
- Patch up all components to the correct / latest levels, including obscure things like the network card firmware?
- Install, configure, and TEST the antivirus, backup, and monitoring modules? Did you run a test backup AND a test restore?
- Verify that every step worked?

If the answer is no to ANY of those, you or your people are fucking up on a regular basis, whether you know it or not. I've lost count of the number of times I've seen "enterprise clusters" where 2 of the 5 nodes have a different date & time. It's a trivial thing, yes, but MY GOD DOES IT BREAK THINGS if you forget.

Re:Get rid of the dinosaurs

[NIckGorton]

They reflexively cry out against 'cookbook' medicine and 'socialized medicine' while ignoring sound scientific advice.

Huh? WTF has concerns about cookbook medicine to do with the need for universal access to health care? I am one of those physicians who reflexively cries out FOR universal health insurance. Of course I don't call that 'socialized medicine' because its not different than our 'socialized' school, EMS, Fire, Police, and Highway systems to name a few.

And while I think that there are a lot of potential problems with 'cookbook medicine' there are also a lot of potential benefits. Its like any tool that can be used to help or hinder the practice of medicine. For example, an EMR that reminds me my diabetic patient is overdue for annual eye screen and should be on an ACE-Inhibitor is a great idea. However if the same system forces my hand on the ACE-I when I know that patient has had multiple episodes of fainting due to low blood pressure (which an ACE-I would exacerbate) its a problem. Similarly, if I spend all my time inputting data into the EMR it becomes more of a hindrance than a help.

In the case of ICU checklists, nurses every year are required to do more and more documentation (an average of 18 pieces of paper for a new non-ICU admission to my hospital) and every checklist or additional page you add to that is taking time away from patient care. So what sounds like a great idea may in fact cause worse outcomes because it puts the nurses focus on a paper rather than their patient.

Of course what I always find to be funny is the very same people who have zero tolerance for any risk or error and decry doctors for an untoward fear of cookbook medicine are the first ones to ask you to depart from standard practice for their personal special case... the antibiotics they want for a virus, the expensive lab test or MRI that is really not necessary, the pricey new drug on TV they want when a safer older drug with a proven track record works just as well. That innate lack of trust of health care providers and assumption of laziness or ill will translates to their own relationships with their physician in different but recognizable ways.

The actual paper

[argiedot]

You can find the actual paper in the New England Journal of Medicine. I think many here are missing the point. Peter Pronovost's suggestion to use a checklist is to ensure that commonly done tasks are done properly, not that a surgeon will have to look at a piece of paper before he moves each vein aside. And, as he has demonstrated, it works.

It's not as glamorous as discovering a cure for some new disease, but it works great.

Notice the other things mentioned in the New Yorker article:

• Nurses were authorised by the hospital administration to correct doctors when they skipped part of a procedure.
• Mundane processes were pushed to a checklist, so recalling them was no longer a human task, letting the doctors focus on the parts that actually require them to think.

In fact, the most important part of the whole article is in these paragraph:

First, they helped with memory recall, especially with mundane matters that are easily overlooked in patients undergoing more drastic events. (When you're worrying about what treatment to give a woman who won't stop seizing, it's hard to remember to make sure that the head of her bed is in the right position.) A second effect was to make explicit the minimum, expected steps in complex processes. Pronovost was surprised to discover how often even experienced personnel failed to grasp the importance of certain precautions. In a survey of I.C.U. staff taken before introducing the ventilator checklists, he found that half hadn't realized that there was evidence strongly supporting giving ventilated patients antacid medication. Checklists established a higher standard of baseline performance.

Re:Importantly

[hairyfeet]

Not to mention the fact that all it takes is a single misstep to fuck you up. Believe me I know. When I was a kid we went from rolling in dough to 3 years on welfare nearly starving. Why? Because my dad who was making damn good money at the time as a special job truck driver(because he could make a semi dance and get it into places that people would swear a truck wouldn't fit) and went to help a fellow driver who had got his load hung up. After he got it loose he went up top to check on the damage and a piece of the top crumpled under him and launched him head first off the roof onto concrete. The medical bills for putting his skull back together, plus the ICU bills quickly blew through the insurance and savings and due to the damage he was not able to work for nearly 3 years.

That is why every chance I get I try to help out those that don't have as much as me, and spend a decent chunk of my free time talking SMBs out of and raiding junk shops for PCs that I then rebuild and give to those that don't have one. Knowing that I can use my time and knowledge to make life a little better for a single mom, a kid needing a decent machine to do his homework on, or a small church that uses their donations to help the poor not only gives me a good feeling but makes their lives a little easier and at the same time saves a working PC from becoming just another pile of e-waste in a landfill. For example I helped set up a little network of donated PCs for a battered womens shelter that uses them to teach office skills.

So in this Xmas season don't think the only way you have to help is by throwing cash in a charities coffers, there are many ways that someone who knows IT can make someone's life a little easier. Believe me there are plenty of groups out there doing good work whose computers and/or networks are about to fall apart. A little of your time and some donated gear can make a big difference.

Re:At $107 per life...

[Chapter80]

While I'm a firm believer in checklists, I have my doubts about the study. It sounds to me like this might have been conducted by a guy who has a bias toward having checklists (which I do too).

It'd be difficult to measure how many lives were saved by the checklist (as opposed to other factors). I bet he looked at the deaths, and then looked for mistakes in procedures (that could have been avoided with a checklist).

But there are some serious issues with this methodology. Were these patients going to die of something else? Would the checklist have prevented the mistake? And how many patients inadvertently lived due to mistakes that would have died, had checklist procedures been followed?

This message was posted in Slashdot tradition - without reading the article.

Re:At $107 per life...

[khanyisa]

Yes the methodology is important - and in this case the article is a fascinating read - basically he compared the number of mistakes beforehand and afterwards, but also looked at the correlation with the number of deaths before hand and afterwards.

In this case the causes are a known problem (especially line infections) and they could directly correlate the adoption of checklists with a drop in the incidence of line infections, and subsequently with a lower death rate. Seems like fair enough science to me, and logical as well :-)

Re:Get rid of the dinosaurs

[jimicus]

It's amazing how resistant 'modern' medicine is to basic proven work flow improvements such as checklists, treatment guidelines.

It's not just modern medicine - this has been a problem since more-or-less forever. Go and look up a little medical history about the early use of antiseptics, anaesthetics and even such basic practices as good hygiene.

There was a documentary shown a few weeks ago in the UK about a 19th century doctor who noticed that births attended by doctors had a much higher fatality rate than those attended by midwives - he eventually figured out that hygiene had something to do with it and started making sure he and those working under him washed before visiting the maternity wards. His fatality rate plummeted but still the majority of doctors refused to change how they worked and he wound up literally driven insane because he had worked out how one could easily save thousands of lives but nobody was prepared to even give his idea a go.

Unfortunately I forget his name now so I can't easily find more information to point you at.

Treatment errors kill

[spfoo]

around 25% of the patients that die in western hospitals. It's ranked 3 on causes of patient death in hospitals - right after cancer and cardiovascular disease which together account for 50% of deaths. Fatal errors in treatment have been proven to drop to 8% in hospitals using computer technology for managing patients.

Re:Get rid of the dinosaurs

[The Wannabe King]

Probably Semmelweis.

Re:The VA system isn't broken, it's ignored.

[Gilmoure]

Ironically, the VA system provides better, cheaper care because it does not suffer from many of the inefficiencies of a market-based health care system-- preventive care and unified standards within a single provider make treatment much more effective and cheaper in the long run.

When I was a medic, I asked a doc what his opinion was of socialized medicine (had just read article about Canada's system). He said: Look around. I prefer being a military doctor. I make less, but then I don't have to worry about business expenses. personal insurance, and having insurance companies looking over my shoulder when I'm working with a customer.