Medication errors found in half of surgeries | Hacker News Reader

Medication errors found in half of surgeries

news.harvard.edu

100 pointsDrScump31 comments

"Of the almost 3,675 medication administrations in the observed operations, 193 events, involving 153 medication errors and 91 adverse drug events, were recorded ... Almost 80 percent of those events were determined to have been preventable. One-third of the observed medication errors led to an adverse drug event, and the remainder had the potential to cause an adverse event. Of the adverse drug events that were recorded, 20 percent were not associated with a medication error."

that tangle resolves to:

  3,675 observations
  122 (3.3%) preventable medication errors
  31 (0.8%) un-preventable errors
  18 (0.05%) adverse drug events not caused by medication error

3% preventable error rate is shocking - but does include things like documentation problems as well as incorrect dosage

btw, the clickbait headline (which is Harvard's fault, not OPs) is seriously getting in the way of communication since I went in to RTFA primed to find garbage.

Do you have experience in healthcare? 3% isn't shocking to me, in fact a number of studies of IV med administration have error rates narrowly on either side of 50% and serious error rates in the double digits. This report suggests that you're safer in surgery than elsewhere in the hospital :|

I spent over a decade working on technology solutions to prevent medication errors in hospitals but the bottom line problems that make it a hard problem to solve are:

1. There is a culture of individual blame and so (certainly most) errors go un-reported at all. Systematic identification and process solution is still not healthy, despite 16 years having passed since IOM did an admirable job shining light on the problem [1]

2. There isn't strong enough motivation to fix. It's easier to not report medication errors than to fix them. You or someone you know has probably been victim of a medication given in error one way or another. You just aren't likely aware of it. Many errors aren't very serious, or result in lengthened hospital stays (which guess what!? the responsible institution probably gets paid for). Even the consequences of serious errors aren't often (maybe even usually) attributed to the source mistake.

Like many things, money has driven the status quo (more money in ignoring than acknowledging) and it'll drive the solution. Accountability and solutions to this problem will come in the form of payers [your insurance company] finding ways to measure and not reimburse for consequences of errors. Once that motivation exists in a major way solutions will show up very quickly... it's a non-trivial, but also sub-ridiculously-hard problem to solve, from a process/technical perspective.

[Edit: I've been out of this problem space for 6-7 years now (because fuck selling things to hospitals [2]) so would love to hear from anyone who knows whether things have changed significantly and my take is now wrong/dated.]

[1] https://iom.nationalacademies.org/~/media/Files/Report%20Fil...

[2] Not that it isn't super profitable - particularly if you spend way more on steak-dinner-sales than making shit that actually solves problems! An ironically terrible/frustrating market for anyone who actually cares about clinicians or patients, though.

My background is hedge fund and consumer website management - not healthcare. My ignorance aside* - your points, and solution seem right on.

It's surprising that in these lightly-regulated, profit-driven, fields we seem to have set up effective processes for error reduction that are not standard in medicine! In trading operations we made (successful) efforts to avoid individual blame, track mistakes, record lessons learned, and continuously develop the processes and systems -- all motivated by the client's money at the end of the day though it was professional culture that was at work minute-by-minute.

* if only...

Is trading lightly regulated? Terrifying :) Even then I assume it's not hard to arrive at a hard-dollar-ROI story that drove the improvements.

Medication errors end up being a bit of a staring contest: Obviously there's the potential for value in terms of outcomes that are costing someone (payer) money.. but I think historically it's been a hard-enough problem to solve and a hard-enough problem for payers (or anyone, really) to see. Kind of a natural deadlock.

The problem has gotten some attention, though. The IMO study lit a fire, and companies like the ones I worked for chased it down with an obviously vested interest in shining a light on the problem. Bottom line is that hospitals (maybe even most of them now?) buy shit to make this not happen anymore. I wonder if the fashion around buying those things has waned though. I don't really hear about buzz/movement in this space the way I used to, and I severely doubt it's a Solved Problem yet.

Remaining (and super fundamental) problem is that it's way easier and cheaper to sell hospitals garbage than it is to build great solutions (and even if you build great solutions, it doesn't make them massively easier to sell to hospitals).

So until there's some external measurement and accountability, the entire problem space will be dominated by perception-management and not actually moving needles that matter enough.

There's a great documentary comparing the safety cultures of healthcare (individual blame) vs airlines (blame-free.)

Listen here: http://www.bbc.co.uk/programmes/p02x3vwh

I would be interested in seeing the error rates from countries with nationalised healthcare.

Yeah. If your hypothesis is that it might be lower when the payer and provider are effectively the same entity you could also look closely at the VA or Kaiser in the US for comparison. Both have historically had reputations in the past for doing better-than-average with regard to patient safety but I'm not sure whether there are apples-to-apples numbers that bear that out.

In one sense you'd think having responsibility for both sides of the equation would drive quality up by attaching responsibility for care with responsibility for cost... but in another sense it the lack of tension from going from two-actors to one can make things worse. To fix a problem like this you've got to surface it first, and surfacing it leads to lots of legal exposure while you figure out how to fix it.

See the VA scandal of the past year or two for instance. It seems unlikely those kinds of lapses in care would have been as prevalent if an outside party was taking on risk and picking up the bill for services rendered (or the consequences of not rendering them, in this case)... That's just me speculating though.

Working in medtech in the implant space, my recent revelation is that choosing the incorrect implant only 10% of the time is a good error rate. Many hospitals are closer to 25%. The only motivation to fix it is that they want to make sure they put cheaper implants into people expected to die sooner.

How is the headline clickbait? It's a true finding of the study. I would say that a headline leading with the 3% figure would be much more misleading, since the relevant figure for someone considering whether to receive medical treatment (for example) is the "1 in 2" figure. I don't care if the doctors only mess up 3% of my medication interactions; just one is bad enough.

I called it clickbaity for 2 reasons :

1) The buzzfeed style (gotta have some digits!) didn't help because any reasonable person would have said "half".

2) While 1:2 is technically true it's also the scariest number they could choose and is useless out of context. the level of fear and outrage goes down upon reading the article.

also, it really does matter how many erroneous medication interactions you have - not just operations-with-at-least-one. Most of the errors did not have dire consequences (there were no fatalities in the study). Errors are bad, but it's not all errors are equal.

Most people have no clue how much medication they receive during surgery, so saying 5% of medication interactions have errors is almost useless.

What most people care about: "I'm having surgery today, what are the odds someone will make a mistake during my surgery?"; the headline answers that question directly.

Ok, we'll put "half" in the title. Seems pretty mild on the linkbait spectrum, but still, that's a good edit.

My mom was recovering from surgery a few years back. She was in lots of pain. I should mention that it was hard getting the time of day from doctors or nurses about next steps (how did the surgery go, how long would we be at the hospital, etc).

My mom gradually got quieter and went to sleep. I little while later they came in and took her to the ICU. Turns out they had accidentally overdosed her on narcotics and depressed her respiration to unacceptable levels.

The nurses and doctors were quite friendly for a couple hours after that. It turns out that after a nurse almost kills a loved one you a small window of decent service from hospital staff.

I have family members who have collectively lost years of their lives due to bad medications and negligence, but they a) aren't the kind of people who sue b) don't have the resources or willpower to do so even if they wanted to.

I don't get it, don't lawyers work for percentage? Any lawyer would take that case as you described.

Maybe they didn't want to sue - maybe they believed it was just an accident (shit happens), and without direct consequences to day-to-day life, they felt suing a hospital (and taking money & time out of system that generally helps people) is kind of an assole thing to do?

I know I would feel like that even if I got almost killed by hospital staff, but eventually recovered to full physical and mental capacity (if I ended up somehow disabled or permanently ill, then it would be another story).

Then they would choose not to sue as a matter of principle, not because they didn't not have the resources.

Not denying what you are describing as good reasons, I'm just saying that listing "a lack of resources" a reason one can't sue in the United States is false.

Fair enough. I don't much about the legal system in the US - so assuming you can get a lawyer to work for % of settlement / damages, would that shield the family from spending enormous amounts of time on the case? My impression was that lawsuits are not only a money burner, but also a huge time sink.

Taking a medical malpractice case on contingency to trial is a huge risk for a lawyer. Nearly every lawyer would be happy to hear your story, call up the insurance company, and immediately settle for a pittance. (Hint: if you've seen a "personal injury" lawyer in an ad, that's what he/she does.) Because of student loans and other expenses, many lawyers are not in a position to do anything else on contingency.

However, unlike most insurance policies, medical malpractice policies typically include the services of defense attorneys who are happy to take a case that isn't open-and-shut to court, rather than settling. Doctors make lots of money, and they are happy to spend quite a bit as a business expense to get good insurance. Therefore plaintiff's attorneys in medical malpractice have to be choosier about which cases they take. Any number of factors can mean that a patient who really did suffer needlessly is not a good risk for the lawyer. It totally does make a difference if the plaintiff or family can chip in for expenses along the way. The situation described by the thread parent, in which the additional medical expenses caused by the malpractice ended up being owed mostly to the hospital whose nurse made the error, would probably be better handled by the hospital's accounts receivable department than by the courts. (Although since the error was ascribed to a nurse we might not expect the same blue-ribbon defense that a physician would get.)

I once served on a jury who found for the defendant in a medical malpractice case.

I will be looking for frequency of medication errors per medication. Giving someone a triple dose of advil or local anesthetic probably not too big of a deal, but a triple dose of phenylephrine or labetolol is asking for trouble. In other words, I'm curious of the rate of scary mistakes versus meh-don't-worry-about-it mistakes, though that terminology is not politically acceptable. Explaining any difference there would be fascinating.

Is this bit from the article what you are looking for?

Of all the observed adverse drug events and medication errors that could have resulted in patient harm — four of which were intercepted by operating room staff before affecting the patient — 30 percent were considered significant, 69 percent serious, and less than 2 percent life-threatening; none were fatal.

"Scary" really depends on the clinical situation. It's much more difficult to define clearly than it seems. Too much NSAID or local anesthetic is usually worse than too much of a short acting, easily countered drug like phenylephrine or labetalol.

From the article:

“Errors” were defined as any kind of mistake in the process of ordering or administering a drug or an adverse drug event, which includes harm or injury to a patient related to a drug, whether or not it was caused by an error.

How else could they conduct this study? It would be completely meaningless if observers had to second-guess after an observed harm or injury whether or not it was an "error" or just bad luck. It makes no sense to catalogue interactions which could have caused harm (but may or may not have done so) while ignoring the ones which did cause harm (but weren't obviously errors).

Then the headline shouldn't call them all errors, if they do not have evidence that they were in fact errors. "Medication errors or adverse reactions found in 1 out of 2 surgeries" is a less sexy but actually accurate title.

80% of their "errors" were actual errors. So, the number drops to 1 out of 3 surgeries for "Preventable Medication Errors".

There's a method known as "chart review" that's meant for exactly this purpose - essentially reviewing the evidence available to see if the same decision would have been made. It's expensive, but doable.

As a patient, whether or not to have surgery is a tough decision. Surgery really sucks. Even if everything goes as expected, they are cutting you up, and putting you under drugs that mess with you so badly that you don't notice them cutting you up. You spend at least a few days, if not weeks, recovering, usually involving more drugs which come with side effects and addiction risk.

And that is when it goes right.

There are always small percentages thrown at you of the chance of possible complications, so a 3% risk of a drug problem is another one to add to the list. Even if there is a much higher chance of something going "wrong" during the procedure, we're talking about a scenario where I am already trusting my medical team to handle my body, deal with what happens, and bring me back to the recovery room . So an "Oops, oh crap, better fix this." moment is not shocking to me... I just assume it happens, and trust the doctors to react appropriately.

But all of that risk and all of the problems with surgery ... you choose that to fix what is usually a worse problem. For example, I once got to make the decision to have a surgery that had a 97% chance of letting me walk again. And today, I am walking.

Would it be great if there was a 0% chance of error? Of course! But I take it much more as an area in which we can improve, not a shocking metric that would scare me away from surgery.

Medication errors and adverse drug events are a terrifyingly common cause of hospitalization and even death, especially in the elderly. Thankfully some companies are really doing neat things to try and improve the quality of care that elders receive. For example Carekinesis[1], one of my old employers, is effectively engineering solutions to help new-school eldercare paradigms - specifically PACE programs[2] - to decrease adverse drug events to nearly zero.

If you haven't heard of PACE as an alternative model for elder care (as opposed to long term care) I highly recommend you learn about it. The statistics on its effectiveness versus LTC are astounding.

[1] http://www.carekinesis.com [2] https://en.wikipedia.org/wiki/Program_of_All-Inclusive_Care_...

There's a list of common medications I can't take because of a fairly benign condition I have, which is and has been in my medical records for years. 90% of the time doctors prescribe me things I can't take, even though the information is simple and at their fingertips. I can't imagine the number of mistakes that must get made in more complex cases.

And people wonder why I would prefer to have an IBM Watson powered robot treating me. (Actually I would prefer the doctor required to work in conjunction with Watson. So many simple errors would be caught).

Now that is an industry ripe for disruption.

Reply on news.ycombinator.com