Medical errors may be the third biggest cause of death in America.

[The Romulan Republic]

http://www.pbs.org/newshour/bb/is-fatal ... -of-death/

HARI SREENIVASAN: For the better part of two decades, there’s been a growing recognition that medical errors kill too many patients in the U.S.

While exact numbers are elusive, a new analysis and estimate portrays an even grimmer picture. The new paper finds that as many as 250,000 people die each year from errors in hospitals and other health care facilities. That would make it the third leading cause of death in the U.S., ahead of respiratory disease, accidents and even stroke.

Dr. Martin Makary, a professor of surgery at the Johns Hopkins University School of Medicine who led the research, joins me now.

So, how did we get to this number? What did your research find?

DR. MARTIN MAKARY, Johns Hopkins University School of Medicine: Well, we took the best available studies, the data from the medical literature, and we basically came up with a meta-analysis point estimate, and then asked, where would that fall if medical error were counted as a disease?

It turns out that we learned that the CDC doesn’t consider medical error to be a cause of death in listing our national health statistics each year, even though the point estimate comes right in between number two and number three on the list, which means medical error is the number three cause of death in the United States. We’re just not measuring it.

HARI SREENIVASAN: OK. So, let’s talk a little bit about the methodology.

If the numbers are scarce, are these studies representative enough sample sets to be able to extrapolate this quarter-million?

DR. MARTIN MAKARY: These are studies of hundreds of thousands of hospitalizations in the top medical journals. And they are updating the 1999 Institute of Medicine report.

And there’s broad consensus that the range is somewhere between 200,000 and 400,000. Our analysis came up with 251,000. No matter what number you pick, it is well above the currently listed number three cause of death. And it turns out that the reason it’s not being counted is that the system relies on billing codes to compile our national health statistics.

But people don’t always die of a billing code. They can die from diagnostic errors, fragmented care, preventable complications. These are not things that are captured in national health statistics. That list of most common causes of death in the United States, that list is a big deal.

It informs all of our research funding priorities as a country, all of our public health campaigns. We spend a lot of time and money on heart disease and cancer, but we haven’t even really recognized that the third leading burden on health in America in terms of death is medical error in its many forms.

HARI SREENIVASAN: So, how do systems change to try to adapt for this? I remember there was a book written a while ago about the checklist, and actually preventing surgical errors just by something as simple as that.

Are there systemic improvements that we can make to try to decrease this error rate?

DR. MARTIN MAKARY: Well, there are so many great homegrown ideas by doctors around the country, hospital associations, national collaboratives.

But the important work that they’re doing is vastly underfunded and underappreciated. Our large research center in patient safety at Johns Hopkins has applied for numerous federal grants, and we keep getting message back, this is not within the scope of the NIH. This is not within the scope of the National Cancer Institute.

And all of these grants are relegated to a very small agency with a fraction of the budget, $300 million for the entire agency, including the grants. If you look at the number of people that die from breast cancer, it’s about a fourth or a fifth of the number of people that die from medical care gone wrong.

And yet they have billions more because of the great lobbying efforts and the vocal advocacy work of that group. Well, it turns out that it’s not proportional to the burden of preventable health in America.

HARI SREENIVASAN: How do you capture the number of people who might not be killed by a medical error, but might be with some serious negative health outcomes when they leave the hospital?

DR. MARTIN MAKARY: Well, studies in — even in “The New England Journal of Medicine” show that as many as one in four patients in the hospital will have some medical error that they experience, almost always nonconsequential.

And it’s estimated that about half of 1 percent to a little more than 1 percent of these errors could actually be fatal. If you extrapolate the numbers to all U.S. hospitalization, that’s where this 250,000 estimate comes from. That’s not even counting people that die at home or sometimes through limited insurance networks or cracks in the system that result in deaths.

It doesn’t include outpatient office deaths or ambulatory surgery deaths. So, we think that the estimate is a solid estimate. There’s broad consensus in the field. It’s in that range. And it doesn’t even include a lot of other types of medical errors that lead to death.

HARI SREENIVASAN: All right, so I’m a patient. How do I figure out the hospital I might be taken to or that I’m already in is kind of the lowest that it can be on these error rates? Or what do I do to inform myself? What kind of questions do I ask a doctor or a hospital about my care?

DR. MARTIN MAKARY: Well, on a national level, this is exactly why we need to measure the problem.

On a bedside level, you should always go into your office visit or your hospitalization with a loved one or family member. They’re an important safety net. And, certainly, patients that we see that come in with that support system are often critical in coordinating care.

Also, ask about a second opinion. If you’re going to have something major, like an operation or start a medication, sometimes, it’s worth getting a second opinion, because about 20 percent of second opinions are different than the first opinion. So, it’s good to know all the treatment options, be well-read, and come in with a loved one.

HARI SREENIVASAN: All right, Dr. Martin Makary of Johns Hopkins, thanks so much.

DR. MARTIN MAKARY: Great to be with you.

What the bloody fuck?

Just... how? How can their be that many fuck-ups?

[Terralthra]

The study is being posted around, but reading it, it's actually defining "medical error" extremely broadly, including "an unintended act (either of omission or commission) or one that does not achieve its intended outcome, the failure of a planned action to be completed as intended (an error of execution), the use of a wrong plan to achieve an aim (an error of planning), or a deviation from the process of care that may or may not cause harm to the patient."

The most problematic portion of that definition is "an act [...] that does not achieve its intended outcome". That means, for example, you come into the hospital with, say, an arm that was smashed in a car accident and stuck between two pieces of car for hours. The EMTs cut you out and rush you to the hospital for treatment. The doctors evaluate you and find that you're suffering from crush syndrome and recommend immediate amputation, else you will certainly die from renal failure. You agree to the procedure, your arm is amputated, but enough toxins from the crush syndrome got into the rest of your circulatory system that your kidneys pack it in anyway and you die. Thus, the doctors performed an act - amputation - intended to prevent renal failure from crush syndrome, which it did not. Did you die of "medical error"? This study would code that as a yes, but I would say no. It seems, to me, highly questionable to include procedures that are unsuccessful as errors. Some times, procedures don't work, even if executed perfectly from a technical standpoint.

[Patroklos]

I imagine routine misdiagnoses for conditions and diseases that are recognized as legitimately being hard to diagnose would be included here too.

To go along with what Terra said, how many people die from a heart attack or stroke every year after the doctors had rendered care that could have saved them but just didn't? How many people die of cancer after or chemotherapy?

[The Romulan Republic]

Well, there's a difference between an actual error and "things just didn't work out", of course. Sometimes people do a good job and something still goes wrong.

I figured, given the source, that this is probably a reasonably legit. study rather than some new age anti-medicine quackery or something, but it wouldn't surprise me if they're being overly broad. Its actually rather reassuring if they are, given the circumstances.

[Esquire]

As a general rule, if an article doesn't link directly to the academic journal article it mentions, you can safely disregard it. More seriously, Terralthra's exactly correct; the reported operationalizations are so broad as to be effectively useless - it's true that a statistically-significant fraction of medical procedures go awry, but not all of those are due to mistakes in the usual sense of the word, especially in a world where a disproportionate number of hospitalizations are of sick old people with a massively increased risk for complications that's got nothing to do with the actual medical system.

[The Jester]

I recall C0nc0rdance posting a video touching on the issue of medical errors about six years ago. One key point he brings up is that because our capacity to keep people alive is so advanced, so long as we make correct decisions in terms of treatment, we have excellent odds of keeping an individual alive. As a consequence of that though, it's only natural that the odds of death from a wrong decision are going to increase as a patient continues to live and more medical decisions need to be made.

If you keep rolling dice, you're eventually going to get snake eyes.

[Lord Revan]

We also have the matter that longer a person lives higher the chance of something "unexpected" popping up is after all we've yet to be able to halt the aging process, instead of just treating of issue that come with natural aging.

basically you don't have to even make wrong decision per say, there could jsut be something that causes a nasty complication when treating other issues.

[Simon_Jester]

Well, there's a difference between an actual error and "things just didn't work out", of course. Sometimes people do a good job and something still goes wrong.

I figured, given the source, that this is probably a reasonably legit. study rather than some new age anti-medicine quackery or something, but it wouldn't surprise me if they're being overly broad. Its actually rather reassuring if they are, given the circumstances.

In one sense the article's conclusion is valid even if the survey is too broad- because the conclusion then becomes that many patients die because a procedure did not go well enough or something was done incorrectly. Improving the efficiency of existing medical treatments then becomes a good way to save lives, even without developing new treatments.

On another level, yes, because "a quarter of a million Americans die due to medical error every year" sounds a lot more like "doctors are fuckups." Especially compared to "Every year X Americans die of medical error every year, and Y Americans die after receiving medical treatment that doesn't work even after the doctor does everything right, and Z Americans die because their disease escaped diagnosis, and X+Y+Z equals a quarter million."

I recall C0nc0rdance posting a video touching on the issue of medical errors about six years ago. One key point he brings up is that because our capacity to keep people alive is so advanced, so long as we make correct decisions in terms of treatment, we have excellent odds of keeping an individual alive. As a consequence of that though, it's only natural that the odds of death from a wrong decision are going to increase as a patient continues to live and more medical decisions need to be made.

If you keep rolling dice, you're eventually going to get snake eyes.

This is a great point. On the other hand, it still leads to a conclusion like the one in the article. Namely, that at some point the best, most cost-effective way to spend money to prolong life and increase health is to reduce the odds of snake eyes coming up. The article mentions things like checklists for surgery, and off the top of my head I can think of things like better databases for cross-referencing a patient's condition and any likely side-effects so that multiple physicians all know the same facts, and so forth.

[Knife]

The definitions are too broad to be meaningless. A lot of people come into care with chronic conditions no where even close to managed and even in some cases where the facility manages them correctly, the patient is too far gone and have sentinel events. That would count as an 'error' in the study. Americans are too used to 'the hospital will fix me no matter what' and only get there when they are sooooo messed up, that standard protocols won't fix them, nor are they really interested in being fixed, just want the pill that will fix them.

When I get a patient with blood sugars of 300 and an A1C of 12, but refuses insulin (and if they straight out refuse, can't hold them down and force them) and even refuses it after education on diabetes and effects on circulatory systems and risk of stroke, blindness, and other issues, what do you do? Besides treat the CVA a couple months later. Or kidney failure next year. Or treat the RBKA the year after?

[LadyTevar]

Technically, Nitram died of medical error --
They were x-raying him for the broken hip and a possible concussion, and TOTALLY MISSED how his stomach was swelling larger from the internal bleeding. It wasn't until I asked the ER Doc why his stomach was bigger, and why it was changing color that they realized what was happening.

Which has led to "If Only's" in the past. If Only I'd ignored the Paramedics, who told me I shouldn't ride with them because I'd need a way home, I'd have been in the ER and seen the swelling sooner. Instead, I drove, and I really don't know how long it was before I got to go back. I do know I was asked to step out and talk to someone (can't remember who) just before Martin Coded the first time.
And after the first... he wasn't able to talk, and his eyes never focused. I still wonder if he consciously knew I was there, or heard anything I said.

SHit... stop dwelling on it... it only brings tears, and nothign else....

[PainRack]

http://www.pbs.org/newshour/bb/is-fatal ... -of-death/

What the bloody fuck?

Just... how? How can their be that many fuck-ups?

Medical errors doesn't mean someone makes a mistake and cause harm. In the IOM 2001 To err is human, it's includes lapses or complications that could had been treated, so, for example, hospital acquired pneumonia, sepsis from central lines and etc are all counted.

It also includes stuff like drug side effects and harm from side effects like falls, having a blood clot post operation or pregnancy .

And as the IOM identifies, it's due to the complexity of healthcare, in particular chronic disease and the fragmented care that the result of the system. It's why you get models like primary physicians and gatekeepers. An act that does not achieve its intended outcome is used for incidents where patients aren't compliant to medications, such as they ran out of pills, or they had low blood sugar after taking insulin and etc. It's why hba1c guidelines were changed and you supposed to teach patients how to manage the side effects. So, while normal people might think knowing how many people go to the ED for low blood sugar related to medications, because they ate or drank less, or because they got a cold or etc isn't an error, defining it this way as an error allows us to study it and reduce the deaths and problems it cause.

Having said all that, the article is more about trying to draw attention and funding, because the US was the first to measure and then implement interventions to reduce errors. Zero Harm comes from U.S. Hospitals and you getting stuff like zero central lines infection in Michigan after Dr Peter Pravanoich of Mayo guidelines, and this is now set into JCI guidelines worldwide.


There's also a mixture of engineering, from engineering oxygen tubes to NOT fit into suction ports, the IA lines and etc from ICU care involved.
I also listed one such invention, the ice cream label test from the single payer thread, because that measures literacy and hence compliance/errors.


And we learn from errors... A culture that SBM is trying badly to innovate further but money..... But hey, Europe errors in giving Velcade now allow us to further minimise the side effects from this miracle cancer drug by changing the route of administration..... Sadly, it's also Europe which gave us lessons on how NOT to give the drug and kill patients by accident...

[Knife]

http://www.pbs.org/newshour/bb/is-fatal ... -of-death/

What the bloody fuck?

Just... how? How can their be that many fuck-ups?

Medical errors doesn't mean someone makes a mistake and cause harm. In the IOM 2001 To err is human, it's includes lapses or complications that could had been treated, so, for example, hospital acquired pneumonia, sepsis from central lines and etc are all counted.

It also includes stuff like drug side effects and harm from side effects like falls, having a blood clot post operation or pregnancy .

And as the IOM identifies, it's due to the complexity of healthcare, in particular chronic disease and the fragmented care that the result of the system. It's why you get models like primary physicians and gatekeepers. An act that does not achieve its intended outcome is used for incidents where patients aren't compliant to medications, such as they ran out of pills, or they had low blood sugar after taking insulin and etc. It's why hba1c guidelines were changed and you supposed to teach patients how to manage the side effects. So, while normal people might think knowing how many people go to the ED for low blood sugar related to medications, because they ate or drank less, or because they got a cold or etc isn't an error, defining it this way as an error allows us to study it and reduce the deaths and problems it cause.

Having said all that, the article is more about trying to draw attention and funding, because the US was the first to measure and then implement interventions to reduce errors. Zero Harm comes from U.S. Hospitals and you getting stuff like zero central lines infection in Michigan after Dr Peter Pravanoich of Mayo guidelines, and this is now set into JCI guidelines worldwide.


There's also a mixture of engineering, from engineering oxygen tubes to NOT fit into suction ports, the IA lines and etc from ICU care involved.
I also listed one such invention, the ice cream label test from the single payer thread, because that measures literacy and hence compliance/errors.


And we learn from errors... A culture that SBM is trying badly to innovate further but money..... But hey, Europe errors in giving Velcade now allow us to further minimise the side effects from this miracle cancer drug by changing the route of administration..... Sadly, it's also Europe which gave us lessons on how NOT to give the drug and kill patients by accident...

Sing it sister. I pass a thousand a day, what is my error rate?

[PainRack]

Sing it sister. I pass a thousand a day, what is my error rate?

The way I see it, it's an embarrassing use of language because you need to shock healthcare professionals and management into investing to reduce such incidents, instead of relying on the old tired excuses that people shouldn't make mistakes. It's just that the inflammatory language used to agitate for reform is shocking to laypeople.


And yeah... Is it an error because a person didn't remain to 1 liter fluid restriction after discharge when his kidneys are failing? Yes... Because the idea is to get admin to both investigate the rates of non compliance, as well as measure how interventions like a nurse doing a follow up call a week later can help enforce compliance.similarly, it's easier to get everyone to realise it's not because somebody was at fault that the wrong tubing was attached, rather, we should spend money to engineer ICU lines that won't fit into the wrong section or suction port. However, the very idea that people are human seems to strike some ego nerve amongst healthcare leaders...

I still remember the kind of shocked outrage i could feel emanating from my email when I told the council that their blood transfusion checklist violated every safety principle published and is non evidence based....

[LadyTevar]

Another error that has happened more than once: Oxygen pump set too high.

A pre-term newborn needed oxygen after birth. The person who set up the oxygen pump set it for an Adult patient. The pump over-inflated the newborn's lungs. The baby's lungs failed under pressure.

Yes, they were in a hurry to get the child on oxygen, so part of the error was being in a rush and not checking everything. But why was there a unit set for Adult Patients in the Neo-Natal Unit? Why didn't NICU have units set specifically for Neo-Natal and Pre-term babies?

This has happened twice in the same WV hospital: Once in 1994, to my cousin's newborn. A second time in 2013, to a family friend's newborn.

[U.P. Cinnabar]

Yes, they were in a hurry to get the child on oxygen, so part of the error was being in a rush and not checking everything. But why was there a unit set for Adult Patients in the Neo-Natal Unit? Why didn't NICU have units set specifically for Neo-Natal and Pre-term babies?

Because Maintenance/Physical Plant/Whoever delivered the wrong unit to the wrong ward, and no one bothered to check. Happens on every job site, even ones where health and safety are paramount.

[PainRack]

Yes, they were in a hurry to get the child on oxygen, so part of the error was being in a rush and not checking everything. But why was there a unit set for Adult Patients in the Neo-Natal Unit? Why didn't NICU have units set specifically for Neo-Natal and Pre-term babies?

Because Maintenance/Physical Plant/Whoever delivered the wrong unit to the wrong ward, and no one bothered to check. Happens on every job site, even ones where health and safety are paramount.

I think she's referring to the ventilator.

I not from paeds but my understanding is the error usually results from weight.
http://www.uichildrens.org/childrens-co ... ?id=234530

There's two different recommended settings involved due to different tidal volume.

Alternatively, they did set 500ml for a ventilator as the machine itself isn't unique and not all the gear are new enough to use a dedicated paediatric ventilator... I know I haven't seen one before but again, not paeds. But my understanding is that setting 500 for paeds will cause the alarm to go off even for normal machines, because it trigger other abnormalities like Fi02. But then again, I not qualified to use ventilator ...

[U.P. Cinnabar]

I thought they were pre-set, PainRack. I appreciate the explanation and clarification. Thank you.

[PainRack]

I thought they were pre-set, PainRack. I appreciate the explanation and clarification. Thank you.

Paeds is....annoyingly difficult because the formulas are all different and you actually need to consider the formulas, as opposed to standard set 500ml tidal volume, 20 breaths and then adjust based on response. So, having preset units can actually be troublesome....

And while computers are going to help, it requires you to key in the correct parameters in the first place... Not something that's always easy. I know I won't be able to transit to a paediatric unit without months of studying...And my instincts and experience will be totally off. However, in places like the UK with many NHS facing funding shortfalls, that's what's happening as they hire more temp nurses and etc to meet shortfalls.

Which leads to increased rates of errors and etc....

But honestly.... I mean, I can even use approximate to simplify dopamine drug formulas now and that was the last bit of complex drug calculation left.(please don't ask me how to use the original). You can't do that for paeds and their drug calculations are annoyingly prone to errors when using old school handwritten records.... Ug and mg and g.... Looks very similar when they packed next to per kg and you have doctors handwriting..


The other is sometimes technology being too advanced for its own good. A preset error that changed mg to g meant a precalculated drug rate got programmed wrong, leading to two drug pumps delivering wrong amount of chemotherapy and one death.And the thing is, having a human check is all well and good but some defects aren't easily checked... Such as when one line was not slotted in properly and for some reason, the alarm failed to trigger, leading the staff to think it was running in at the programmed rate until a few hours later when staff checking on the patient realised the infusion bag was visibly smaller. Thank God that we weren't using old school plastic bottles but bags, because you can estimate volume infused better.
And like James Reason said, errors happen when the holes align. Alarm fatigue from too sensitive equipment is a serious area of study in the ICU.

The impetus to bring in science from airliners and other industries is long overdue.