Number Needed to Harm: Treatments Can Hurt You
Вставка
- Опубліковано 24 лип 2024
- Last week, we discussed the NNT, or number needed to treat. I'm sure it made a lot of you upset to realize that many therapies you've been sold as "awesome" were, in fact, somewhat incremental with respect to benefits. But another problem is that a lot of those therapies are anything but benign. They come not only with costs, but also with side effects or problems.
We can quantify harms, too. Watch and learn about NNH!
Almost all of the data for this came from the amazing website TheNNT (thennt.com). You can go there to see those and more.
Additionally, Aaron's new book is out! Please consider buying a copy. He'd really appreciate it! dontputthatinthere.com/#buy_th...
John Green -- Executive Producer
Stan Muller -- Director, Producer
Aaron Carroll -- Writer
Mark Olsen -- Graphics
/ aaronecarroll
/ crashcoursestan
/ realjohngreen
/ olsenvideo
I'm coming up with a new acronym. NNE, number needed to educate. The NNE for watching Healthcare Triage's videos is 1. That means if just 1 person watches Healthcare Triage's videos, that 1 person is instantly educated, and I think the likes/dislikes bar proves that.
Lol great video as always, keep up the fantastic work.
Wow. right on.
I was a critical care RN, who was prescribed a statin, at 29 years old, due to a cholesterol level in the 270-280's, and every pharma funded conference I attended, and what I learned in nursing school as well as my 12 years as an RN. And I conceded and agreed to take lipitor 10mg, as I desperately wanted to drastically reduce my risk of a heart attack, stroke or death, and from what I was taught, by taking my statin, my chances of benefit, greatly surpassed any potential adverse effect. (Wish I would have learned about statistical manipulation). Fast forward 2002, I was hospitalized for 28 days, with Alzheimer's like dementia, I wet and soiled myself, I could not walk, nor speak coherently, I did not know my wife, nor my 3 year old son, my parents nor siblings. Brain MRI revealed dozens of scattered lesions throughout the grey and white matter, biopsy revealed the lesions to be Apoptosis (Programmed cell death), Electron microscopy revealed Mitochondrial DNA mutations most similar to Mitochondrial Encephalomyopathy with Lactic Acidosis and Stroke Like Episodes (MELAS). Doctors prepared my wife for my grave diagnosis, and were waiting to transfer me to a nursing home at 34 years of age. When a visiting professor from Johns Hopkins was asked to evaluate me, and started me on a Mitochondrial cocktail (essentially a handful of vitamins, amino acids, and one KEY componend, a robust dose of CoQ10..and within 24 hours of starting this cocktail, I began to be able to communicate the need to use the restroom, I recognized my wife and my parents, I could speak coherently, walk with assistance, and ultimately was sent home with aggressive rehab. Shortly after discharge, I saw Good Morning America's Diane Sawyer, interviewing Dr Beatrice Golomb MD PhD, principal investigator in the UCSD Statin Effects Study, discussing statins and the unpublished risks they have, etc I enrolled in the study, sent my records from UW Madison, and when the study concluded I was called by dr Golomb, who informed me that I was one of a handful of patients in her study, with strikingly similar courses of Illness, and biopsy findings, she referred our cases to Mitochondrial Disease Expert, Dr Doug Wallace PhD or UC Irvine, who opined that "my use of Lipitor was the causal contributor to the holes in my brain (apoptosis) as well as the mitochondrial DNA mutations. Ive been disabled since Oct 10, 2002 due to the lasting effects of the brain damage from Lipitor. I have poured my heart and soul into researching, and do everything in my power to spread these truths like wildfire, to anyone who will listen to me, Thanks Doc Carroll
As a parent, I'm really glad you've made this and the previous video. I feel like it will give me a better way to communicate with doctor's and allow me to make an informed decision. Thanks.
personal notes: NNH aspirin 3333 major bleeding event. NNT statin MI 60, stroke 268; NNH diabetes 50, muscle 10.
Ty for your simple explanation! I am a PhD student in nursing and your channel is a great asset to my studies!
I'll be showing this video to my parents. I'm sure I'll have much more success convincing them about understanding the risks of meds with this! Thanks! :)
I love how you explain complicated medical situations in easy-to-understand wordings. Thanks!
This stuff is complicated for most people, including me, to understand. VERY good information is being put on this channel and I appreciate it greatly.
Really liked this and the last video, puts things into prospective. Thank you!
Everybody should know this.Sadly I did not till now.
Nice work. Love the risk stuff.
I love you for making this. I hope millions of people worldwide see this so the world can be better
Thanks for doing this. This should be taught in schools.
I wonder that all the time...
Definitely need to share this!
Super interesting again! Thanks. I love this channel.
The two videos about NNH and NNT are awesome considering all the pandemic related debates going on. And more importantly we know the content is not biased due to covid-19. It may be biased for other reasons. But knowing people who prepared this video doesn't have bias for or against covid-19 vaccines is a stress and anxiety relieving. Thank you to all involved.
Another great video, thanks.
Very informative. Thank you!
I wish this kind of information was more easily available. Doctors should inform their patients about these things thoroughly.
Thank you for this episode.
You guys should do another Questions video.
I'd love to see a video on ADHD either from an effective treatment or a diagnosis standpoint . Thanks! I look forward to this show every week.
I still love these fundamental videos. They are so important.
It is very nice ,simple way to explain NNH, good jop👍
Another great video! You brought up a few treatments that have greater NNH than NNT in terms of the risks versus benefits are - aside from Statins and antibiotics for ear infections, I wonder what some notorious cases of that are in terms of medical treatments out there.
Love your stuff! You should do a video on whether certain drinks like soda, coffee, and alcohol cause dehydration.
Well done
Thank you!
I wish Dr. Carroll was my doctor when I was a kid...
What a hero! Fight that antibacterial resistance AND promote efficient well-being! This guy's literally my hero!
Thanks so much
What about the NNH of NOT giving antibiotics for ear infection? Isn't the reason to treat with antibiotics to prevent tympanic membrane rupture, or mastoiditis?
Same question I had. I think he should have pointed out the risks of not taking the antibiotics. Maybe if enough people like your comment he will.
Most ear infections in kids is self-limiting, they go away on their own without any complications in a week or so. Antibiotics have been shown to only reduce the infection time by a day or two, and even then the evidence is sketchy. There's very little difference in risks between taking the antibiotics and not.
Practically, ear infections usually don't require antibiotics. Ear infections mainly become a bigger concern if they are recurrent, in which case the best management is grommets, not antibiotics.
Some kids, especially those who continue to have ear infections or who have a particularly nasty microorganism (which is pretty rare), do get antibiotics to help reduce recurrences, but its not a definitive treatment, more an adjuvant.
Did you not see the part where I said that "none would have less serious complications?" :)
Go see www.thennt.com/nnt/antibiotics-for-otitis-media/ . Kids on antibiotics did not have fewer serious complications. In fact, in the studies of 7000+ complications, the only case of mastoiditis was in a kid on antibiotics.
Todd Klopfer See below.
Healthcare Triage Fair enough.
Love the videos. Is there a way to get a transcript to keep on hand?
Fascinating! So I'm right most of the time to just grin and bear it with common illnesses?
This video seems to have the highest approval percentage out of all the others.
I would very much like to know more about developing tolerances for certain drugs. My mother has a very high tolerance for pain meds, which has made her a borderline addict of pain medication. I no longer live with her, but I know that back when I was she was taking pain medication every day. High dosage Motrin most of the time, sometimes vicodin. Her usage and the pain levels she had regardless always scared me, and now most of the time, when I can, I tend to suffer through pains instead of taking something. I'd just like to make a more informed decision about how often I choose to take pain medication.
Also perhaps a refresher on the types of pain medications, and what they're best used for? I think that would be a very useful resource.
I'm understanding this, but it's still a lot of information to take in. XD
I haf the low-dose aspirin discussion with my Optometrist... The reason I did was that a study showed taking low-dose aspirin increased the risk of retinal bleeding. His answer was that he took low-dose aspirin because retinal bleeding could be fairly easily treated, but the consequences of a heart attack (other benefits of aspirin aside) were much more severe... potentially fatal which could not be fixed.
"I recently discussed this in a piece in the New York Times." #HumbleBrag? ;)
uhhh...nope. A humblebrag requires that you use humility or self-humilitiation as a vehicle for bragging, and he wasn't pretending to be humble or really bragging about it.
lol Ikr
Are the NNT and NNH easy to find? If so, where? Do doctors have better access to this information? Anyway, just curious. Great video and interesting topic.
I would love to know where to easily find this stuff as well. My gf has Chron's and if we're gonna be a thing I want to make sure she's getting the best treatment and the treatment she wants. If I can find these numbers I can ensure that.
You can find that type of information in medical articles. Check out pubmed www.ncbi.nlm.nih.gov/pubmed, It's a search engine by the NIH that keeps track of most of medical literature in the world. You won't be able to get most of them though, unless you have a personal/institutional subscription.
Serge B yeah I always forget I have access to way more information as a student, which is absolutely not how things should be
Pay to be a member of a university library (which has thousands of journal subscriptions) and you'll have as good resources as any doctor. But that's not the hard part. Doctors have better access because they are trained to find and filter relevant information.
@@Srewtheshadow check out www.thennt.com
Lovely. All your stats stuff is reasonable. Esp the Bayes & result interpretation, & that is just the tip of the iceburg, & no Drs have time to know what test is even being done at which lab, let alone current debates in that specialty. Yet patients being active in their health is discouraged. But is it getting better or worse? I shall consult the Tarot at once!
For things like statins or asprin which are taken continuously, are NNT and NNH calculated per year of treatment?
can you do a video explaining everything there is to know about the placebo affect?
this is good stuff. much appreciated. -er doc
As a pharmacist I am your big fan 💕
We need one of those White House Petitions now:@aaronecarroll for Surgeon General!
Have you made one?
Even with the simple language Dr. Carroll uses, this is still a very complicated concept to get your head around. Still, though, it's very useful information!
"I recently discussed this in a piece in the New York Times." Woah! Name drop! :D
All this made me think of was that player's chant in the ball field scene from Parenthood, "When you're feeling lots of thirst and you're sliding into first, DIARRHEA." So, a minute it, I was done.
Would you do a video on soy? There is so much misinformation regarding its benefits and drawbacks etc
Interesting videos on NNT and NNH, but I'm wondering about how this can be so absolute like in so and so many people "will see no benefit at all". Isn't the actual outcome often somewhere in between? For example diarrhea can be mild or worse. Is there a way to take that into consideration? As an extension to that maybe you can do a video on that and explain how they come up with those numbers to begin with.
Wow, you sound like a really good doctor. I think I'll buy your book. Do you know if it comes in eBook format?
I don't know for certain it does, but I would strongly suspect so.
It does!
Yaay, thanks for replying! I'm going to go read it now.
I wish I could send my kids to this guy when they got sick.
Real health (coming from ACTUALLY right diet not trendy crap) is proven by your kids not getting cold or flu or ANY of the related inflammatory "diseases" for at least 3-5 years in a row, or more. Then and only then do you know that they are getting fed right!
I came here to learn more for journal club and now I'm just questioning everything I learned about statins.
Can you do a video about veganism/vegetarianism and compare the health benefits/risks what to be aware of and so on and so forth? It would be really interesting to watch :)
Is there a way to use the NNHs for all the significant side effects of a treatment and the risk of death from those side effects to get a number needed to kill or something?
I found your explanations about NNH and NNT very clear and eye opening, but I was left with some questions: What about taking antibiotics after you've had a culture taken, where they check responses of your current bacterial infection to various types of antibiotics? What is the NNT in that case, and is it not near 1? What about the chance of a complication, like an ear infection spreading to the nerves? What about bladder infections that might spread to the kidneys or cause death (?) if untreated? Can the body overcome these common diseases? I think not for many common diseases, because in the past people simply died of infections and such. Also if we take cancer for example, even though there are harsh, sometimes lifelong, side effects to the treatment, and even if their chances of success are not high, people often opt for treatment because their high chance of death of they don't. Also, sometimes you try a treatment and if it doesn't work you try a second one, and then the combined NNT may be lower.
Ah yes, the other reason I hate taking antibiotics. I'd rather be sick for a week without an upset stomach than be sick for 5-6 days with an upset stomach. I realize it may be different for other people, but I'll stick with trying to fight things off on my own first.
do one on tap water! bottled water! distilled water!! and the benefits/imeditments of them all!
Anastrozole has a NNT of 25 (from what I was able to find). The side effects are horrible. What is considered an acceptable NNT when you consider side effects and cancer vs non-cancer outcome?
Aaaaah! Math! Run awaaaaaaay!
But in all seriousness, a very informative video. What I understood of it was, anyway.
This is quickly becoming my favorite UA-cam channel. Thank you for explaining the often times (intentionally) difficult to understand world of medicine in plain language.
Under 301 club! Keep up the great videos
right now .. you are my favorite
Hello! I loved the video, it's awesome, BUT...there's a typo around 3:45 (Unnecessary Surgery). You might want to make a notation of some sort before the views get too high. Great stuff though.
Can u please give ur opinion on dental X-rays on children 5 to 10 year old ? Thank u
I am NOT a doctor. Kindly take this post with a grain of salt. That being said, the occasional X-ray is fairly benign. So long as your child is getting one every week, its unlikely to do much, if any, harm.
However, that doesn't mean we should go around giving people X-rays just because we can. Is there some benefit to getting the X-ray? If so, what?
Are there also studies about long term effects?
I'm wondering if taking medication for every little thing harms our natural immune system
Question: with a lot of these new meds saying that it will cause suicidal actions and with most states punishing those who try to take their life. Will that person on that medicine, assuming it was the medicines fault, still be punished?
how are states supposed to punish suicide victims?
There is been a big push by NICE on statins in the UK which has the medical profession in uproar
So in terms of the children who have the ear infection I take it that the infection is a rather minor one that will subside without the use of anti-biotics? What are the risk factors when you do not give the child anti-biotics?
This is a fair question my neckbearded-fedora-wearing friend, an actually very relavent.
TBH though, having had tons of ear infections in my life... You'll be fine.
Srew the Shadow I figured that. I was just wondering what the complications were and what the chances of complications are. Mostly because the last time I went to the doctor with an ear infection they gave me both anti-biotics and painkillers. So it is mostly out of curiosity. Thanks though!
As he said in the video, the NNT to avoid a complication was pretty much infinite.
I don't agree with this. It is purely mathematical and depends on how much you trust the publications. You need to evaluate on a case by case basis, not a paper.
😮great
Seeing as I'm at the beach this week, I kinda wish Sunscreen was one higher in the schedule. :l
I recommend you tackle food coloring!
Can I get reference for statins of NNH infinity
Nice
He continues to show me that I did not pay enough attention in math class.
is there any situation where relative risk is better than absolute risk?
Does the Dr. have a column on the NYT? Or he just talked for someone else's column?
As far as I know, he doesn't have a regular column. He has, however, had his pieces included in many large newspapers and TV shows. Here's his bio - theincidentaleconomist.com/about/about-aaron/
I write regularly for The Upshot at the NYT. In fact, I have a piece there today: www.nytimes.com/2014/07/15/upshot/why-improving-access-to-health-care-does-not-save-money.html?rref=upshot
Healthcare Triage
Thanks for the reply. I'll check it out. I really enjoy the videos :)
This should be taught in grade school. It would solidify a better understanding of maths.
If what he said about statins is true, which I don't doubt since everyone I know who's taken them has had awful side affects, why is the NHS in the UK recommending EVERYONE over the age of 40 should be taking them weather or not they have any risk factors for either hard attack or stroke?
Way to point out the problems with statins!
Next topic, suncreen. You read my mind.
I believe the NNH for stains, overall, is probably around 5.
Well shoot. I had an ear infection that lasted for over three months and ruptured my ear drum. We went with antibiotics, after the doctors suggestion, and the infection was gone shorty after one round. I was told "most people" experience diarrhea but I was completely fine throughout the full round. I guess that makes me one lucky son of a gun.
Is it my speakers or is the sound a little off....ringing..in the beginning??
I'm having difficulty phrasing this eloquently, so please bear with me ... Is there a reason not to state a change in absolute risk in terms of the number of people who, in the absence of treatment, would not have experienced the effect in question, but who will after receiving the treatment (ie. excluding from the pool the number who would have experienced that effect anyway, even in the absence of treatment)?
This would mean that a change from 10% absolute risk in the absence of treatment to 20% absolute risk with treatment would be phrased as 11% (1/9) risk introduced by the treatment.
I would think that this would be more useful than either relative change (2 times more likely) or change in absolute risk (10%).
But then again, this is my first time really contemplating the issue, so I might be missing something gigantic and obvious.
That was expert product/service placement at the end there.
What, saying that pain meds reduce pain more than antibiotics?
That's not product placement, that's common sense. Product placement is when a specific brand/product is mentioned or advocated for. Did you hear any brand names?
I know I only use Extra Strength Pain Medication®.
You do realize that by pain medication, he was probably referring to Children's Tylenol or one of its many *GENERIC* and *INEXPENSIVE* variants.
you have absolutely not idea what you're talking about
yep.....he managed to simultaneously push every single pain killer on the market at the same time....oh wait...that's not product placement
The ear infections he's talking about must be different than the ones I get. I've had at least 100 in my life (I'm 33, still get 1-2 a year) and antibiotics *always* make things better within 12-36 hours. It doesn't matter if I start the antibiotics as soon as I get the ear infection, or a week later (I was in a foreign country, it sucked). But they always work immediately.
You glossed over the fact that all of these events are more or less not mutually exclusive if not independent. While getting diarrhea may or may not be independent from having a benefit from antibiotics, it's not like you can't get both a benefit and a harm, and it's also entirely possible for nothing to occur.
5:42 Guess which option people normally tend to choose... *hint hint* antibiotic resistant bacteria are on the rise.
"So what I told you was true... from a certain point of view." Obi-Wan Kenobi.
I feel so smart now 0__o
Dr carrol
i think the problem with teaching laymen about absolute risk reduction is that people outside science and even amongst ourselves have a poor concept of the purpose of PREVENTION. For example, the absolute risk reduction of mortality in smoking heavily vs not smoking in men nears 25%, but in a "its not going to happen to me" mentality people might go: but there is still a 75% chance that i WONT get cancer... Thus #YOLO .... its also hard to make others understand that it increases the risk factor for other diseases, since we can do this same exercise ad nauseum for each other risk factor.
its also hard to make the case for vaccination if laymen only use absolute risk reduction and dont factor in the exponential increase in incidence that occurs as herd inmunity decreases, again the person might go: (very unlikely) to happen to me or my children (right now) so... #YOLO
that being said. We run into this very problem with this recent paper on home-births
www.ajog.org/article/S0002-9378%2814%2900275-0/abstract
big relative risks but small absolute risks
so... should we advise our healthy patients to try home births since they are much more cost effective than a hospital birth?
Do you see the problem here? How do we advise our patients to grasp the purpose of our interventions if we can even show them that the ARR is that large? How do we promote prevention and lifestyle changes? How can we even justify annual checkups, patient follow ups, ancillary studies to monitor inpatient care and diagnostic studies?
I would suggest that the solution to poor understanding of principles of statistics and other scientific fields is not to maintain that poor understanding by neglecting to teach 'laymen' about those subjects 'because they won't understand it'.
Do a better job of teaching, then.
Statistics is an area of knowledge that many healthcare practitioners, scientists and even mathematicians and epidemiologists still struggle to fully grasp (and often have to agree to disagree) what exactly is the best study for a given set of data, how to present said findings, and how to define its relevance in the real world? (Patients)
this isnt the same as discussing diabetes with someone, these concepts are quite abstract. The level of education and understanding of our patients also vary greatly.
How would you suggest we discuss these topics to our patients within our average 15-20 minute encounter when someone is complaining of an ailment that merits direct attention?
Wait, so statins are not worth it? That'd mean so many people use a drug that is more dangerous than effective. Is that really the case?
They are worth it when you think the benefits outweigh the harms. That's likely true for some people and not true for others!
Healthcare Triage like infant genital mutilation.
Anne Wunsch +
But let's not discredit the RR, that's a very important statistic for epidemiological studies.
He has no intention to discredit, he has every intention to educate us enough to decide for ourselves if we want X treatment and to seek the numbers to back our decisions.
Srew the Shadow I know he knows that, my concern is with the lay folk
swill128 His concern is with lay folk, who may hear their risk of negative outcome will DOUBLE and decided to not have a treatment, when the absolute risk only went from 0.1% to 0.2%
low fat milk has only 1% less fat than regular milk, but that's half the fat per serving. I know that's not how RRs and ARs work, but my point is that changing the stats you use doesn't change the result. The AR of lung cancer in smokers and non smokers isn't much, but the RR is WAY higher among smokers. Does anyone think smoking is healthy?
swill128 That is some seriously twisted logic. Smoking isn't healthy, so we shouldn't put the risk in perspective and instead only tell smokers their relative risk in order to scare them more? We should probably multiply by 100 and add a percent sign too to get a really scare number. 2? how about 200%! I mean OK... maybe that would be for the public good in the case of smoking, which has NO health benefit, and so should be indiscriminately minimized, but makes no sense when weighing things with positive and negative outcome potentials.
So, wait, are you telling me that you generally do not treat ear infections with antibiotics? Are you speaking in terms of only oral abx? Do these children use topical antibiotics? Drops? Or are they just put on a pain medication and then fingers are crossed that the infection doesn't get worse? This doesn't seem like good medical advice to me. I'm not a pediatrician, though... but I'd like more clarification on that.
Yea, I understand it wasn't the point of the video, but it certainly did pique my interest.
daan, nobody ever told me this, including my doctors,from china
I feel as though you have rational patients :)
The problem with today's world is everything is a lawsuit. I guarantee you if one of those people you recommend no antibiotics to ended up in the hospital, that you would definitely have a subpoena on your hands. There should be better protections for doctors. It's not fair to ask practitioners to reduce antibiotic use but at the same time penalize them for getting things wrong.
I heard though, there is a test coming out that will help doctors determine that if an infection is viral or bacterial. This will help reduce antibiotic prescribing.
That's the reason why homeopathic medicine was so successful in its time. Sure, it didn't actually helped you, but back then the alternatives (I believe mercury was a popular "remedy" at the time) were more likely to kill you than make you well.
Dear Mark Olsen (or whoever does the thumbnail): some Healthcare Triage video thumbnails look kind of washed out. I am not critiquing your style, but the effect looks exactly the same as the effect UA-cam uses to show that you've already watched a video.
This caused me to miss a couple videos. You may want to change this, especially on this one and the Number Needed to Treat video.