Number needed to treat

So here's what I mean by that.

If you look at number needed to treat for all of these different interventions, what you typically find for them is that the NNT is around four or five or six or seven.

So what that means is that relative to placebo, which is a huge topic here that we haven't dug into yet, but I'm hoping to, each of these interventions, something like Cymbalta, which is an SNRI, it's sometimes prescribed for depression and anxiety disorder, but it's also prescribed for chronic pain, uh, including chronic musculoskeletal pain, fibromyalgia, all of that.

About four in 10 people on this ne- uh, hit 50% pain relief.

Big problem, one in six people also stop taking the drug because it gives them stomach distress that is so intense that they don't want to be on it anymore.

So this is not great for the medication side of the spectrum.

Problem is that for, like, CBT for chronic pain, as near as I can tell, NNT is about one in four.

So it's better, but it's not great, and a lot of those effects tend to go away over time.

The thesis was that the diagnosis of anatomic disease led to more statin use.

But when you look at Scott Hart, the delta in statin use was really small, and if it was due to statins, you'd have to posit an NNT of about 3 for the statins, which is hugely higher than what we know statins do.

So if you can explain why Scott Hart showed that CTCA was an event reducer, But CTCA in a higher risk group of patients in this trial did nothing.

I'm all ears.

The absolute risk reduction, the actual difference between 6.8% and 0.7% is 6.1 percentage points.

So walk us through the number needed to treat the NNT.

Right.

The number needed to treat is the most practical metric for a bedside clinician.

It tells you exactly how much effort you have to expend to get one positive result.

You just take 100 and divide it by the absolute risk reduction of 6.1.

which gives us an NNT of roughly 16.

Correct.

It is.

So we naturally wanna look at the number needed to treat, or NNT.

The clinical bottom line.

Right.

The data tells us you would need to use a Macintosh-style video laryngoscope on roughly 21.5 patients to achieve one additional first-pass success compared to just using a direct laryngoscope.

Okay.

For the hyperangulated blade, that NNT drops down to about 10.7.

Which honestly, in the realm of clinical trials, an NNT of roughly eleven to prevent a procedural failure is quite impressive.

It provides zero practical context.

This is why the concept of number needed to treat, or NNT, is the ultimate tool for a practicing nurse or physician.

NNT translates abstract probabilities into a tangible clinical reality.

Yeah.

The text highlights a trial treating pregnant women for gestational diabetes-

Okay.

The absolute risk difference between the treatment and control groups was point zero nine three.

To calculate the NNT, you divide one by that absolute risk difference.

It just truly is what it is, and you have to be able to understand that in order to have a conversation with your patients and have a shared decision-making conversation.

In order to calculate the a- NNT, you need to know the absolute difference.

The absolute difference is just subtracting the two numbers.

So let's do some practice.

Relative risk can sound amazing even when the benefit's small.

Read the article.

Calculate the NNT because shared decision-making means giving patients numbers they can see.

Who was studied? Inclusion and exclusion.