Using a large database of (deidentified) medical records, the researchers compared the outcomes of people taking a GLP-1 for their type 2 diabetes to people taking other common diabetes drugs between 2017 and 2025. GLP-1 users were significantly less likely to be diagnosed with TB for up to a five-year span, they found.
Both of these studies are observational and retrospective, meaning they can only show a correlation between GLP-1 use and reduced (or at least less severe) infections, not prove a direct cause-and-effect relationship. At the same time, these are the only latest pieces of evidence pointing to a genuine germ-busting benefit from GLP-1 drugs.
I wonder if the likelihood that people taking a GLP-1 are probably better off financially, have a health care provider willing to spend on the drugs (and therefore probably a better medical system) or a combination of these and other traits are the real reason there are fewer infections.
They use a method called propensity score matching to try their best to match patients on both sides using a simple linear model with various features that try to ensure that only pairs of closely matched patient histories are compared.
Unfortunately this is rarely clean. Its also easy to make mistakes. Sometimes two arms are fundamentally incomparable. The quality and rigor of the comparison is often determined by a lot of extra checks and validations, and different journals demand different levels of rigor. I need to read it carefully to judge if this is good or not.
It looks like both do standard individual covariate checks for post-match balance, with SMDs. I'm surprised they haven't assessed balance for at least pairwise interactions, too -- we should be balancing out joint risk factors too, no?
I haven't worked on these designs, but I remember the methodologist that taught me this in grad school giving us a lecture about this.
EDIT: the BMJ article (laudably) provides access to the analyis code, although I won't have time to review it:
They already do acknowledge socioeconomic (and other confounding factors) in the analysis.
The primary analysis they perform is a ‘Propensity Score Match’ which is a technique used specifically to address for confounders in observational studies, and they do report balanced cohorts.
Still they write in their discussion “Although we adjus-
ted for several available proxies of socioeconomic and lifestyle status,
direct measures of income, insurance coverage, or out-of-pocket
payment were not available in the TriNetX database. Residual confounding related to unmeasured socioeconomic factors, therefore,
cannot be excluded“
Given the size of the dataset, the effect size, significance and sensitivity testing they did I think it’s very strong evidence for GLP1s causing this and it would be very very surprising to me to see the effect disappear even if they had perfect socioeconomic data.
With so many signs of improvement, even if some of them happens to be just flukes, from so many different areas, it feels like a penicillin moment. And i suppose penicillin also was initially mostly available to the ones with better socio-economy.
I’m sure you can just look up the studies, but note that GLP1s are widely prescribed to people without weight issues —— weight loss itself was originally an off-target effect. I have friends who run marathons who are on semaglutide.
This video by a medical doctor cites many studies showing that GLP-1 health benefits go beyond weight loss (see video description for link to papers): https://www.youtube.com/watch?v=yKPaVhpomks
The benefit is probably from the removal of fat, not a direct antibacterial/antiviral effect. Fat plays a complex immunoregulatory role in human physiology: it down-regulates some pathways, while up-regulating others (notoriously, the production of IL6 is carried out, in part, by adipocytes). The overall effect of fat on the immune system, however, is negative: it tends to increase the chances of rheumatological disorders, cancers, and many other diseases. Alternatively, the effect may be due to some sociological factor that their analysis failed to account for.
Nobody's denying this, but the interesting story about GLP-1s is that after you control for fat loss (for instance, by taking cohorts of patients that aren't losing weight) you still get evidence of these off-target effects.
That's true, and it's also one of those factoids that everybody already knows but certain people can't help themselves from bringing up every time it's even tenuously relevant.
The fact that it was the top comment only shows that people agreed or liked it, not that it was actually valuable, accurate, or useful. Popularity and value aren't the same thing, correlation is not causation, etc, etc...
Most observational studies do in fact account for the first thing a random HN poster can come up with 5 seconds after reading the title. So there is very little value in such a comment.
This was published in Nature, I'm pretty sure the reviewers understand and account for that. HN and reasonably educated people always think this is some huge gotcha, second only to "CORRELATION != CAUSATION!!!!"
Nature isn't an arbiter of truth, it's a publisher of papers. It's also meant to be read critically by people who see a paper like this and and think "interesting, but this is just an observational study, I'd like to see more research, maybe something with more controls before I believe this".
Say what you want about the "correlation doesn't imply causation", but many people do not understand why that is. As a shorthand for "hey, you don't know the statistics behind this, so don't take the pretty picture at face value", it's pretty good.
I would imagine people going out of their way to take GLP-1 might also lose weight, be more active, and be also likely to eat better. Which I imagine would have a large effect on "germ busting"
Comments
Using a large database of (deidentified) medical records, the researchers compared the outcomes of people taking a GLP-1 for their type 2 diabetes to people taking other common diabetes drugs between 2017 and 2025. GLP-1 users were significantly less likely to be diagnosed with TB for up to a five-year span, they found.
Both of these studies are observational and retrospective, meaning they can only show a correlation between GLP-1 use and reduced (or at least less severe) infections, not prove a direct cause-and-effect relationship. At the same time, these are the only latest pieces of evidence pointing to a genuine germ-busting benefit from GLP-1 drugs.
I wonder if the likelihood that people taking a GLP-1 are probably better off financially, have a health care provider willing to spend on the drugs (and therefore probably a better medical system) or a combination of these and other traits are the real reason there are fewer infections.
They use a method called propensity score matching to try their best to match patients on both sides using a simple linear model with various features that try to ensure that only pairs of closely matched patient histories are compared.
Unfortunately this is rarely clean. Its also easy to make mistakes. Sometimes two arms are fundamentally incomparable. The quality and rigor of the comparison is often determined by a lot of extra checks and validations, and different journals demand different levels of rigor. I need to read it carefully to judge if this is good or not.
It looks like both do standard individual covariate checks for post-match balance, with SMDs. I'm surprised they haven't assessed balance for at least pairwise interactions, too -- we should be balancing out joint risk factors too, no?
I haven't worked on these designs, but I remember the methodologist that taught me this in grad school giving us a lecture about this.
EDIT: the BMJ article (laudably) provides access to the analyis code, although I won't have time to review it:
github.com/nilskruger/Tirzepatide-and-the-Risk-of-Atherosclerotic-Cardiovascular-Events
They already do acknowledge socioeconomic (and other confounding factors) in the analysis. The primary analysis they perform is a ‘Propensity Score Match’ which is a technique used specifically to address for confounders in observational studies, and they do report balanced cohorts. Still they write in their discussion “Although we adjus- ted for several available proxies of socioeconomic and lifestyle status, direct measures of income, insurance coverage, or out-of-pocket payment were not available in the TriNetX database. Residual confounding related to unmeasured socioeconomic factors, therefore, cannot be excluded“
Given the size of the dataset, the effect size, significance and sensitivity testing they did I think it’s very strong evidence for GLP1s causing this and it would be very very surprising to me to see the effect disappear even if they had perfect socioeconomic data.
With so many signs of improvement, even if some of them happens to be just flukes, from so many different areas, it feels like a penicillin moment. And i suppose penicillin also was initially mostly available to the ones with better socio-economy.
Or having a lower % of body fat (within healthy limits) is the factor improving a better immune response?
Generally studies showing off-target effects with GLP1s are at least attempting to control for this.
I’d be interested to see the data for that claim, since I would imagine a strong correlation between glp use and lower body fat.
I’m sure you can just look up the studies, but note that GLP1s are widely prescribed to people without weight issues —— weight loss itself was originally an off-target effect. I have friends who run marathons who are on semaglutide.
Because they need to medically or it helps athletically ?
Because they need it medically.
This video by a medical doctor cites many studies showing that GLP-1 health benefits go beyond weight loss (see video description for link to papers): https://www.youtube.com/watch?v=yKPaVhpomks
It may be that those without repeated toxic caloric exposure allows the immune system to focus on other aspects.
Is that a thing?
They are also anti-inflammatory, so it could be related to less systemic inflammation.
The benefit is probably from the removal of fat, not a direct antibacterial/antiviral effect. Fat plays a complex immunoregulatory role in human physiology: it down-regulates some pathways, while up-regulating others (notoriously, the production of IL6 is carried out, in part, by adipocytes). The overall effect of fat on the immune system, however, is negative: it tends to increase the chances of rheumatological disorders, cancers, and many other diseases. Alternatively, the effect may be due to some sociological factor that their analysis failed to account for.
(I am not a medical doctor)
Many of the health benefits, including cardiovascular and kidney health, have been shown to go beyond or be unrelated to changes in body weight.
https://www.youtube.com/watch?v=yKPaVhpomks
Lower visceral fat as a percentage of body weight may bring benefits that are greater than proportional to total body mass lost.
Nobody's denying this, but the interesting story about GLP-1s is that after you control for fat loss (for instance, by taking cohorts of patients that aren't losing weight) you still get evidence of these off-target effects.
How much of “fat” also includes biofilmed infection stifling your electrical system and indeed signaling your immune system to not work as well?
(You can look this up regarding biofilms, I just did today.)
Then you must have some links available
Wow, bet they never thought of that. If only the researchers had thought to ask HN first.
Would you prefer that people accept claims uncritically? It’s a valid critique of the results.
If the commenter read the study and found that they did not, in fact, account for that then it would be valid to point it out here.
Otherwise, it's just a waste of our time.
By its fundamental design, an observational study cannot account for everything. That is the critique and it is a valid one.
That's true, and it's also one of those factoids that everybody already knows but certain people can't help themselves from bringing up every time it's even tenuously relevant.
It might be common sense if you work with statistics, but most people don’t, and the only way they would know these things is if someone says them.
Regardless, it was the top comment so people clearly think it was valuable enough to vote for.
The fact that it was the top comment only shows that people agreed or liked it, not that it was actually valuable, accurate, or useful. Popularity and value aren't the same thing, correlation is not causation, etc, etc...
Most observational studies do in fact account for the first thing a random HN poster can come up with 5 seconds after reading the title. So there is very little value in such a comment.
This was published in Nature, I'm pretty sure the reviewers understand and account for that. HN and reasonably educated people always think this is some huge gotcha, second only to "CORRELATION != CAUSATION!!!!"
Nature isn't an arbiter of truth, it's a publisher of papers. It's also meant to be read critically by people who see a paper like this and and think "interesting, but this is just an observational study, I'd like to see more research, maybe something with more controls before I believe this".
Say what you want about the "correlation doesn't imply causation", but many people do not understand why that is. As a shorthand for "hey, you don't know the statistics behind this, so don't take the pretty picture at face value", it's pretty good.
I would imagine people going out of their way to take GLP-1 might also lose weight, be more active, and be also likely to eat better. Which I imagine would have a large effect on "germ busting"