F. Perry Wilson, MD
@fperrywilson.bsky.social
Director, Clinical and Translational Research Accelerator @Yale. Columnist @medscape.
How Medicine Works and When It Doesn't in bookstores now!
How Medicine Works and When It Doesn't in bookstores now!
Full column here: buff.ly/kuENZTy (12/12)
January 12, 2026 at 10:50 PM
Full column here: buff.ly/kuENZTy (12/12)
Meanwhile, CMS administrator Mehmet Oz (that feels strange to write) says alcohol is a good "social lubricant" but not to "have it for breakfast." Not exactly the Dry January messaging I was hoping for. (11/12)
buff.ly/GTew7LQ
buff.ly/GTew7LQ
January 12, 2026 at 10:50 PM
Meanwhile, CMS administrator Mehmet Oz (that feels strange to write) says alcohol is a good "social lubricant" but not to "have it for breakfast." Not exactly the Dry January messaging I was hoping for. (11/12)
buff.ly/GTew7LQ
buff.ly/GTew7LQ
My take: GLP-1s clearly reduce alcohol intake. What I'm not sure about is how much better they are than existing drugs. No head-to-head trials yet. But for someone with AUD starting a GLP-1 for another reason? Icing on the cake. (10/12)
January 12, 2026 at 10:50 PM
My take: GLP-1s clearly reduce alcohol intake. What I'm not sure about is how much better they are than existing drugs. No head-to-head trials yet. But for someone with AUD starting a GLP-1 for another reason? Icing on the cake. (10/12)
A JAMA Psychiatry trial randomized 48 people with AUD to semaglutide vs placebo. In a lab setting, semaglutide users drank almost two full drinks less. In the real world, fewer heavy drinking days. Oh, and the smokers smoked less too. (9/12)
buff.ly/7tHORC9
buff.ly/7tHORC9
January 12, 2026 at 10:50 PM
A JAMA Psychiatry trial randomized 48 people with AUD to semaglutide vs placebo. In a lab setting, semaglutide users drank almost two full drinks less. In the real world, fewer heavy drinking days. Oh, and the smokers smoked less too. (9/12)
buff.ly/7tHORC9
buff.ly/7tHORC9
Of course, observational data could be confounded. Maybe naltrexone is given to people with worse pathology. What we need is a randomized trial. And we finally have one. (8/12)
January 12, 2026 at 10:50 PM
Of course, observational data could be confounded. Maybe naltrexone is given to people with worse pathology. What we need is a randomized trial. And we finally have one. (8/12)
Observational data backs this up. A Swedish cohort of 200,000+ people with AUD found those on semaglutide were 36% less likely to be hospitalized for alcohol-related issues. A bigger effect than the FDA-approved AUD drugs. (7/12)
January 12, 2026 at 10:50 PM
Observational data backs this up. A Swedish cohort of 200,000+ people with AUD found those on semaglutide were 36% less likely to be hospitalized for alcohol-related issues. A bigger effect than the FDA-approved AUD drugs. (7/12)
Enter the GLP-1s. From early trials, it was clear more was happening than just delayed gastric emptying. Anecdotal reports suggested people were consuming less... everything. Food, cigarettes, gambling, compulsive shopping, and yes, alcohol. (6/12)
January 12, 2026 at 10:50 PM
Enter the GLP-1s. From early trials, it was clear more was happening than just delayed gastric emptying. Anecdotal reports suggested people were consuming less... everything. Food, cigarettes, gambling, compulsive shopping, and yes, alcohol. (6/12)
There are a few other options, but overall the AUD treatment landscape is this: we have multiple drugs, but due to modest efficacy or tolerability issues, none of them are real game-changers. (5/12)
January 12, 2026 at 10:50 PM
There are a few other options, but overall the AUD treatment landscape is this: we have multiple drugs, but due to modest efficacy or tolerability issues, none of them are real game-changers. (5/12)
Naltrexone is the flip side. Antabuse makes you feel bad when you drink; naltrexone prevents you from feeling good. It's the most commonly prescribed AUD drug. More modest effects: ~20% reduction in heavy drinking, ~15% increase in abstinence. (4/12)
January 12, 2026 at 10:50 PM
Naltrexone is the flip side. Antabuse makes you feel bad when you drink; naltrexone prevents you from feeling good. It's the most commonly prescribed AUD drug. More modest effects: ~20% reduction in heavy drinking, ~15% increase in abstinence. (4/12)
Let's set the baseline. The oldest drug is Antabuse. It blocks alcohol breakdown, making you feel terrible if you drink. Under supervision: 80% reduction in heavy drinking. Without supervision? Almost no benefit. People skip doses so they can drink. (3/12)
January 12, 2026 at 10:50 PM
Let's set the baseline. The oldest drug is Antabuse. It blocks alcohol breakdown, making you feel terrible if you drink. Under supervision: 80% reduction in heavy drinking. Without supervision? Almost no benefit. People skip doses so they can drink. (3/12)
For some people, cutting back on alcohol requires pharmacologic help. We've had drugs for Alcohol Use Disorder since the 1950s. But over the past few years, a new player has emerged: GLP-1 receptor agonists. That's right. The weight loss drugs. Again. (2/12)
January 12, 2026 at 10:50 PM
For some people, cutting back on alcohol requires pharmacologic help. We've had drugs for Alcohol Use Disorder since the 1950s. But over the past few years, a new player has emerged: GLP-1 receptor agonists. That's right. The weight loss drugs. Again. (2/12)
We're in week two of Dry January and continuing our month of columns dedicated to alcohol, the substance Homer Simpson referred to as the cause of, and solution to, all of life's problems. (1/12)
January 12, 2026 at 10:50 PM
We're in week two of Dry January and continuing our month of columns dedicated to alcohol, the substance Homer Simpson referred to as the cause of, and solution to, all of life's problems. (1/12)
More in my @medscape column this week: buff.ly/EW3RDg6
Let me know if you're doing it! I'm in... so far.
#dryjanuary
(13/13)
Let me know if you're doing it! I'm in... so far.
#dryjanuary
(13/13)
January 6, 2026 at 9:22 PM
More in my @medscape column this week: buff.ly/EW3RDg6
Let me know if you're doing it! I'm in... so far.
#dryjanuary
(13/13)
Let me know if you're doing it! I'm in... so far.
#dryjanuary
(13/13)
Why might this work long-term? Self-efficacy.
Dry January proves you don't need alcohol to navigate social situations or handle stress.
There's real power in proving to yourself that you can do something.
(12/13)
Dry January proves you don't need alcohol to navigate social situations or handle stress.
There's real power in proving to yourself that you can do something.
(12/13)
January 6, 2026 at 9:22 PM
Why might this work long-term? Self-efficacy.
Dry January proves you don't need alcohol to navigate social situations or handle stress.
There's real power in proving to yourself that you can do something.
(12/13)
Dry January proves you don't need alcohol to navigate social situations or handle stress.
There's real power in proving to yourself that you can do something.
(12/13)
Nope. Actually the opposite.
Researchers checked on participants in August.
Drinking days per week: 4.3 → 3.3
Days drunk per month: 3.4 → 2.1
That's some sustained behavior change.
buff.ly/LQJEbvH
(11/13)
Researchers checked on participants in August.
Drinking days per week: 4.3 → 3.3
Days drunk per month: 3.4 → 2.1
That's some sustained behavior change.
buff.ly/LQJEbvH
(11/13)
January 6, 2026 at 9:22 PM
Nope. Actually the opposite.
Researchers checked on participants in August.
Drinking days per week: 4.3 → 3.3
Days drunk per month: 3.4 → 2.1
That's some sustained behavior change.
buff.ly/LQJEbvH
(11/13)
Researchers checked on participants in August.
Drinking days per week: 4.3 → 3.3
Days drunk per month: 3.4 → 2.1
That's some sustained behavior change.
buff.ly/LQJEbvH
(11/13)
But here's the big question: what happens in February?
Does everyone rebound and make up for lost time?
(10/13)
Does everyone rebound and make up for lost time?
(10/13)
January 6, 2026 at 9:22 PM
But here's the big question: what happens in February?
Does everyone rebound and make up for lost time?
(10/13)
Does everyone rebound and make up for lost time?
(10/13)
71% of Dry January participants report sleeping better and having more energy.
Honestly, I had trouble finding a physiologic metric that doesn't improve.
buff.ly/LQJEbvH
(9/13)
Honestly, I had trouble finding a physiologic metric that doesn't improve.
buff.ly/LQJEbvH
(9/13)
January 6, 2026 at 9:22 PM
71% of Dry January participants report sleeping better and having more energy.
Honestly, I had trouble finding a physiologic metric that doesn't improve.
buff.ly/LQJEbvH
(9/13)
Honestly, I had trouble finding a physiologic metric that doesn't improve.
buff.ly/LQJEbvH
(9/13)
Oh, and you'll probably lose about 3 lbs.
Alcohol has 7 calories per gram. More calorie-dense than carbs or protein, almost as bad as fat.
It's not a GLP-1, but it's not nothing either.
(8/13)
Alcohol has 7 calories per gram. More calorie-dense than carbs or protein, almost as bad as fat.
It's not a GLP-1, but it's not nothing either.
(8/13)
January 6, 2026 at 9:22 PM
Oh, and you'll probably lose about 3 lbs.
Alcohol has 7 calories per gram. More calorie-dense than carbs or protein, almost as bad as fat.
It's not a GLP-1, but it's not nothing either.
(8/13)
Alcohol has 7 calories per gram. More calorie-dense than carbs or protein, almost as bad as fat.
It's not a GLP-1, but it's not nothing either.
(8/13)
Other changes in just one month:
25% improvement in insulin resistance
7% drop in systolic blood pressure
42% reduction in VEGF
74% reduction in EGF
Those last two are biomarkers linked to cancer risk.
buff.ly/DRqoIvg
(7/13)
25% improvement in insulin resistance
7% drop in systolic blood pressure
42% reduction in VEGF
74% reduction in EGF
Those last two are biomarkers linked to cancer risk.
buff.ly/DRqoIvg
(7/13)
January 6, 2026 at 9:22 PM
Other changes in just one month:
25% improvement in insulin resistance
7% drop in systolic blood pressure
42% reduction in VEGF
74% reduction in EGF
Those last two are biomarkers linked to cancer risk.
buff.ly/DRqoIvg
(7/13)
25% improvement in insulin resistance
7% drop in systolic blood pressure
42% reduction in VEGF
74% reduction in EGF
Those last two are biomarkers linked to cancer risk.
buff.ly/DRqoIvg
(7/13)
The liver responds fast.
One study put 64 heavy drinkers through 4 weeks of abstinence. 80% saw improved liver stiffness, about a 15% reduction.
The liver just recovers quickly when you stop poisoning it.
buff.ly/JDRJFrl
(6/13)
One study put 64 heavy drinkers through 4 weeks of abstinence. 80% saw improved liver stiffness, about a 15% reduction.
The liver just recovers quickly when you stop poisoning it.
buff.ly/JDRJFrl
(6/13)
January 6, 2026 at 9:22 PM
The liver responds fast.
One study put 64 heavy drinkers through 4 weeks of abstinence. 80% saw improved liver stiffness, about a 15% reduction.
The liver just recovers quickly when you stop poisoning it.
buff.ly/JDRJFrl
(6/13)
One study put 64 heavy drinkers through 4 weeks of abstinence. 80% saw improved liver stiffness, about a 15% reduction.
The liver just recovers quickly when you stop poisoning it.
buff.ly/JDRJFrl
(6/13)
So assuming you actually do it, what happens to your body?
A lot, actually.
(5/13)
A lot, actually.
(5/13)
January 6, 2026 at 9:22 PM
So assuming you actually do it, what happens to your body?
A lot, actually.
(5/13)
A lot, actually.
(5/13)
Social media tells a different story: only 12.7% of tweets claim success while 66% admit failure.
But social media isn't real life.
buff.ly/rR40gwS
(4/13)
But social media isn't real life.
buff.ly/rR40gwS
(4/13)
January 6, 2026 at 9:22 PM
Social media tells a different story: only 12.7% of tweets claim success while 66% admit failure.
But social media isn't real life.
buff.ly/rR40gwS
(4/13)
But social media isn't real life.
buff.ly/rR40gwS
(4/13)
Success rates are actually pretty solid.
About 2/3 of UK participants and 80% of Dutch participants (they call it "IkPas") make it through without drinking.
buff.ly/aTqguFM
(3/13)
About 2/3 of UK participants and 80% of Dutch participants (they call it "IkPas") make it through without drinking.
buff.ly/aTqguFM
(3/13)
January 6, 2026 at 9:22 PM
Success rates are actually pretty solid.
About 2/3 of UK participants and 80% of Dutch participants (they call it "IkPas") make it through without drinking.
buff.ly/aTqguFM
(3/13)
About 2/3 of UK participants and 80% of Dutch participants (they call it "IkPas") make it through without drinking.
buff.ly/aTqguFM
(3/13)
Let's get something out of the way first.
People who do Dry January are a very particular sort: more likely female, higher income, college-educated... and they drink more than average.
Few would call themselves alcoholics. Most definitely had a December.
(2/13)
People who do Dry January are a very particular sort: more likely female, higher income, college-educated... and they drink more than average.
Few would call themselves alcoholics. Most definitely had a December.
(2/13)
January 6, 2026 at 9:22 PM
Let's get something out of the way first.
People who do Dry January are a very particular sort: more likely female, higher income, college-educated... and they drink more than average.
Few would call themselves alcoholics. Most definitely had a December.
(2/13)
People who do Dry January are a very particular sort: more likely female, higher income, college-educated... and they drink more than average.
Few would call themselves alcoholics. Most definitely had a December.
(2/13)
So... for reasons both personal and professional, I decided to dig into the data behind "Dry January". To be honest, I'm impressed. The liver is a remarkable organ. Brief thread on major findings:
🧵
(1/13)
🧵
(1/13)
January 6, 2026 at 9:22 PM
So... for reasons both personal and professional, I decided to dig into the data behind "Dry January". To be honest, I'm impressed. The liver is a remarkable organ. Brief thread on major findings:
🧵
(1/13)
🧵
(1/13)