RSS
_-The-Breakthrough-Antipsychotic-That-Could-Change-Everything.webp?t=1729528747 "KarXT (Cobenfy)_ The Breakthrough Antipsychotic That Could Change Everything.jpg")
Nine studies have tried to find out whether ADHD can start in adulthood, we dig into the data and find some answers to the controversy.
Publication Date: 08/03/2026
Duration: 13 minutes, 07 seconds
Transcript:
KELLIE NEWSOME: Can ADHD start in adulthood? We look at nine studies that tried to find an answer.
CHRIS AIKEN: Welcome to the Carlat Psychiatry Podcast, keeping psychiatry honest since 2003. I’m Chris Aiken, the editor in chief of the Carlat Report.
KELLIE NEWSOME: And I’m Kellie Newsome, a psychiatric NP and a dedicated reader of every issue. Last week, we left off in the mid-2000s. Psychiatry had reached a consensus: ADHD continues into adulthood for many who have it, and this adult ADHD continues to respond to stimulants. Today, we'll look at research that pushed the boundaries further; it claims ADHD can start in adulthood. But first, a preview of the CME quiz for this episode. You'll find the link in the show notes, and the answer in the research update at the end.
1. Which of these medications is classified as an azapirone?
A. Buspirone
B. Trazodone
C. Pindolol
D. Vilazodone
A man in his mid-thirties sits down in your office. He had no childhood diagnosis of ADHD; no teacher notes about disrupting class or daydreaming. He got through college, and his parents thought he was a good kid. But now he can't finish projects at work, loses his keys twice a month, and wonders if he has ADHD. Nine studies have looked at whether ADHD can show up for the first time in adulthood, and the answer is complicated.
CHRIS AIKEN: The controversy traces back to 1972, when researchers in New Zealand enrolled just over a thousand newborns into a cohort study. They followed them with regular medical and psychological workups, at first every few years, then less frequent after childhood. The cohort has one of the highest retention rates of any study like it. In 2015, it produced a startling finding. At that time, they were in their late 30s, and 1 in 30 of the New Zealanders had ADHD. But nearly all of these adult ADHD cases - 90% of them – did not have ADHD as children, not even on the careful cognitive testing the researchers ran in early childhood.
KELLIE NEWSOME: This paper, by Terrie Moffitt and colleagues, was quickly followed by three other papers claiming to identify a late-onset phenotype of ADHD, usually starting in the teens. But a closer read leaves us skeptical. Many of these teens had mild ADHD symptoms as children that didn't cross the threshold for diagnosis, somewhere between 30% and 75% of them. Others had childhood disorders that overlap with ADHD, like conduct disorder or oppositional defiant disorder. So was this teenage onset, or were their problems misclassified in childhood?
CHRIS AIKEN: When we take a closer look at the New Zealand paper, the headline finding becomes more mundane. There were only 27 cases of adult-onset ADHD. Ten of them had conduct disorder as children, which could have been an externalizing sign of ADHD. The rest had adult problems that are known to cause ADHD-like symptoms: 15 had substance use disorders, and 15 had major depressive disorder. Substance use harms the brain in ways that impair cognition, and cognitive symptoms are common in mood disorders even after the depression lifts, but the authors didn't rule these out when they reached for the ADHD diagnosis. The only adult disorder they excluded before diagnosing ADHD was schizophrenia.
KELLIE NEWSOME: For listeners keeping a close count, there were 27 adult-onset cases, but Dr. Aiken counted 40 instances where other causes might better explain the ADHD. Yes, 40 is more than 27, and that’s because some subjects had multiple causes. The authors seemed intent on describing a new diagnostic entity, so these other causes got buried in a table. But they may be the most important part of the paper. Two other studies on late-onset ADHD spelled it out directly: 93% to 95% of cases were better explained by something else, mainly sleep disorders, substance use disorders, or another psychiatric condition.
CHRIS AIKEN: That is backed by a meta-analysis that Lee Taylor's group at Syracuse University ran in 2022. They found nine studies on late-onset ADHD. The low-quality studies concluded it was real, including the New Zealand one. The high-quality studies, the ones that carefully checked for childhood evidence or other causes of adult symptoms, found late-onset ADHD didn't exist. Dr. Taylor's conclusion: people can present with ADHD symptoms for the first time in adulthood, but that doesn't mean they have adult-onset ADHD. Most cases are better explained by one of these three reasons:
A man in his mid-thirties sits down in your office. He had no childhood diagnosis of ADHD; no teacher notes about disrupting class or daydreaming. He got through college, and his parents thought he was a good kid. But now he can't finish projects at work, loses his keys twice a month, and wonders if he has ADHD. Nine studies have looked at whether ADHD can show up for the first time in adulthood, and the answer is complicated.
CHRIS AIKEN: The controversy traces back to 1972, when researchers in New Zealand enrolled just over a thousand newborns into a cohort study. They followed them with regular medical and psychological workups, at first every few years, then less frequent after childhood. The cohort has one of the highest retention rates of any study like it. In 2015, it produced a startling finding. At that time, they were in their late 30s, and 1 in 30 of the New Zealanders had ADHD. But nearly all of these adult ADHD cases - 90% of them – did not have ADHD as children, not even on the careful cognitive testing the researchers ran in early childhood.
KELLIE NEWSOME: This paper, by Terrie Moffitt and colleagues, was quickly followed by three other papers claiming to identify a late-onset phenotype of ADHD, usually starting in the teens. But a closer read leaves us skeptical. Many of these teens had mild ADHD symptoms as children that didn't cross the threshold for diagnosis, somewhere between 30% and 75% of them. Others had childhood disorders that overlap with ADHD, like conduct disorder or oppositional defiant disorder. So was this teenage onset, or were their problems misclassified in childhood?
CHRIS AIKEN: When we take a closer look at the New Zealand paper, the headline finding becomes more mundane. There were only 27 cases of adult-onset ADHD. Ten of them had conduct disorder as children, which could have been an externalizing sign of ADHD. The rest had adult problems that are known to cause ADHD-like symptoms: 15 had substance use disorders, and 15 had major depressive disorder. Substance use harms the brain in ways that impair cognition, and cognitive symptoms are common in mood disorders even after the depression lifts, but the authors didn't rule these out when they reached for the ADHD diagnosis. The only adult disorder they excluded before diagnosing ADHD was schizophrenia.
KELLIE NEWSOME: For listeners keeping a close count, there were 27 adult-onset cases, but Dr. Aiken counted 40 instances where other causes might better explain the ADHD. Yes, 40 is more than 27, and that’s because some subjects had multiple causes. The authors seemed intent on describing a new diagnostic entity, so these other causes got buried in a table. But they may be the most important part of the paper. Two other studies on late-onset ADHD spelled it out directly: 93% to 95% of cases were better explained by something else, mainly sleep disorders, substance use disorders, or another psychiatric condition.
CHRIS AIKEN: That is backed by a meta-analysis that Lee Taylor's group at Syracuse University ran in 2022. They found nine studies on late-onset ADHD. The low-quality studies concluded it was real, including the New Zealand one. The high-quality studies, the ones that carefully checked for childhood evidence or other causes of adult symptoms, found late-onset ADHD didn't exist. Dr. Taylor's conclusion: people can present with ADHD symptoms for the first time in adulthood, but that doesn't mean they have adult-onset ADHD. Most cases are better explained by one of these three reasons:
- Childhood ADHD that got overlooked or misdiagnosed.
- Childhood ADHD that never reached clinical attention because the child had strong support, maybe too strong, like a parent doing the kid's homework.
- An adult cause of cognitive symptoms that gets mistaken for ADHD.
I'd add a fourth reason to keep in mind when interpreting this research: these are community samples, not treatment-seeking adults, and community samples run higher rates of false positives.
KELLIE NEWSOME: Back to our patient. If we stick to the evidence, we have two options. Either he had ADHD symptoms as a child that were never addressed, or an adult cause is at work that we haven't found yet. The list of adult causes is long, but one of them is a total exclusion: schizophrenia. Cognitive problems are universal in that population, and the DSM says you can't diagnose ADHD if the symptoms occur during schizophrenia or another psychotic disorder. Part of the reason is that amphetamines serve as the animal model for psychosis. Researchers tested stimulants in schizophrenia. They didn't improve cognition, and they did worsen psychotic symptoms at alarming rates, in 30% of patients who were in remission from psychosis and in 50% of patients who were actively psychotic.
CHRIS AIKEN: Outside of schizophrenia, the DSM allows an ADHD diagnosis alongside other disorders, but favors skipping it when the symptoms are "better explained" by bipolar disorder, depression, anxiety, dissociation, a personality disorder, or substance use. Next week, we'll look at what "better explained by" means in practice, and why it separates novice diagnosticians from experts.
KELLIE NEWSOME: Last December, a new antidepressant hit the pharmacy shelves: gepirone, branded as Exxua. It's a cousin of buspirone; both meds share an azapirone structure, and both act as agonists at the serotonin 1A receptor. Gepirone had a rocky road to approval. The FDA rejected it four times, in 1999, 2002, 2004, and 2007, before finally clearing it for major depression in 2023. Today's research update explains why.
CHRIS AIKEN: It's a meta-analysis in JAMA Psychiatry by Erick Turner and colleagues. His team pulled every FDA document on the drug: clinical reviews, correspondence, the 2015 Advisory Committee materials, to see how it finally got through. In the stack were 13 randomized trials. Most were never published. Of the 12 short-term trials, only 2 beat placebo. The other 10 didn't, and in 3 of those, gepirone did worse than an SSRI: fluoxetine or paroxetine. The one maintenance trial was negative, too.
CHRIS AIKEN: When all 12 short-term trials are pulled together, and the drug-placebo difference isn't statistically significant, and it isn’t clinically meaningful even if it was real: just under half a point on the Hamilton depression scale. The sponsor argued that the negative trials lacked "assay sensitivity," meaning they couldn't detect a true difference on the drug. But some of those same trials detected a difference between SSRIs and gepirone, so the industry explanation doesn’t fully hold up.
KELLIE NEWSOME: The FDA Advisory Committee voted 9 to 4 against approval. FDA leadership approved it anyway, and the product label only mentions the 2 positive trials.
CHRIS AIKEN: Gepirone's appeal is that it doesn’t cause sexual side effects, but we have other antidepressants that avoid them, like bupropion, mirtazapine, and possibly vortioxetine. And gepirone comes with a unique obstacle: the FDA label requires a baseline ECG to check for QTc prolongation. Its cousin, buspirone, carries no such risk, and this azapirone has better evidence than gepirone as a first-line option for major depression, particularly anxious depression. Buspirone never got approved for depression because the FDA had a more conservative bar back then, so no one uses buspirone first-line for depression. Instead, we use it as an augmentation strategy when antidepressants don’t work, and there’s little support for using it in these refractory cases: buspirone has largely failed as an augmentation strategy.
KELLIE NEWSOME: Long ago, medications earned their approval through the American Medical Association, until the AMA lost the public trust when they started caving to commercial pressures: The AMA earned advertising money for every drug it approved. Because of that debacle, the FDA took over approvals in 1963, and the first few decades were something of a golden era. Since then, approvals have grown more industry-friendly, driven by a revolving door where FDA employees leave the agency for lucrative jobs at the companies they used to regulate. This isn't a partisan story. Gepirone was approved under Biden, and Trump's appointees have lowered the bar further, requiring only one positive trial for approval. The bottom line: if you're looking for an antidepressant without sexual side effects, better options exist. Start with bupropion, mirtazapine, and possibly vortioxetine.
KELLIE NEWSOME: Back to our patient. If we stick to the evidence, we have two options. Either he had ADHD symptoms as a child that were never addressed, or an adult cause is at work that we haven't found yet. The list of adult causes is long, but one of them is a total exclusion: schizophrenia. Cognitive problems are universal in that population, and the DSM says you can't diagnose ADHD if the symptoms occur during schizophrenia or another psychotic disorder. Part of the reason is that amphetamines serve as the animal model for psychosis. Researchers tested stimulants in schizophrenia. They didn't improve cognition, and they did worsen psychotic symptoms at alarming rates, in 30% of patients who were in remission from psychosis and in 50% of patients who were actively psychotic.
CHRIS AIKEN: Outside of schizophrenia, the DSM allows an ADHD diagnosis alongside other disorders, but favors skipping it when the symptoms are "better explained" by bipolar disorder, depression, anxiety, dissociation, a personality disorder, or substance use. Next week, we'll look at what "better explained by" means in practice, and why it separates novice diagnosticians from experts.
KELLIE NEWSOME: Last December, a new antidepressant hit the pharmacy shelves: gepirone, branded as Exxua. It's a cousin of buspirone; both meds share an azapirone structure, and both act as agonists at the serotonin 1A receptor. Gepirone had a rocky road to approval. The FDA rejected it four times, in 1999, 2002, 2004, and 2007, before finally clearing it for major depression in 2023. Today's research update explains why.
CHRIS AIKEN: It's a meta-analysis in JAMA Psychiatry by Erick Turner and colleagues. His team pulled every FDA document on the drug: clinical reviews, correspondence, the 2015 Advisory Committee materials, to see how it finally got through. In the stack were 13 randomized trials. Most were never published. Of the 12 short-term trials, only 2 beat placebo. The other 10 didn't, and in 3 of those, gepirone did worse than an SSRI: fluoxetine or paroxetine. The one maintenance trial was negative, too.
CHRIS AIKEN: When all 12 short-term trials are pulled together, and the drug-placebo difference isn't statistically significant, and it isn’t clinically meaningful even if it was real: just under half a point on the Hamilton depression scale. The sponsor argued that the negative trials lacked "assay sensitivity," meaning they couldn't detect a true difference on the drug. But some of those same trials detected a difference between SSRIs and gepirone, so the industry explanation doesn’t fully hold up.
KELLIE NEWSOME: The FDA Advisory Committee voted 9 to 4 against approval. FDA leadership approved it anyway, and the product label only mentions the 2 positive trials.
CHRIS AIKEN: Gepirone's appeal is that it doesn’t cause sexual side effects, but we have other antidepressants that avoid them, like bupropion, mirtazapine, and possibly vortioxetine. And gepirone comes with a unique obstacle: the FDA label requires a baseline ECG to check for QTc prolongation. Its cousin, buspirone, carries no such risk, and this azapirone has better evidence than gepirone as a first-line option for major depression, particularly anxious depression. Buspirone never got approved for depression because the FDA had a more conservative bar back then, so no one uses buspirone first-line for depression. Instead, we use it as an augmentation strategy when antidepressants don’t work, and there’s little support for using it in these refractory cases: buspirone has largely failed as an augmentation strategy.
KELLIE NEWSOME: Long ago, medications earned their approval through the American Medical Association, until the AMA lost the public trust when they started caving to commercial pressures: The AMA earned advertising money for every drug it approved. Because of that debacle, the FDA took over approvals in 1963, and the first few decades were something of a golden era. Since then, approvals have grown more industry-friendly, driven by a revolving door where FDA employees leave the agency for lucrative jobs at the companies they used to regulate. This isn't a partisan story. Gepirone was approved under Biden, and Trump's appointees have lowered the bar further, requiring only one positive trial for approval. The bottom line: if you're looking for an antidepressant without sexual side effects, better options exist. Start with bupropion, mirtazapine, and possibly vortioxetine.
If you’d like to learn more about ADHD, I recommend Dr. Greg Malzberg’s audio course at the Carlat Report: Adult ADHD: Comprehensive Diagnosis and Treatment. Find it at theCarlatReport.com, click STORE then MULTIMEDIA.
The Carlat CME Institute is accredited by the ACCME to provide continuing medical education for physicians. Carlat CME Institute maintains responsibility for this program and its content. Carlat CME Institute designates this enduring material educational activity for a maximum of one quarter (.25) AMA PRA Category 1 CreditsTM. Physicians or psychologists should claim credit commensurate only with the extent of their participation in the activity.
