New Research Calls for Stronger Standards When Using Large Language Models and Other AI Tools for Psychological Assessment

Researchers find AI models can score well on accuracy measures while missing key signs of mental health concerns; they introduce Text Psychometrics, a new framework for testing these tools more rigorously.

New York, NY — Today the Child Mind Institute, in collaboration with Harvard University and the Massachusetts Institute of Technology, released a new research paper that found that AI tools cannot yet provide valid psychological assessments and require a more rigorous testing process. The paper reaffirms that mental health professionals are still best suited to assess and address the needs of people seeking mental health resources. Additionally, researchers introduced a framework for evaluating whether AI systems can measure mental health in a scientifically sound, reliable, and fair way for future development. The paper is in press at PLOS Digital Health.

Every day, people describe their inner lives in text — including crisis hotline messages, social media posts, therapy transcripts, journal entries, and more. As AI becomes increasingly capable of analyzing language, researchers are beginning to use these tools as a new way to conduct psychological assessments.

However, unlike traditional questionnaires used by psychologists, most AI language tools are insufficiently tested. Experts have spent a century building rigorous statistical standards for questionnaires within the field of psychometrics. Does a test measure what it claims? Does it give consistent answers? And does it work equally well for everyone? AI language tools are rarely held to that same standard.

In this paper, researchers introduce the concept of Text Psychometrics, a framework for evaluating whether AI and natural language processing (NLP) tools can measure psychological constructs in text in a scientifically sound way. Rather than asking only whether a model can match human judgments, the co-authors argue that researchers should also examine whether these tools capture the full range of how people express symptoms, remain reliable across different psychological situations, and work fairly across different patient populations.

“This research reaffirms that AI cannot replace mental health professionals to adequately assess and meet the needs of patients and people seeking mental health resources. An AI model may appear accurate while still missing important ways people express psychological symptoms,” says Daniel Low, PhD, a research scientist at the Child Mind Institute and the study’s lead author. “AI can help researchers and clinicians analyze language at scale and find patterns that would otherwise be very challenging, but these AI models do not currently understand the full approach that experienced clinicians take when making assessments. With our new framework, our goal is not to replace human expertise, but to build tools that are helpful while being more transparent, trustworthy, and scientifically grounded to meet the needs of patients.”

To develop the paper and framework, researchers analyzed thousands of crisis counseling conversations and public social media posts. They compared several approaches for identifying mental health concerns, including large language models and other language-analysis methods. The team found that some AI systems performed well, on average, with commonly used accuracy measures. But they still failed to recognize many clear and important expressions of mental health concerns because they were less common and therefore did not affect the general accuracy. In other words, a model could appear successful while overlooking language that a clinician or researcher would consider highly relevant.

To address this issue, the researchers developed a new framework to evaluate language models that includes different types of validation, including what they call “content validity test sets.” Rather than asking only whether a model agrees with human judgments, the approach examines whether the model recognizes the full range of ways people express a psychological symptom. For example, to screen for eating disorders, a model would need to not only capture mentions of anorexia and bulimia but also purging, chewing and spitting, and other symptoms. 

“A mental health model should not be considered trustworthy simply because it performs well on a single metric,” Dr. Low explains. “Researchers should also ask whether it captures the most important expressions of a symptom, performs consistently, protects privacy, and works fairly across different groups of people, among many other desirable properties we describe.”

In a second study, the team examined whether language from the beginning of a crisis counseling conversation could predict who would later be sent an emergency service intervention towards the end of the conversation. The researchers found that language patterns associated with suicide risk could provide useful signals, while also revealing which factors out of about 50 were most strongly linked to acute risk. Mentions of lethal means for suicide, substance use and overdose, and having the texter write more than the counselor were the most predictive variables for detecting the need for an emergency service intervention.

Human clinicians consider far more than written words when evaluating a person’s mental health, including clinical history, life circumstances, nonverbal cues — if available — and professional judgment. The findings suggest that AI systems are not yet at this level.

The researchers hope Text Psychometrics will help consolidate efforts to improve how AI-based mental health assessments are developed and evaluated. By applying the same rigorous standards used for traditional psychological tests, they believe language-based AI tools can become more reliable, interpretable, and useful for research, clinical care, and public health.

The study was authored by Daniel Low, PhD, Patrick Mair, PhD, and Matthew K. Nock, PhD, of Harvard University, and Satrajit S. Ghosh, PhD, of the Massachusetts Institute of Technology and Harvard Medical School.

The research was supported by training grants from the National Institute of Mental Health and the National Institute on Deafness and Other Communication Disorders. It also received support from the RallyPoint Fellowship, Amelia Peabody Professional Development Award, Chet and Will Griswold Fund, Tommy Fuss Collaborative STAR Fund, and Lann and Christopher Woehrle Psychiatric Research Fund. The funders had no role in the study’s design, data collection and analysis, publication decision, or manuscript preparation.

The authors thank Crisis Text Line for providing access to de-identified data used in this research. Crisis Text Line did not influence the study design, analysis, interpretation of findings, or conclusions.

Learn More About Our Work

Learn more about the AI, Risk, and Contemplative Science (ARC) Lab and the Center for Data Analytics, Innovation, and Rigor (DAIR) at the Child Mind Institute.

Crisis Resources

If you or someone you know is struggling or in crisis:

Call or text 988 anytime in the United States to reach the Suicide & Crisis Lifeline

To reach Crisis Text Line, text HOME or HOLA to 741741 to connect with a live, trained volunteer crisis counselor.

About the Child Mind Institute

The Child Mind Institute is dedicated to transforming the lives of children and families struggling with mental health and learning disorders by giving them the help they need. We’ve become the leading independent nonprofit in children’s mental health by providing gold-standard, evidence-based care, delivering educational resources to millions of families each year, training educators in underserved communities, and developing tomorrow’s breakthrough treatments.

The post New Research Calls for Stronger Standards When Using Large Language Models and Other AI Tools for Psychological Assessment appeared first on Child Mind Institute.

Eating disorders are on the rise. GLP-1s could make things worse.

GLP-1 weight-loss drugs are one of the biggest biotech breakthroughs of the last decade. But to some clinical psychologists, they can represent something more sinister—a potential trigger for eating disorders.

A new report suggests that the global incidence of anorexia nervosa and bulimia nervosa shot up between 1990 and 2023—by a whopping 71%. By 2023, there were an estimated 18.1 million people living with one of these two disorders.

What’s worse is that the number is likely to be much higher than that once we factor in other eating disorders. And, disturbingly, those in the field are already starting to see GLP-1 weight-loss drugs exacerbating the issue.

Cheri Levinson is a clinical psychologist at the University of Louisville in Kentucky who has specialized in eating disorders for almost 20 years. She says she’s now seeing more patients with more severe symptoms than at any other time in her career, and that many of them are misusing GLP-1s. “This is the worst I’ve seen things,” she says.

Eating disorders are mental health conditions in which people’s eating behaviors impact their health. They include anorexia nervosa (defined by a fear of weight gain), bulimia nervosa (in which people binge and purge), binge eating disorder, and others.

It’s thought that some people have a genetic predisposition to certain eating disorders, but that some external factor is what triggers their onset. That might be stress, or social pressure to be skinny. Often, simply restricting food intake can trigger an eating disorder, says Levinson.

Eating disorders—along with the pressure to look thin—were already prevalent in the 1990s (the heroin chic era springs to mind). To find out how things might have changed since then, Ruhai Bai at the Children’s Hospital of Nanjing Medical University in China and his colleagues pulled together data from studies and surveys conducted on populations around the world. The team used statistical models to fill in any data gaps.

That research allowed the researchers to estimate the annual incidence of two eating disorders—anorexia nervosa and bulimia nervosa—for each of 204 countries and territories between 1990 and 2023. They also estimated the impact those disorders had on global health by calculating their disability-adjusted life years (one DALY represents a year of healthy life lost).

The findings, published in JAMA Psychiatry on Wednesday, are unsettling. Between 1990 and 2023, there was a 71% increase in the incidence of those two eating disorders, the team found. Those conditions had a huge impact on global health, too. In 2023, they “cost the world the equivalent of about 3.83 million years of healthy life,” says Bai.

Bai and his colleagues expect these figures to continue to rise. In 2023, an estimated 18.1 million people had either anorexia nervosa or bulimia nervosa; by 2050, the team predict, the two conditions could affect 21.8 million people.

It gets worse. Those two conditions represent less than half of all eating disorders, says Levinson, who was not involved in Bai’s study. The true impact of these disorders is “probably three times the rate that they’re reporting,” she says.

Bai and his colleagues didn’t explore the reasons for this rise (although Bai points out that it can’t be entirely explained by population growth). So how did we get here?

Social media may have played a role. “Social media comparisons, influencers, challenges, diet trends can all be precipitating … events for an eating disorder,” says Cynthia Bulik, a clinical psychologist at the University of North Carolina, who was not involved in the new study. Bulik has been studying and treating eating disorders since the early 1980s. “Between 1990 and 2023 there has clearly been an upsurge of those types of environmental exposures,” she says.

Many social media users are exposed to filtered and modified images of skinny influencers who may have undergone cosmetic surgery or extreme dieting and who are promoting “thinspiration” or “fitspiration” content. But users with eating disorders appear to be more likely to engage with this content—and to be fed more of it as a result.

When researchers at the University of Melbourne investigated this phenomenon a couple of years ago, they found that the algorithms of TikTok users with eating disorders delivered 4343% more “toxic eating disorder videos” than those of users who didn’t have the disorders. They also got more content focused on dieting or exercising. Many were fed over 1000 problematic videos per month. Bulik describes social media as “quicksand” for people with, or at risk of, eating disorders.

The covid-19 pandemic—and the social isolation that accompanied lockdowns—also appears to have contributed to the rise in eating disorders. Research conducted in 2021 found that the number of people hospitalized with eating disorders in the US doubled between 2018 and 2020.

Researchers don’t yet have good data on how trends in eating disorders might have progressed since 2023. But anecdotally, things seem to be getting even worse. “There are more eating disorders, more severe eating disorders… and we’re seeing earlier ages of onset,” says Levinson. “We’re seeing eight- and nine-year-olds with eating disorders.”

“From what I’m hearing and seeing, I am not going to be surprised if that number isn’t at least doubling again,” she adds.

Levinson particularly worries about the impact of GLP-1 weight-loss drugs, which have exploded in popularity in recent years. A Gallup survey conducted earlier this year suggested that 15% of Americans had used one of these drugs for weight loss. And they’re becoming more accessible—drug companies are working on new formulations that are both easier to take and more effective, and many people can buy versions of these drugs from compound pharmacies or dubious online sources.

In a small study published earlier this year, Levinson and her colleagues found that 30% of people with eating disorders had used GLP-1 drugs. The drugs might benefit some of those people. But 10% appeared to have misused the drugs. And Levinson thinks that figure is also an underestimate.

Levinson says she and her colleagues have seen children develop eating disorders after being prescribed GLP-1s for medical reasons. They’ve seen others—even children as young as 11—buy versions of these drugs over TikTok.

“We’re very concerned that GLPs will lead to even faster and more extreme weight loss in people with anorexia nervosa, and that patients will come to us in even more advanced stages of illness that become difficult to reverse,” says Bulik.

Both Bulik and Levinson are continuing to study the impacts of these drugs in eating disorders. In the meantime, both urge medical professionals to consider the risks GLP-1s might pose to people with those disorders. Bulik contributed to a set of research and clinical guidelines, published earlier this month, that recommends doctors screen for eating disorders before prescribing the drugs, for example. The guidelines also recommend more patient education and more robust monitoring of people who take GLP-1s.

If you think you or someone you know might have an eating disorder, you can find information and support at the National Eating Disorders Association.

This article first appeared in The Checkup, MIT Technology Review’s weekly biotech newsletter. To receive it in your inbox every Thursday, and read articles like this first, sign up here.

STAT+: Telehealth leapfrogs FDA approval with off-label GLP-1 prescriptions for addiction

For the popular diabetes and obesity medications known as GLP-1s, addiction medicine is one of many possible promising next acts.

Eli Lilly is investigating a new drug, brenipatide, and testing it in alcohol use disorder. A new company called Baseline Therapeutics is studying a GLP-1 receptor agonist in alcohol, cocaine, and methamphetamine use disorders; Altimmune is testing one called pemvidutide for alcohol. Addiction medicine specialists are hopeful that the Food and Drug Administration may soon approve a new medication for alcohol use disorder — the first in two decades. 

But some telehealth platforms that specialize in treating addiction aren’t waiting. In the last year, a few companies have begun to market off-label prescriptions of GLP-1s, both branded and compounded, to treat alcohol use disorder and other addictions. 

Continue to STAT+ to read the full story…

Forty percent of high school students report binge eating in new CDC survey

About 40% of high school students reported binge eating in the past month in a large new government survey, putting a spotlight on how widespread this common form of disordered eating is among young people in the U.S. 

This is the first time the biennial survey, conducted by the Centers for Disease Control and Prevention, has asked about binge eating, and the first data on disordered eating collected by the Youth Risk Behavior Survey in a decade. Among its other striking findings: 17% of nearly 7,000 students said they had engaged in binge eating within the last three days.  “It was extremely alarming seeing how high those numbers were,” said Ariel Beccia, a researcher in the adolescent and children’s medicine division of Boston Children’s Hospital. 

Read the rest…

STAT+: FDA adcomm members’ ties to industry under scrutiny

Good morning. Yesterday, we got the devastating news that STAT’s infectious disease correspondent, Helen Branswell, passed away. She was a reporter I always looked up to and was also an incredibly generous colleague. Please take the time to read this tribute to her.

The need-to-know this morning

  • Liquidia and United Therapeutics reached a settlement agreement that resolves their ongoing patent litigation over Liquidia’s treatment for pulmonary arterial hypertension. 

New trial to test effect of GLP-1s on social outcomes

Researchers in Pittsburgh are launching a novel type of GLP-1 trial — testing not only if the drug will help people with addiction, but also if it will help reduce recidivism while improving other employment, health, and housing outcomes.

The Center for Addiction Science, Policy and Research (CASPR), a non-profit organizing the effort, and the Allegheny County Department of Human Services are planning to launch the trial in early 2027. They will enroll up to 1,200 adults with alcohol use disorder, opioid use disorder, or both, who are also at high risk of criminal-justice involvement. Half will be randomly assigned to take Novo Nordisk’s Wegovy, according to a press release issued yesterday.

Continue to STAT+ to read the full story…

Increased frequency of rare damaging variants in PLIN4 in patients with anorexia nervosa

BackgroundAnorexia nervosa (AN) is a severe psychiatric disorder characterized by pathological fear of weight gain, persistent restriction of energy intake, and pervasive body image disturbance. Although AN is substantially heritable, the rare coding variants contributing to susceptibility remain incompletely defined. Pedigree-based whole-exome sequencing studies can help to identify rare, high-effect protein-coding variants and thereby elucidate disease mechanisms.MethodsWhole-exome sequencing was performed on AN-dense family pedigrees to identify candidate genes. The frequency of rare (minor allele frequency <1%), putatively damaging [combined annotation-dependent depletion (CADD) >15, including predicted loss-of-function (LOF)] variants in 32 candidate genes was assessed in 154 unrelated probands of European ancestry. Carrier counts were compared against gnomAD v4 non-Finnish European reference using a TRAPD-style coverage-adjusted denominator (maximum allele number/2 across qualifying sites per gene). Significance was assessed by one-sided Fisher exact test, Bonferroni-corrected across 32 candidate genes.ResultsPLIN4 was the only gene enriched in AN cases at Bonferroni-corrected significance: 17 of 154 patients (11.0%) carried a rare damaging variant versus 3.2% in gnomAD non-Finnish Europeans [odds ratio (OR) 3.86, 95% confidence interval (CI) 2.35 to 6.35; one-sided Fisher p = 9.98e−06; Bonferroni-corrected p = 3.19e−04]. The result is preserved with the index proband removed (n = 153; OR 3.64, 95% CI 2.18 to 6.07; corrected p = 1.13e−03). Six of 154 patients carried predicted LOF variants; 4 of 6 had at least one AN-affected first- or second-degree relative carrying the same variant, providing co-segregation evidence. The LOF-only burden did not survive Bonferroni correction in isolation.ConclusionsPatients with AN carry a statistically significant excess of rare, putatively damaging variants in PLIN4, at a threshold surviving Bonferroni correction across 32 candidate genes. These findings implicate lipid-droplet biology as a candidate contributor to AN pathophysiology.