The scale of the problem motivating the study is considerable. New psychoactive substances—a shifting family of synthetic compounds designed to mimic the effects of controlled drugs while evading legal restrictions—have become one of the most difficult fronts in global drug policy. Because their chemical structures change constantly, their health consequences are poorly characterized, and people who begin using them in adolescence face heightened risks of lasting neurobiological and psychological harm. The adolescent brain, with its early-maturing reward circuitry and still-developing self-control systems, is particularly vulnerable to substance initiation and to the entrenched patterns of craving that follow.

To uncover the hidden structure of craving, the researchers assessed 1,008 patients in Tianjin whose substance use disorder began during adolescence. Rather than reducing craving to a single score, the team measured three separate dimensions: the intensity of craving itself, the degree to which it interfered with daily functioning, and the effort patients expended resisting it. Latent profile analysis—a form of statistical modeling that sorts individuals into unobserved subgroups based on patterns across multiple measures—was then applied to see whether distinct combinations of these dimensions existed within the sample.

Three profiles emerged, and they were strikingly uneven in size. The largest group, accounting for 54 percent of participants, was labeled relatively stable, showing moderate craving without overwhelming functional disruption. A second group, 40.2 percent of the sample, fell into a high craving struggle category marked by intense urges and an exhausting internal battle to resist them. The smallest but most concerning group, 6 percent, was described as functionally impaired, in which craving severely disrupted everyday life. That two in five patients occupied the high struggle profile underscores how common severe craving is in this population—and how misleading it would be to treat them all identically.

Identifying the profiles was only the first step. The team then asked whether psychosocial information could predict which profile a patient belonged to, using three machine learning models of increasing sophistication: logistic regression, a classical statistical baseline; random forest, an ensemble method that averages many decision trees; and XGBoost, a gradient-boosted algorithm known for its performance on tabular medical data. Performance was evaluated using macro-AUC, accuracy, and Macro-F1 scores, metrics that account for the imbalanced group sizes. XGBoost achieved the strongest discrimination among the three, although the researchers are candid that overall classification performance remained modest—a limitation they attribute to the exploratory nature of the work and the complexity of the underlying constructs.

Perhaps the most clinically valuable output came from interpreting the models. Using SHAP, an explainability technique that quantifies each feature’s contribution to predictions, the researchers identified the psychosocial variables that most strongly distinguished the craving subgroups. Social support emerged as a leading factor, consistent with a growing body of evidence that supportive networks buffer against relapse. Resistance to peer influence also ranked highly—a finding that resonates with developmental research showing adolescence as the period of peak susceptibility to peer pressure. Coping style, childhood adversity, age, and age at first drug use completed the list of key contributors, weaving together life history, developmental timing, and present-day resources into a single predictive picture.

These findings align closely with the biopsychosocial model of addiction, which holds that substance use disorders arise from interactions among biological vulnerability, psychological processes, and social context rather than from any single cause. The prominence of childhood adversity in the models echoes systematic reviews linking early maltreatment to elevated risk of substance misuse, while the role of social support reflects studies demonstrating that perceived support moderates the relationship between stress and relapse. By embedding these factors in a quantitative, subgroup-aware framework, the study gives that conceptual model a practical, data-driven form.

The authors are careful to frame the work as exploratory rather than definitive. The cross-sectional design captures a single moment in time and cannot establish causal direction—for instance, whether low social support drives intense craving or vice versa. The modest discriminative performance means the models are not yet ready for individual-level clinical decisions, and the data, which contain sensitive personal information, are not publicly available for reasons of privacy protection. Still, the researchers argue that the approach may provide preliminary evidence for risk stratification and for subgroup-sensitive addiction nursing support, in which care is tailored to whether a patient fits the stable, high struggle, or functionally impaired pattern.

If future longitudinal studies confirm and refine these profiles, the implications could be far-reaching. A patient in the functionally impaired group might warrant immediate intensive intervention, while someone in the high craving struggle group might benefit most from coping-skills training and peer-resistance support, and those in the stable group from relapse prevention and social strengthening. In an era when new psychoactive substances continually outpace regulation and traditional treatment models, the ability to sort patients by their actual craving signature—rather than a one-size-fits-all label—could mark a meaningful shift toward precision medicine in addiction care. The study, funded by the Ministry of Education in China, is an early but provocative demonstration that the algorithms of machine learning can find order in one of addiction’s most stubborn and subjective experiences.

Subject of Research: Exploratory risk stratification of adolescent-onset new psychoactive substance use disorder using latent craving profiles and machine learning

Article Title: Exploratory Risk Stratification of Patients With Adolescent-Onset New Psychoactive Substance Use Disorder: Integrating Latent Craving Profiles, Psychosocial Factors, and Machine Learning

Article References: Zhuang, S., Che, X., Song, Y., Li, S., Shang, X., Shi, L., Zhang, X., Song, C., Jing, J., Fan, J., Qin, S., Xu, Z., Zhang, J., Li, J., Li, L., Hou, K., & Jiang, Y. (2026). Exploratory Risk Stratification of Patients With Adolescent-Onset New Psychoactive Substance Use Disorder: Integrating Latent Craving Profiles, Psychosocial Factors, and Machine Learning. International Journal of Mental Health and Addiction. https://doi.org/10.1007/s11469-026-01714-3

Image Credits: AI Generated

DOI: 10.1007/s11469-026-01714-3

Keywords: new psychoactive substances, adolescent-onset substance use disorder, craving, latent profile analysis, machine learning, XGBoost, SHAP, social support, childhood adversity, risk stratification, addiction nursing, peer influence

A recovery philosophy once associated with personal experimentation is moving into the center of an increasingly complicated cannabis debate. Known as “California sober,” the approach generally means abstaining from alcohol and other intoxicating drugs while continuing to use cannabis and, in some interpretations, psychedelics. A new commentary in the International Journal of Mental Health and Addiction argues that the expanding cannabis market is forcing addiction medicine to reconsider what sobriety means—and whether flexible recovery models can be made safe without replacing one harmful dependence with another.

The concept challenges the definition of recovery used by traditional abstinence-based programs, including Alcoholics Anonymous. In those settings, sobriety typically means complete cessation of intoxicating substances. California sober instead treats recovery as a potentially individualized process, allowing a person to eliminate substances that have caused the greatest harm while retaining cannabis or another psychoactive substance. For some people, the authors suggest, this may feel more achievable than an all-or-nothing demand for abstinence. But the commentary emphasizes that California sober is not a standardized clinical treatment, diagnostic category or evidence-based protocol. Its meaning can vary substantially from one person to another.

That ambiguity matters because cannabis is not a single, uniform exposure. The modern cannabis marketplace includes smoked and vaporized flower, concentrates, oils, tinctures, capsules, edibles, beverages and products derived from hemp. Their chemical compositions, delivery routes and time courses differ. Inhaled tetrahydrocannabinol, or THC, reaches the bloodstream rapidly and can produce a relatively fast onset of intoxication. Edible products are absorbed more slowly, are metabolized partly through the liver into 11-hydroxy-THC and may produce effects that last considerably longer. Concentrates can deliver much higher THC doses than traditional plant material. Cannabidiol, or CBD, is not intoxicating in the same way as THC, but products marketed as CBD may contain varying amounts of other cannabinoids or contaminants, particularly in loosely regulated markets.

The authors connect this product proliferation to a growing public-health challenge. Cannabis legalization in parts of the United States has expanded commercial access while leaving major differences among state and federal rules. Hemp-derived THC products have added another layer of complexity, because products sold outside conventional cannabis dispensaries may be subject to different testing, labeling and age restrictions. The result is a landscape in which consumers may have difficulty determining how much THC they are taking, how quickly it will act or whether a product contains what its packaging claims. Greater physical availability may also affect patterns of use. The commentary cites research examining links between retail availability, frequent cannabis consumption and related health harms, underscoring that convenience and normalization can influence exposure.

The central clinical concern is cannabis use disorder, a condition characterized by impaired control over cannabis use despite negative consequences. Symptoms can include unsuccessful efforts to cut down, persistent craving, spending substantial time obtaining or using cannabis, and continuing to use despite problems at work, school, in relationships or in physical and mental health. Regular exposure to THC can produce neuroadaptations in the brain’s endocannabinoid and reward systems, potentially contributing to tolerance, withdrawal and compulsive use. Withdrawal may involve irritability, anxiety, sleep disturbance, reduced appetite and intense urges to resume use. These symptoms are generally less medically dangerous than withdrawal from alcohol or some sedative drugs, but they can still undermine recovery and make a person’s chosen limits difficult to maintain.

That creates the possibility of what clinicians call addiction substitution: replacing one problematic substance or behavior with another. The commentary does not claim that every person who uses cannabis during recovery will develop cannabis use disorder, nor does it conclude that cannabis use is always more harmful than continued use of alcohol or other drugs. Instead, the authors argue that substitution should not be assumed to be either inherently successful or inherently dangerous. Its effects may depend on the person’s history, the substance being replaced, the pattern and dose of cannabis use, co-occurring psychiatric conditions, social environment and recovery goals. A strategy that reduces exposure to a highly lethal drug may produce benefits in one context, while heavy cannabis use may create new impairment or destabilize recovery in another.

The distinction between harm reduction and recovery is crucial to this debate. Harm reduction seeks to reduce the health and social consequences of substance use without requiring immediate abstinence. In practice, that might mean helping someone avoid overdose, reduce risky combinations, use less frequently or shift away from a more dangerous substance. Recovery, by contrast, is a broader and more subjective concept that may include improved health, stable housing, restored relationships, psychological well-being and sustained control over substance use. The authors note that people may define a sober life in different ways, and that treatment systems can lose patients when they impose rigid rules or discharge anyone who does not meet an abstinence standard. Yet a flexible label alone cannot ensure that a person is safer. It must be paired with monitoring, honest discussion of risks and the ability to change course when harms emerge.

Cannabis can also complicate recovery through its interactions with other substances. Research cited in the commentary has associated alcohol, tobacco and marijuana use with increased odds of using multiple substances on the same day. Such co-use may be especially concerning because the effects of one drug can obscure or amplify those of another. Cannabis-related impairment can affect attention, reaction time, memory and judgment, while combining substances may make it harder for individuals to recognize escalating risk. For people recovering from opioid or stimulant addiction, cannabis may be perceived as a safer substitute, but “safer” does not mean risk-free, and the substitution may not address the psychological, social or environmental factors that sustain addiction. The authors therefore call for clinicians to assess outcomes rather than rely on assumptions about any particular substance.

The policy environment makes these decisions harder. Cannabis remains federally controlled in the United States even as states have legalized medical or recreational markets, and proposed or ongoing changes in scheduling and regulation may influence research, commercial access and clinical practice. Meanwhile, products containing THC can appear in foods, drinks and other consumer formats that do not resemble traditional drug use. This rapid evolution has outpaced the development of consistent standards for describing or evaluating California sober recovery. The commentary argues that treatment providers need clearer language: patients should know whether a program supports abstinence, harm reduction, monitored cannabis use or a broader individualized plan. Without such clarity, the same phrase may communicate very different expectations to patients, families and clinicians.

The authors ultimately frame California sober not as a universal answer, but as a test of whether addiction care can become more personalized without becoming less rigorous. Any decision to incorporate cannabis into a recovery plan should consider the individual’s prior substance use, current symptoms, medical and psychiatric history, risk of relapse, cannabis dose and route, functional effects and personal goals. Clinicians may need to track craving, withdrawal, frequency of use, escalating tolerance and consequences over time, while also asking whether cannabis is helping a person move toward stability or becoming the new center of daily life. The authors acknowledge that evidence remains limited and that much of the debate has developed faster than formal research. Their warning is therefore less a verdict on California sober than a demand for better studies, consistent terminology and safety-focused care. As cannabis products continue to multiply and social attitudes shift, recovery may no longer fit neatly into a single definition—but flexibility without evidence could leave vulnerable people navigating a rapidly changing drug landscape without reliable guardrails.