The scale of the investigation is what sets it apart. The research team, led by Zihan Dong of Karolinska Institutet in Sweden together with an international consortium spanning more than thirty institutions, identified 1,929,027 adults who received a first recorded diagnosis of type 2 diabetes between 2010 and 2020. Of these, roughly 47 percent were male. Because the study drew on national health registries, insurance claims databases, and prescription records rather than self-report, the researchers could capture ADHD diagnoses and medication dispensations with a degree of completeness that smaller single-country studies cannot match. Each national dataset was analysed under a standardized analytical protocol before the results were combined, a design choice that minimizes the methodological heterogeneity that has plagued earlier cross-study comparisons.

The headline result is stark: adults with type 2 diabetes and co-occurring ADHD had a hazard ratio of 2.42 for all-cause mortality after their diabetes diagnosis, with a 95 percent confidence interval of 1.72 to 3.41. In practical terms, even after accounting for the uncertainty inherent in observational data, the death rate among these patients was more than double that of their diabetes-only counterparts. Crucially, the association held after the researchers adjusted for sex, age, calendar year at diabetes diagnosis, psychiatric history, use of other psychotropic medications, and use of cardiometabolic drugs—the standard arsenal of confounders that critics of pharmacoepidemiology studies typically raise. The consistency of the signal across seven healthcare systems with different diagnostic cultures, treatment guidelines, and record-keeping infrastructures strengthens the case that the finding reflects a genuine clinical phenomenon rather than an artifact of any single country’s data.

Perhaps more surprising is what the study did not find. ADHD was not consistently associated with major adverse cardiac events—a composite outcome encompassing acute myocardial infarction, stroke, and cardiovascular disease-related death. The pooled hazard ratio for MACE was 1.04, with a confidence interval of 0.81 to 1.35, meaning the data cannot distinguish the true effect from no effect at all. This dissociation between mortality and cardiac events is scientifically intriguing. It suggests that the excess deaths among diabetes patients with ADHD may not be driven primarily by the classical cardiovascular pathway that dominates outcomes research in type 2 diabetes. Alternative explanations could include non-cardiac causes of death, difficulties with diabetes self-management, higher rates of psychiatric comorbidity, medication non-adherence, or lifestyle factors that are harder to capture in registry data. The authors note that the mortality signal was robust while the cardiac one was not, a pattern that warrants deeper investigation into the specific causes of death in this population.

The study also examined a second, clinically urgent question: among adults who have both type 2 diabetes and ADHD, does taking ADHD medication change cardiovascular risk? Here the answer was conditional and age-dependent. Overall, ADHD medication use was not associated with elevated MACE risk across the full cohort. But when the researchers stratified by age at diabetes onset, a clear signal emerged: among patients diagnosed with type 2 diabetes before age 45—so-called early-onset type 2 diabetes—ADHD medication use was associated with a hazard ratio of 1.58 for major adverse cardiac events, with a confidence interval of 1.18 to 2.10. That interval excludes the null value, indicating a statistically reliable elevation in risk within this younger subgroup.

The age stratification matters because early-onset type 2 diabetes is itself an aggressive disease. Patients diagnosed in early adulthood accumulate longer lifetime exposure to hyperglycemia and typically face higher cumulative risks of retinopathy, nephropathy, and cardiovascular complications than those diagnosed later in life. Layering stimulant and non-stimulant ADHD medications onto an already vulnerable metabolic profile may compound that risk. Stimulant medications such as methylphenidate and amphetamine derivatives act on the sympathetic nervous system, raising heart rate and blood pressure—effects that are generally well tolerated in healthy adults but potentially consequential in patients with insulin resistance, dyslipidemia, and emerging vascular disease. The study’s time-varying exposure design, which allowed medication status to change over the follow-up period rather than being fixed at baseline, lends additional credibility to this subgroup finding.

Methodologically, the researchers employed cause-specific Cox proportional hazards models, the workhorse of survival analysis, to estimate hazard ratios with 95 percent confidence intervals for each country separately. These country-specific estimates were then pooled using random-effects meta-analysis, a technique that assumes the true effect may vary across populations and quantifies that variability rather than pretending it does not exist. Treating both ADHD status and medication use as time-varying exposures was a deliberate safeguard against immortal time bias, a subtle distortion that arises when researchers classify patients by a status they could only achieve after surviving a certain period. The adherence to the STROBE reporting guidelines for observational epidemiology further signals the team’s attention to transparency and reproducibility.

The study was supported by the European Union Horizon 2020 Research and Innovation Programme, along with national funders including the Australian National Health and Medical Research Council, and the funders had no role in the design, analysis, or publication decisions. The authors declared a range of competing interests typical of a large psychiatric-pharmacoepidemiology collaboration, including speaker fees and research funding from pharmaceutical companies for several senior investigators, all reported as outside the submitted work. The article is published open access under a Creative Commons Attribution license, and the team has shared supplementary materials detailing the country-specific analyses.

For clinicians, the message is twofold. First, a diagnosis of ADHD in an adult with type 2 diabetes should prompt intensified monitoring and integrated somatic-psychiatric care, because the co-occurrence marks a population at substantially elevated risk of dying. Adults with ADHD are, on average, diagnosed with type 2 diabetes at younger ages and may experience greater difficulty adhering to complex medication regimens, dietary recommendations, and glucose monitoring routines—mechanisms the study’s authors flag as plausible contributors to the mortality gap. Second, the decision to prescribe ADHD medication to a patient with early-onset type 2 diabetes deserves careful individualized consideration, weighing the demonstrated benefits of treatment on function and safety against a possible cardiovascular signal in this vulnerable group. The researchers emphasize that their findings suggest a potential need for integrated care models and careful medication stewardship, particularly for younger-onset diabetes, rather than any blanket restriction on ADHD treatment.

The broader significance of the work lies in its demonstration that psychiatric comorbidity is not a peripheral concern in metabolic disease. As ADHD diagnoses in adults continue to rise and as type 2 diabetes increasingly affects younger populations, the overlap between the two conditions will grow. Studies of this scale, harnessing the combined statistical power of seven national data systems, provide the evidentiary foundation that single-country analyses cannot. What remains to be established is the mechanism: whether the doubled mortality reflects biological interactions, behavioral pathways, healthcare access disparities, or some combination of all three. Answering that question will determine whether the intervention that matters most is cardiological, psychiatric, or structural—and for now, the study’s authors argue, the safest bet is closer collaboration between the specialists who treat the brain and those who treat the blood sugar.

Subject of Research: The association of ADHD and ADHD medication use with major adverse cardiac events and mortality in adults with type 2 diabetes

Article Title: Major adverse cardiac events and mortality associated with ADHD and its treatment in adults with type 2 diabetes: a multi-national cohort study

Article References: Dong, Z., Zhou, Y., Yao, H., Liu, S., Astrup, A., Engeland, A., Gao, L., Gillies, M. B., Lunde, A., Zhang-James, Y., Araki, M., Brikell, I., Butwicka, A., Dalsgaard, S., D’Onofrio, B. M., Faraone, S. V., Gudbjornsdottir, S., Haavik, J., Kuja-Halkola, R., … Du Rietz, E. (2026). Major adverse cardiac events and mortality associated with ADHD and its treatment in adults with type 2 diabetes: a multi-national cohort study. BMC Medicine, 24(1), Article 527. https://doi.org/10.1186/s12916-026-05246-4

Image Credits: AI Generated

DOI: 10.1186/s12916-026-05246-4

Keywords: ADHD, type 2 diabetes, major adverse cardiac events, mortality, ADHD medication, cardiovascular risk, cohort study, pharmacoepidemiology, early-onset diabetes, multinational study, psychiatric comorbidity, BMC Medicine

The scale of the problem is difficult to overstate. According to the study, nearly 39 percent of the adolescents surveyed reported having experienced at least one serious injury, defined through standardized questionnaires administered in school settings. The burden was not distributed evenly between the sexes: 45.60 percent of boys reported a serious injury compared with 33.32 percent of girls, a gap consistent with earlier research showing that boys tend to engage in more risk-taking behavior and are exposed more frequently to hazardous activities such as road traffic, sports, and physical violence. Global Burden of Disease analyses cited by the authors have repeatedly identified transport and unintentional injuries as dominant causes of mortality among people aged 10 to 24, making any modifiable risk factor for injury a priority target for public health intervention.

At the heart of the analysis was a straightforward but consequential question: are adolescents who use tobacco or substitute nicotine products, or who breathe other people’s smoke, more likely to suffer serious injuries than their unexposed peers? To answer it, the research team, led by Manman Chen and Zichen Ye of the Chinese Academy of Medical Sciences and Peking Union Medical College, applied multivariable logistic regression models to the pooled survey data. These statistical models allow investigators to estimate the association between an exposure and an outcome while adjusting for other variables that could confound the relationship. The primary analysis produced a clear signal. Adolescents who currently used tobacco or substitute products had roughly twice the odds of reporting a serious injury compared with non-users, with an odds ratio of 2.07 and a 95 percent confidence interval of 1.95 to 2.19.

Secondhand smoke told a similar story. Among adolescents who did not smoke themselves but were exposed to smoke in their environment, the odds of reporting a serious injury were 43 percent higher than among unexposed non-smokers, corresponding to an odds ratio of 1.43 with a confidence interval of 1.37 to 1.49. That a passive exposure, one the adolescents did not choose, was associated with injury risk is perhaps the most sobering element of the findings. It implies that the injury burden linked to tobacco extends into homes and communities where children cannot control their exposure, echoing a growing body of literature on the harms of secondhand exposure, including recent systematic reviews of secondhand exposure to e-cigarette aerosols.

The researchers then subjected their primary results to a series of robustness checks, and the picture that emerged was more nuanced than the headline numbers suggest. When the models were additionally adjusted for the year in which each survey was administered, the association for current tobacco and substitute use was attenuated, dropping from an odds ratio of 2.07 to 1.49, though it remained statistically significant with a confidence interval of 1.44 to 1.55. The estimate for secondhand smoke exposure, by contrast, was largely unchanged by the extra adjustment. This pattern indicates that some of the raw association between active use and injury may reflect temporal or cohort effects captured by survey year, but that a genuine residual association persists even after accounting for when the data were collected.

One of the most technically interesting aspects of the study is its treatment of ambient air pollution as a potential modifier of the tobacco-injury relationship. The authors stratified their analyses by country-level concentrations of fine particulate matter, or PM2.5, the microscopic airborne particles smaller than 2.5 micrometers that penetrate deep into the lungs and are implicated in oxidative stress and systemic inflammation. Prior research has linked long-term PM2.5 exposure to fall-related injuries in adults in low- and middle-income countries, and animal studies have shown that pollution-derived PM2.5 induces persistent oxidative stress and inflammation in lung tissue even after exposure ceases. The hypothesis underlying the stratification is that adolescents living in highly polluted environments may already carry a physiological load that interacts with tobacco smoke exposure, compounding its effects.

The stratified results offered partial support for that idea. Association estimates were slightly higher among girls than boys and slightly higher in countries with elevated PM2.5 levels, although most of these subgroup differences did not reach statistical significance. One exception stood out: among girls, the association between secondhand smoke exposure and serious injury was significantly stronger in the high-PM2.5 country group, with a P value for the subgroup difference below 0.001. This finding resonates with earlier evidence that females may be particularly vulnerable to some of the health effects of smoking, and it raises the possibility that the combination of polluted air and passive smoke exposure imposes a compounded burden on girls in the most contaminated environments. The authors are careful, however, to frame these as variations in the magnitude of associations rather than definitive evidence of biological interaction.

What mechanisms might connect tobacco exposure to injuries in the first place? The study, being cross-sectional, cannot establish causation, but the authors situate their findings within a rich mechanistic literature. Nicotine and tobacco smoke affect vascular function, cognition, and potentially attention and reaction time, any of which could plausibly increase the likelihood of accidents. Clinical research has shown that cigarette smoking increases complications following fractures and impairs bone healing, and cohort studies have documented that smoking cessation reduces fracture risk. There is also a behavioral pathway: adolescents who use tobacco are more likely to cluster with other risk behaviors, including alcohol use, violence, and reduced physical activity, a clustering phenomenon documented in analyses of hundreds of thousands of adolescents across 89 countries. Prior multi-country analyses have also shown that substance use among adolescents is associated with injuries partly through its mediating relationship with interpersonal violence.

The public health implications are considerable. If tobacco and substitute use roughly double the odds of serious injury among adolescents, then tobacco control measures, taxation, age restrictions, advertising bans, and school-based prevention programs, could deliver injury-reduction benefits that are rarely counted in their cost-effectiveness calculations. Likewise, the strong and stable association for secondhand smoke exposure underscores the value of smoke-free home and public space policies, particularly in low- and middle-income countries where the Global School-based Student Health Survey is concentrated and where regulatory protections are often weakest. Systematic reviews of interventions to prevent unintentional injuries in school-aged children have identified school-based education programs and environmental modifications as effective tools, and integrating tobacco prevention into these frameworks could multiply their impact.

As with any observational study, important caveats apply. The injury data are self-reported and may be subject to recall bias, the exposure measures rely on adolescent self-report of tobacco use and smoke exposure, and the cross-sectional design precludes any claim that tobacco use causes injuries rather than merely co-occurring with them. Residual confounding by factors such as alcohol use, socioeconomic status, and psychological distress cannot be excluded. Nevertheless, the sheer scale of the dataset, spanning 58 countries and more than 176,000 young people, lends the findings a weight that few single-country studies can match. The work was supported by the National Natural Science Foundation of China, and the authors report no competing interests. For parents, educators, and policymakers, the message is unambiguous: protecting adolescents from tobacco, both the smoke they inhale by choice and the smoke they breathe by circumstance, may be as much an injury-prevention strategy as a lung-health one, and in a world where nearly four in ten teenagers report being seriously hurt, every modifiable risk factor matters.

Subject of Research: Associations of adolescent tobacco use, substitute product use, and secondhand smoke exposure with serious injuries across 58 countries

Article Title: Tobacco and substitute use, secondhand smoke exposure, and serious injuries among adolescents: A multinational analysis of 58 countries

Article References: Chen, M., Ye, Z., Zhao, Y., Qu, Y., Wang, X., Wang, W., & Gu, M. (2026). Tobacco and substitute use, secondhand smoke exposure, and serious injuries among adolescents: A multinational analysis of 58 countries. Air Quality, Atmosphere & Health, 19(10), Article 223. https://doi.org/10.1007/s11869-026-02112-3

Image Credits: AI Generated

DOI: 10.1007/s11869-026-02112-3

Keywords: adolescent health, tobacco use, secondhand smoke, serious injuries, PM2.5, air pollution, Global School-based Student Health Survey, logistic regression, injury prevention, public health, nicotine, multinational study