The research team, led by Renae D. Schmidt and Daniel J. Feaster at the University of Miami’s Miller School of Medicine, together with colleagues at Columbia University and the University of Chicago, analyzed data from a nationally representative survey of U.S. primary care and emergency department physicians conducted between October 2020 and October 2022. The survey achieved a Council of American Survey Research Organizations response rate of 53.6 percent, a respectable figure for physician surveys, and the analysis included 855 physicians. Because the sample was designed to be nationally representative, the results offer one of the clearer windows to date into how training and attitudes intersect across the frontline settings where most Americans first encounter the health system.

The researchers examined three distinct strands of addiction education: hours of training received during medical school, hours received during residency, and participation in Continuing Medical Education, the ongoing coursework practicing physicians complete throughout their careers. They also measured each physician’s perceived capability to care for individuals with substance use disorders, essentially a self-assessed sense of competence. Stigma was assessed with validated instruments, including the Drug and Drug Problems Perceptions Questionnaire and the Medical Condition Regard Scale, which capture how providers emotionally and cognitively regard patients with drug-related problems. Clinical practice was measured by whether physicians reported providing treatments such as medication for opioid use disorder, behavioral counseling, and treatment for alcohol and other drug use disorders.

The statistical architecture of the study was as important as its findings. Because primary care physicians and emergency department physicians were surveyed with overlapping instruments, the team used generalized estimating equations, a regression framework that handles correlated observations and allows estimates to be pooled across physician types while adjusting for covariates including sex, race and ethnicity, years since degree, and physician type. This approach let the investigators isolate the association of training with stigma and practice while holding demographic and career factors constant, and then to test whether perceived capability sat on the causal pathway between training and stigma.

The headline result is strikingly consistent across substances. More addiction training was associated with lower stigma toward patients with alcohol use disorders, with a standardized coefficient of negative 2.42 that was statistically significant at the one percent level. The same pattern held for stimulant use disorders, where the coefficient was negative 2.78, and for opioid use disorders, where it was negative 2.48. In other words, physicians who reported more training expressed measurably less stigmatizing regard for patients regardless of which substance was involved, a notable finding given that stimulant use disorders lack FDA-approved medications and are often viewed with particular pessimism in clinical settings.

Training was also linked to what physicians actually did, not just how they felt. Each increment of training was associated with higher odds of providing treatment: a prevalence ratio of 1.19 for treating alcohol use disorders, 1.09 for providing medication for opioid use disorder, 1.10 for delivering behavioral counseling for opioid use disorder, and 1.14 for treating disorders related to other drugs. These ratios were statistically significant, and while the magnitudes may look modest, they compound across a workforce of hundreds of thousands of physicians and millions of patients. In a country where only a small fraction of people with substance use disorders receive any treatment at all, shifting the delivery behavior of frontline physicians even slightly could translate into a substantial public health effect.

The most conceptually interesting finding concerns mediation. Higher perceived capability was itself strongly associated with lower stigma toward patients with alcohol, stimulant, and opioid use disorders, with coefficients of negative 0.55, negative 0.51, and negative 0.59 respectively, all highly significant. When the researchers accounted for perceived capability in their models, the direct effect of training on stigma disappeared. Yet the indirect effect of training, operating through perceived capability, remained significant for all three disorder types, with coefficients of negative 0.11 for alcohol, negative 0.09 for stimulants, and negative 0.11 for opioids. This pattern indicates full mediation: training reduces stigma not by directly changing attitudes, but by building physicians’ confidence in their own clinical competence.

That mechanism carries real weight for how medical education should be designed. If stigma were driven primarily by moral judgment or personal bias, remediation would require attitude-focused interventions, such as contact-based anti-stigma campaigns. But if, as this study suggests, stigmatizing attitudes toward patients with substance use disorders flow substantially from a sense of being unprepared, then the remedy is skills-based: teach physicians to screen, diagnose, prescribe, and counsel, and the negative regard tends to recede on its own. This reframing aligns with a broader shift in addiction medicine away from viewing provider stigma as an intractable cultural problem and toward viewing it as a trainable deficit, one that responds to the same pedagogical tools used for any other clinical competency.

The study’s limitations deserve honest attention. As an exploratory secondary analysis of cross-sectional data, it cannot establish causality or direction; it is possible that less stigmatizing physicians seek out more training rather than training reducing stigma, or that unmeasured factors, such as institutional culture or personal exposure to addiction, shape both training uptake and attitudes. Stigma and capability were self-reported, which introduces the possibility of social desirability bias, and the survey period overlapped with the COVID-19 pandemic, a time of unusual strain on both primary care and emergency departments. The authors also note that the long-term durability of training effects, and how specific training content maps onto specific outcomes, remains under-explored, making this an open field for longitudinal work.

Even with those caveats, the implications are concrete. The research was supported by the National Institute on Drug Abuse through the National Drug Abuse Treatment Clinical Trials Network, a infrastructure designed to move evidence-based addiction treatment into routine care, and the message for that mission is clear. Medical schools and residency programs that treat addiction medicine as a marginal elective may be inadvertently cultivating the very stigma they hope to eliminate, while continuing education requirements that build hands-on capability, including the prescribing of medications for opioid use disorder, offer a scalable lever for change. With overdose and alcohol-related mortality remaining leading causes of death in the United States, and with primary care and emergency departments serving as the front door to care for most patients, equipping physicians with genuine competence may be the most efficient anti-stigma intervention available. The study, published open access under a Creative Commons license, adds to a growing body of evidence that the antidote to stigma is not simply empathy training but mastery, and that every hour of addiction education a physician receives may ripple outward into better, more equitable care for one of medicine’s most underserved patient populations.

Subject of Research: Associations between physician addiction training, stigma toward patients with substance use disorders, and clinical treatment practices in U.S. primary care and emergency medicine

Article Title: Associations of addiction training with attitudes and clinical practices: a cross-sectional survey of primary care and emergency department physicians in the US

Article References: Schmidt, R. D., Gravlin, A., Duan, R., Gonzalez, S. T., DiCamillo, C. A., Gooden, L. K., Pollack, H. A., Parish, C. L., Metsch, L. R., Horigian, V. E., & Feaster, D. J. (2026). Associations of addiction training with attitudes and clinical practices: a cross-sectional survey of primary care and emergency department physicians in the US. Addiction Science & Clinical Practice. https://doi.org/10.1186/s13722-026-00722-x

Image Credits: AI Generated

DOI: 10.1186/s13722-026-00722-x

Keywords: addiction training, substance use disorder, provider stigma, medical education, primary care, emergency medicine, opioid use disorder, alcohol use disorder, continuing medical education, perceived capability, cross-sectional survey, medication for opioid use disorder