Kelsey Isman

Kelsey Isman

Auto-generated profile, based on publicly available information

Kelsey Isman

National Institute on Drug Abuse Intramural Research Program

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
26 June 2024 in Psychiatry Research
Verified
Journal Article
Open access
SGSalvatore Giorgi
Salvatore Giorgi
KIKelsey Isman
Kelsey Isman
TLTingting Liu
Tingting Liu
ZFZachary Fried
Zachary Fried
JSJoão Sedoc
João Sedoc
BCBrenda Curtis
Brenda Curtis
Generative Artificial Intelligence (AI) systems such as OpenAI's ChatGPT, capable of an unprecedented ability to generate human-like text and converse in real time, hold potential for large-scale deployment in clinical settings such as substance use treatment. Treatment for substance use disorders (SUDs) is particularly high stakes, requiring evidence-based clinical treatment, mental health expertise, and peer support. Thus, promises of AI systems addressing deficient healthcare resources and structural bias are relevant within this domain, especially in an anonymous setting. This study explores the effectiveness of generative AI in answering real-world substance use and recovery questions. We collect questions from online recovery forums, use ChatGPT and Meta's LLaMA-2 for responses, and have SUD clinicians rate these AI responses. While clinicians rated the AI-generated responses as high quality, we discovered instances of dangerous disinformation, including disregard for suicidal ideation, incorrect emergency helplines, and endorsement of home detox. Moreover, the AI systems produced inconsistent advice depending on question phrasing. These findings indicate a risky mix of seemingly high-quality, accurate responses upon initial inspection that contain inaccurate and potentially deadly medical advice. Consequently, while generative AI shows promise, its real-world application in sensitive healthcare domains necessitates further safeguards and clinical validation.

31 July 2025 in The American Journal of Drug and Alcohol Abuse
Verified
Journal Article
Closed access
KIKelsey Isman
Kelsey Isman
SGSalvatore Giorgi
Salvatore Giorgi
JEJennifer D. Ellis
Jennifer D. Ellis
AHAndrew S. Huhn
Andrew S. Huhn
TLTingting Liu
Tingting Liu
BCBrenda Curtis
Brenda Curtis
Background: Perceived Substance Use Disorder (SUD) stigma, defined as the awareness of negative societal attitudes toward individuals with SUDs, may discourage treatment-seeking and completion. Unlike self-stigma (negative beliefs about oneself), perceived stigma reflects individuals' perceptions of stigma from the public. While self-stigma has been widely studied, research on perceived stigma's role in shaping treatment outcomes remains limited.Objectives: This study aimed to examine whether higher perceived stigma at treatment intake predicts premature treatment discontinuation and hypothesized that greater perceived stigma would be associated with increased rates of premature treatment discontinuation.Methods: A total of 7,591 participants (70.2% male) from 75 SUD treatment facilities across the United States completed surveys at treatment intake and early in treatment. Perceived stigma was assessed using the Perceived Stigma of Addiction Scale (PSAS). Treatment completion, defined as standard discharge (recommended duration of care), served as the primary outcome. Mixed-effects models evaluated the relationship between perceived stigma and treatment discontinuation while adjusting for demographic, SUD, and mental health-related covariates.Results: Higher perceived stigma significantly predicted an increased likelihood of premature discontinuation (adjusted odds ratio [AOR] = 0.97, 95% CI [0.95, 0.99], p < .001). For each one-unit increase in PSAS score, the odds of treatment completion decreased by 3%. This relationship persisted across all models, even after accounting for covariates.Conclusion: These findings underscore the importance of addressing perceived stigma at treatment intake and its role in predicting treatment retention. Routine screening for stigma and implementing stigma-reduction interventions during care may contribute to better treatment outcomes for individuals with SUDs.

22 May 2025 in NPP—Digital Psychiatry and Neuroscience
Verified
Journal Article
Open access
PDF file
AAAnkit Aich
Ankit Aich
TLTingting Liu
Tingting Liu
SGSalvatore Giorgi
Salvatore Giorgi
KIKelsey Isman
Kelsey Isman
RBRuhshana Bobojonova
Ruhshana Bobojonova
BCBrenda Curtis
Brenda Curtis
Linguistic hurdles in healthcare, such as complex language, significantly affect patient outcomes, including satisfaction with interaction, comprehension of healthcare materials, and engagement with the healthcare system. Reducing these hurdles has been a focus in healthcare delivery, as they significantly hinder patient engagement and adherence to treatments. The growing use of large language models (LLMs) in healthcare opens the possibility to reduce these linguistic hurdles. This study evaluates the ability of five prominent LLMs—GPT-3.5, GPT-4, GPT-4o, LLaMA-3, and Mistral—to simplify healthcare information to the standard recommended by the American Journal of Medicine. Our results indicate that while LLMs can approximate targeted reading levels, their outputs are inconsistent, with significant variability in reading level and deviation from the topic, making them unsuitable for deployment in healthcare settings.

21 August 2024 in PsyArXiv
Verified
Preprint
Open access
KIKelsey Isman
Kelsey Isman
SGSalvatore Giorgi
Salvatore Giorgi
JEJennifer Ellis
Jennifer Ellis
AHAndrew S. Huhn
Andrew S. Huhn
TLTingting Liu
Tingting Liu
BCBrenda Curtis
Brenda Curtis
Perceived stigma (the perception of public attitudes towards members of a marginalized group) toward people who use substances hinders treatment seeking, impacts the quality of care received, and affects treatment success. Despite its relevance to treatment, research on stigma’s association with SUD-related outcomes is limited. This study aims to examine the associations between perceived stigma, types of substances used, mental health, and treatment type in a large sample across treatment sites. The present study involved survey completion upon treatment intake and early in treatment. Participants were individuals receiving treatment for substance use disorder (SUD; N=7591) in 75 treatment centers across the United States. Participants responded to the Perceived Stigma of Substance Abuse Scale (PSSAS). Participants’ primary substances were collected. Participants completed mental health (anxiety, depression, stress), quality of life, and substance use-related surveys (craving and pursuit of abstinence). Finally, the treatment type received by each patient was reported. Reporting primarily using alcohol predicted lower PSSAS scores, while primarily using benzodiazepines, cocaine, heroin, methamphetamines, and opioids predicted higher PSSAS scores. Higher PSSAS scores also predicted higher levels of anxiety, depression, stress, and craving and lower levels of quality of life and pursuit of abstinence. Receiving intensive outpatient treatment predicted lower PSSAS scores. SUD patients report different levels of stigma depending on the substance they use and the type of treatment they receive. The differences in stigma levels predict distinctions in mental health outcomes. These findings highlight the need for tailored stigma-reduction interventions for those receiving SUD treatment. This is a preprint and, thus, not peer reviewed.