Robert Michael Lundin
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Robert Michael Lundin has a total of 165 co-authors. Here are the top co-authors by number of shared publications:
Ben Sharif
Cardiff University Research Society (CUReS), Cardiff University School of Medicine, Cardiff, United Kingdom
3 shared publicationsCaity Morrison
3 shared publicationsCraig Harvey
Mental Health Drugs and Alcohol Service (MHDAS), Barwon Health, Geelong, Australia
3 shared publications- Tamas Szakmany
Department of Anaesthesia, Intensive Care and Pain Medicine, Division of Population Medicine, Cardiff University, Cardiff, UK
3 shared publications Abbas Khosravi
Institute for Intelligent Systems Research and Innovation (IISRI), Deakin University, Geelong, VIC, Australia
2 shared publications- Ajeet B. Singh
Institute for Mental and Physical Health and Clinical Translation (IMPACT), School of Medicine, Deakin University, Melbourne, VIC, Australia
2 shared publications
Latest Publications
- 22 October 2026
- Published by Elsevier in International Journal of Drug Policy
- Vol. 158
Perceptions of risk among GHB users: A rapid qualitative evidence synthesis
- 22 October 2026
- Published by Elsevier in International Journal of Drug Policy
- Vol. 158
Perceptions of risk among GHB users: A rapid qualitative evidence synthesis
- Open Access
- 1 October 2026
- Published by Wiley in Drug and Alcohol Review
- Vol. 45 (7) , e70272
Evaluating Harm Reduction Advice From Large Language Models
Introduction: Large language models (LLM) and generative Artificial Intelligence are increasingly providing health information. We have applied a novel method to evaluate the harm reduction information generated by popular LLMs. We focused on the filtration of oral medications for injection and the availability of local harm reduction services. Methods: We designed a scoring rubric to evaluate the accuracy of the information provided by each LLM and applied it to four unique, commonly used LLMs. Results: Gemini 2.5 Flash (87%) and ChatGPT‐4o (77%) scored higher than DeepSeek R1 Distill Llama 3.3 70B (22%) and Llama 4 (18%). Overall, both Gemini and ChatGPT‐4o gave comprehensive and accurate instructions for the use of micron filters for drug injection, while neither DeepSeek R1 nor Llama 4 provided instructions on their use. Gemini 2.5 and ChatGPT‐4o both scored highly for harm reduction general information and situational questions, while DeepSeek R1 and Llama 4 provided limited to no responses. Gemini 2.5 and ChatGPT‐4o both provided accurate, location‐specific information on local harm reduction services. Specific models also generated misinformation, further affirming the need for in‐person, nuanced discussions with harm reduction specialists. Discussion and Conclusions: In this study, we presented a novel method for evaluating LLMs' potential effectiveness for harm reduction across common techniques and situations. Most of the tested LLMs provided reasonable information in a non‐judgmental manner, but they also provided misinformation, and some wouldn't answer questions due to built‐in safety parameters. The developed scoring rubric can be easily adapted to other harm reduction topics for ongoing evaluation.
- 1 October 2026
- Published by Wiley in Drug and Alcohol Review
- Vol. 45 (7) , e70272
Open AccessEvaluating Harm Reduction Advice From Large Language Models
Introduction: Large language models (LLM) and generative Artificial Intelligence are increasingly providing health information. We have applied a novel method to evaluate the harm reduction information generated by popular LLMs. We focused on the filtration of oral medications for injection and the availability of local harm reduction services. Methods: We designed a scoring rubric to evaluate the accuracy of the information provided by each LLM and applied it to four unique, commonly used LLMs. Results: Gemini 2.5 Flash (87%) and ChatGPT‐4o (77%) scored higher than DeepSeek R1 Distill Llama 3.3 70B (22%) and Llama 4 (18%). Overall, both Gemini and ChatGPT‐4o gave comprehensive and accurate instructions for the use of micron filters for drug injection, while neither DeepSeek R1 nor Llama 4 provided instructions on their use. Gemini 2.5 and ChatGPT‐4o both scored highly for harm reduction general information and situational questions, while DeepSeek R1 and Llama 4 provided limited to no responses. Gemini 2.5 and ChatGPT‐4o both provided accurate, location‐specific information on local harm reduction services. Specific models also generated misinformation, further affirming the need for in‐person, nuanced discussions with harm reduction specialists. Discussion and Conclusions: In this study, we presented a novel method for evaluating LLMs' potential effectiveness for harm reduction across common techniques and situations. Most of the tested LLMs provided reasonable information in a non‐judgmental manner, but they also provided misinformation, and some wouldn't answer questions due to built‐in safety parameters. The developed scoring rubric can be easily adapted to other harm reduction topics for ongoing evaluation.
- 9 June 2026
- Published by SAGE Publications in Australasian Psychiatry
Recognising and treating ADHD in adults in public mental health services: A scoping review of key concepts
Objectives: To summarise service models, facilitators and barriers relevant to recognising and treating attention deficit/hyperactivity disorder (ADHD) within adult public mental health services. Methods: A scoping review following PRISMA-ScR guidelines examined international literature describing adult ADHD care delivered within public mental health systems. Data were charted on models of care, implementation facilitators and barriers, and evidence gaps were identified. Results: Thirteen sources met inclusion criteria. The literature was limited and largely conceptual, consisting predominantly of editorials, commentaries and clinical guidance, with minimal empirical evaluation of implementation or outcomes. Three broad approaches to service delivery were described: embedding ADHD assessment and treatment within existing adult mental health teams; establishing specialist adult ADHD or neurodevelopmental services; and shared-care and primary-care–led models supported by specialist input. Facilitators included targeted workforce training and collaboration with primary care services. Barriers related to diagnostic complexity, clinician confidence, service capacity and concerns regarding prioritisation within public systems. Conclusions: The literature consistently supports embedding ADHD care within routine adult public mental health teams, supported by targeted registrar and psychiatrist training and structured collaboration with general practitioners. Empirical evaluation of service models is needed to inform sustainable and equitable implementation within publicly funded mental health services.
- 9 June 2026
- Published by SAGE Publications in Australasian Psychiatry
Recognising and treating ADHD in adults in public mental health services: A scoping review of key concepts
Objectives: To summarise service models, facilitators and barriers relevant to recognising and treating attention deficit/hyperactivity disorder (ADHD) within adult public mental health services. Methods: A scoping review following PRISMA-ScR guidelines examined international literature describing adult ADHD care delivered within public mental health systems. Data were charted on models of care, implementation facilitators and barriers, and evidence gaps were identified. Results: Thirteen sources met inclusion criteria. The literature was limited and largely conceptual, consisting predominantly of editorials, commentaries and clinical guidance, with minimal empirical evaluation of implementation or outcomes. Three broad approaches to service delivery were described: embedding ADHD assessment and treatment within existing adult mental health teams; establishing specialist adult ADHD or neurodevelopmental services; and shared-care and primary-care–led models supported by specialist input. Facilitators included targeted workforce training and collaboration with primary care services. Barriers related to diagnostic complexity, clinician confidence, service capacity and concerns regarding prioritisation within public systems. Conclusions: The literature consistently supports embedding ADHD care within routine adult public mental health teams, supported by targeted registrar and psychiatrist training and structured collaboration with general practitioners. Empirical evaluation of service models is needed to inform sustainable and equitable implementation within publicly funded mental health services.
- Open Access
- 14 April 2026
- Published by Bond University in Australian Journal of Clinical Education
- Vol. 15 (1)
Evaluating Paid Positions for Medical Students in Australia: An Online Survey
Objectives: This study evaluates the impact of remunerated medical student (RMS) positions introduced during the COVID-19 pandemic, focusing on final-year students and recent graduates participating since January 2022. Methods: An anonymous, cross-sectional online survey using Likert scales (1–5) was distributed across Australian medical schools between March and September 2024. A total of 890 students participated: 141 RMS and 749 non-RMS. Recruitment occurred via social media, emails, and notice boards. The survey explored RMS recruitment, support, job satisfaction, financial compensation, and perceived effects on clinical skills and career preparation. It also examined non-RMS students’ views on the program’s impact on education and equity. Results: RMS students reported high job satisfaction (mean = 4.46) and strong preparedness for hospital roles (mean = 4.64). Metropolitan students viewed RMS more positively than rural peers. Nearly half (47.6%) of RMS students reported scheduling conflicts, and 22.6% experienced discrimination, harassment, or bullying. Among non-RMS students, 61.3% reported differential treatment during placements, though 88% supported the program. Conclusion: RMS roles enhance clinical skills and hospital readiness, with strong support from RMS and non-RMS students. The program should be retained and expanded, improving rural access, scheduling flexibility, and support. Further research should assess long-term outcomes.
- 14 April 2026
- Published by Bond University in Australian Journal of Clinical Education
- Vol. 15 (1)
Open AccessEvaluating Paid Positions for Medical Students in Australia: An Online Survey
Objectives: This study evaluates the impact of remunerated medical student (RMS) positions introduced during the COVID-19 pandemic, focusing on final-year students and recent graduates participating since January 2022. Methods: An anonymous, cross-sectional online survey using Likert scales (1–5) was distributed across Australian medical schools between March and September 2024. A total of 890 students participated: 141 RMS and 749 non-RMS. Recruitment occurred via social media, emails, and notice boards. The survey explored RMS recruitment, support, job satisfaction, financial compensation, and perceived effects on clinical skills and career preparation. It also examined non-RMS students’ views on the program’s impact on education and equity. Results: RMS students reported high job satisfaction (mean = 4.46) and strong preparedness for hospital roles (mean = 4.64). Metropolitan students viewed RMS more positively than rural peers. Nearly half (47.6%) of RMS students reported scheduling conflicts, and 22.6% experienced discrimination, harassment, or bullying. Among non-RMS students, 61.3% reported differential treatment during placements, though 88% supported the program. Conclusion: RMS roles enhance clinical skills and hospital readiness, with strong support from RMS and non-RMS students. The program should be retained and expanded, improving rural access, scheduling flexibility, and support. Further research should assess long-term outcomes.
- Open Access
- 6 October 2025
- Published by Springer Nature in Harm Reduction Journal
- Vol. 22 (1) , 1-44
Harm reduction approaches for the use of benzodiazepines: a scoping review
Benzodiazepines are widely prescribed but are associated with significant risks, particularly when used long-term. The anxiolytic and hypnotic properties of these medications increase their risk of dependence, which can lead to nonmedical and illicit use. Illicit use further compounds these harms, particularly with the emergence of potent novel benzodiazepines on the unregulated market. While tapering remains the standard treatment, not all individuals seek discontinuation. In such cases, harm reduction becomes a key approach to minimise associated risks. This review aimed to identify and synthesise existing harm reduction approaches for people using benzodiazepines. A systematic search was conducted across four databases, PsycINFO (n = 183), MEDLINE (n = 345), Web of Science (n = 382), and Embase (n = 940), following the PRISMA guidelines. Searches were carried out between February 14 and March 30, 2024, using terms related to harm reduction and benzodiazepines. The search was re-run on July 7, 2025, using the same strategy across all four databases. Thirty-five studies were included and grouped into the following themes: direct interventions (n = 16), policy approaches (n = 9), and population-specific approaches (n = 10). Among direct interventions, drug checking was the most frequently reported approach, with advanced techniques improving the detection of novel benzodiazepines and prompting safer use practices. Benzodiazepine agonist prescribing during the COVID-19 pandemic has yielded positive outcomes; conversely, policy responses such as rescheduling and prescribing changes indicated mixed results. While some studies reported reduced use and improved treatment engagement, others highlighted unintended consequences that may displace or exacerbate harm. Specific populations, such as young people, those who inject benzodiazepines, and members of online communities, highlight the diverse demographics of people who use benzodiazepines and emphasise the importance of developing tailored responses to address unique needs. Drug checking emerged as the most widely reported harm reduction approach for benzodiazepine use, with consistent positive outcomes across studies. Prescribing and policy interventions demonstrated variable impacts, often influenced by broader systemic factors. Critically, a clear gap remains in harm reduction approaches for those not seeking treatment, highlighting a need for inclusive, flexible and pragmatic responses. There is also a need for more robust evaluation of harm reduction interventions to strengthen the evidence base and inform practice.
- 6 October 2025
- Published by Springer Nature in Harm Reduction Journal
- Vol. 22 (1) , 1-44
Open AccessHarm reduction approaches for the use of benzodiazepines: a scoping review
Benzodiazepines are widely prescribed but are associated with significant risks, particularly when used long-term. The anxiolytic and hypnotic properties of these medications increase their risk of dependence, which can lead to nonmedical and illicit use. Illicit use further compounds these harms, particularly with the emergence of potent novel benzodiazepines on the unregulated market. While tapering remains the standard treatment, not all individuals seek discontinuation. In such cases, harm reduction becomes a key approach to minimise associated risks. This review aimed to identify and synthesise existing harm reduction approaches for people using benzodiazepines. A systematic search was conducted across four databases, PsycINFO (n = 183), MEDLINE (n = 345), Web of Science (n = 382), and Embase (n = 940), following the PRISMA guidelines. Searches were carried out between February 14 and March 30, 2024, using terms related to harm reduction and benzodiazepines. The search was re-run on July 7, 2025, using the same strategy across all four databases. Thirty-five studies were included and grouped into the following themes: direct interventions (n = 16), policy approaches (n = 9), and population-specific approaches (n = 10). Among direct interventions, drug checking was the most frequently reported approach, with advanced techniques improving the detection of novel benzodiazepines and prompting safer use practices. Benzodiazepine agonist prescribing during the COVID-19 pandemic has yielded positive outcomes; conversely, policy responses such as rescheduling and prescribing changes indicated mixed results. While some studies reported reduced use and improved treatment engagement, others highlighted unintended consequences that may displace or exacerbate harm. Specific populations, such as young people, those who inject benzodiazepines, and members of online communities, highlight the diverse demographics of people who use benzodiazepines and emphasise the importance of developing tailored responses to address unique needs. Drug checking emerged as the most widely reported harm reduction approach for benzodiazepine use, with consistent positive outcomes across studies. Prescribing and policy interventions demonstrated variable impacts, often influenced by broader systemic factors. Critically, a clear gap remains in harm reduction approaches for those not seeking treatment, highlighting a need for inclusive, flexible and pragmatic responses. There is also a need for more robust evaluation of harm reduction interventions to strengthen the evidence base and inform practice.
- Open Access
- 29 August 2025
- Published by Frontiers Media SA in Frontiers in Medicine
- Vol. 12, 1512703
A review of individual multidisciplinary team roles in alcohol and other drugs
Alcohol and Other Drugs (AOD) usage remains high globally and in Australia and results in a high amount of morbidity and mortality in our community. The Australian AOD workforce comprises a highly diverse range of healthcare professionals. Since medical doctors and AOD roles without specific credentials have already been covered in the literature, this review summarizes on the other direct patient facing roles in the AOD multidisciplinary team. We focused on harm reduction workers, peer workers, social workers, pharmacists, occupational therapists, psychologists, nurses and nurse practitioners. The findings of this review show that previous non-specific roles such as peer and harm reduction workers are developing into their distinct disciplines with their own specific training. Experienced clinicians in their field such as nurse practitioners and senior pharmacists are given increasingly more scope of practice that aligns with medical practitioners. However, there is still limited research to some established disciplines such as the role occupational therapy and social work. Future research should focus on how these roles can be fully utilized in the AOD multidisciplinary team.
- 29 August 2025
- Published by Frontiers Media SA in Frontiers in Medicine
- Vol. 12, 1512703
Open AccessA review of individual multidisciplinary team roles in alcohol and other drugs
Alcohol and Other Drugs (AOD) usage remains high globally and in Australia and results in a high amount of morbidity and mortality in our community. The Australian AOD workforce comprises a highly diverse range of healthcare professionals. Since medical doctors and AOD roles without specific credentials have already been covered in the literature, this review summarizes on the other direct patient facing roles in the AOD multidisciplinary team. We focused on harm reduction workers, peer workers, social workers, pharmacists, occupational therapists, psychologists, nurses and nurse practitioners. The findings of this review show that previous non-specific roles such as peer and harm reduction workers are developing into their distinct disciplines with their own specific training. Experienced clinicians in their field such as nurse practitioners and senior pharmacists are given increasingly more scope of practice that aligns with medical practitioners. However, there is still limited research to some established disciplines such as the role occupational therapy and social work. Future research should focus on how these roles can be fully utilized in the AOD multidisciplinary team.
- Open Access
Predicting the need for electroconvulsive therapy via machine learning trained on electronic health record data
Objective: Electroconvulsive therapy (ECT) is an effective treatment of severe manifestations of mental illness. Since delay in initiation of ECT can have detrimental effects, prediction of the need for ECT could improve outcomes via more timely treatment initiation. This study aimed to predict the need for ECT following admission to a psychiatric hospital. Methods: This cohort study was based on electronic health record (EHR) data from routine clinical practice. Adult patients admitted to a hospital within the Psychiatric Services of the Central Denmark Region between January 2013 and November 2021 were included in the study. The outcome was initiation of ECT >7 days (to not include patients admitted for planned ECT) and ≤67 days after admission. The data was randomly split into an 85% training set and a 15% test set. On the 7 th day of the inpatient stay, machine learning models (extreme gradient boosting (XGBoost)) were trained to predict initiation of ECT and subsequently tested on the test set, using the area under the receiver operating characteristic curve (AUROC) as the main performance measure. Results: The cohort consisted of 41,610 patients with 164,961 admissions eligible for prediction. In the test set the trained model predicted ECT initiation with an AUROC of 0.94, 47% sensitivity, 98% specificity, positive predictive value of 24% and negative predictive value of 99%. The top predictors were the highest suicide assessment score and the mean Brøset violence checklist score in the preceding three months. Conclusions: EHR data from routine clinical practice can be used to predict need for ECT. This may lead to more timely treatment initiation.
Open AccessPredicting the need for electroconvulsive therapy via machine learning trained on electronic health record data
Objective: Electroconvulsive therapy (ECT) is an effective treatment of severe manifestations of mental illness. Since delay in initiation of ECT can have detrimental effects, prediction of the need for ECT could improve outcomes via more timely treatment initiation. This study aimed to predict the need for ECT following admission to a psychiatric hospital. Methods: This cohort study was based on electronic health record (EHR) data from routine clinical practice. Adult patients admitted to a hospital within the Psychiatric Services of the Central Denmark Region between January 2013 and November 2021 were included in the study. The outcome was initiation of ECT >7 days (to not include patients admitted for planned ECT) and ≤67 days after admission. The data was randomly split into an 85% training set and a 15% test set. On the 7 th day of the inpatient stay, machine learning models (extreme gradient boosting (XGBoost)) were trained to predict initiation of ECT and subsequently tested on the test set, using the area under the receiver operating characteristic curve (AUROC) as the main performance measure. Results: The cohort consisted of 41,610 patients with 164,961 admissions eligible for prediction. In the test set the trained model predicted ECT initiation with an AUROC of 0.94, 47% sensitivity, 98% specificity, positive predictive value of 24% and negative predictive value of 99%. The top predictors were the highest suicide assessment score and the mean Brøset violence checklist score in the preceding three months. Conclusions: EHR data from routine clinical practice can be used to predict need for ECT. This may lead to more timely treatment initiation.
- Open Access
- 17 March 2025
- Published by SAGE Publications in Australian & New Zealand Journal of Psychiatry
- Vol. 59 (5) , 423-432
It is time to reassess reporting of electroconvulsive therapy data in New Zealand: A 17-year retrospective analysis of treatment data from Waikato
Objective: The New Zealand Government has provided brief annual reports on electroconvulsive therapy treatment since 2004. Despite this, only limited information is made available to clinicians to guide clinical improvement and refine guidelines. Beyond an audit from Otago detailing 10 years of electroconvulsive therapy treatments, limited information is available about electroconvulsive therapy treatments in New Zealand. This paper reports on the use of electroconvulsive therapy over the past 17 years in one New Zealand District Health Board. Methods: It covers 7126 treatments for 333 patients between 2004 and 2020. Results: Despite an increasing number of treatments, there has been no per capita growth when corrected for the population. Despite criticism for the disproportionate use of electroconvulsive therapy in women, treatment equity for men and women has been evidenced over the latter 3 years. The majority of treatments were given under voluntary consent, even among patients admitted under the Mental Health Act. Clinical practice is moving towards bifrontal treatment over other electrode placements in response to clinical guidance. Conclusion: While COVID-19 had broad impacts across healthcare services, it has not led to an overall change in treatments. There was, however, a noticeable shift towards more psychotic disorders treated during the pandemic. This study also provides data that Māori and Pacific Islanders are accessing electroconvulsive therapy, though rates of electroconvulsive therapy usage are still lesser on a proportional population basis. Conclusions are provided to improve national data collection and reporting standards.
- 17 March 2025
- Published by SAGE Publications in Australian & New Zealand Journal of Psychiatry
- Vol. 59 (5) , 423-432
Open AccessIt is time to reassess reporting of electroconvulsive therapy data in New Zealand: A 17-year retrospective analysis of treatment data from Waikato
Objective: The New Zealand Government has provided brief annual reports on electroconvulsive therapy treatment since 2004. Despite this, only limited information is made available to clinicians to guide clinical improvement and refine guidelines. Beyond an audit from Otago detailing 10 years of electroconvulsive therapy treatments, limited information is available about electroconvulsive therapy treatments in New Zealand. This paper reports on the use of electroconvulsive therapy over the past 17 years in one New Zealand District Health Board. Methods: It covers 7126 treatments for 333 patients between 2004 and 2020. Results: Despite an increasing number of treatments, there has been no per capita growth when corrected for the population. Despite criticism for the disproportionate use of electroconvulsive therapy in women, treatment equity for men and women has been evidenced over the latter 3 years. The majority of treatments were given under voluntary consent, even among patients admitted under the Mental Health Act. Clinical practice is moving towards bifrontal treatment over other electrode placements in response to clinical guidance. Conclusion: While COVID-19 had broad impacts across healthcare services, it has not led to an overall change in treatments. There was, however, a noticeable shift towards more psychotic disorders treated during the pandemic. This study also provides data that Māori and Pacific Islanders are accessing electroconvulsive therapy, though rates of electroconvulsive therapy usage are still lesser on a proportional population basis. Conclusions are provided to improve national data collection and reporting standards.
- Top Cited PapersOpen Access
- 11 November 2024
- Published by Cambridge University Press (CUP) in Acta Neuropsychiatrica
- Vol. 37, 1-14
Use of generative artificial intelligence (AI) in psychiatry and mental health care: a systematic review
Objectives: Tools based on generative artificial intelligence (AI) such as ChatGPT have the potential to transform modern society, including the field of medicine. Due to the prominent role of language in psychiatry, e.g., for diagnostic assessment and psychotherapy, these tools may be particularly useful within this medical field. Therefore, the aim of this study was to systematically review the literature on generative AI applications in psychiatry and mental health. Methods: We conducted a systematic review following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The search was conducted across three databases, and the resulting articles were screened independently by two researchers. The content, themes, and findings of the articles were qualitatively assessed. Results: The search and screening process resulted in the inclusion of 40 studies. The median year of publication was 2023. The themes covered in the articles were mainly mental health and well-being in general – with less emphasis on specific mental disorders (substance use disorder being the most prevalent). The majority of studies were conducted as prompt experiments, with the remaining studies comprising surveys, pilot studies, and case reports. Most studies focused on models that generate language, ChatGPT in particular. Conclusions: Generative AI in psychiatry and mental health is a nascent but quickly expanding field. The literature mainly focuses on applications of ChatGPT, and finds that generative AI performs well, but notes that it is limited by significant safety and ethical concerns. Future research should strive to enhance transparency of methods, use experimental designs, ensure clinical relevance, and involve users/patients in the design phase.
- 11 November 2024
- Published by Cambridge University Press (CUP) in Acta Neuropsychiatrica
- Vol. 37, 1-14
Top Cited PapersOpen AccessUse of generative artificial intelligence (AI) in psychiatry and mental health care: a systematic review
Objectives: Tools based on generative artificial intelligence (AI) such as ChatGPT have the potential to transform modern society, including the field of medicine. Due to the prominent role of language in psychiatry, e.g., for diagnostic assessment and psychotherapy, these tools may be particularly useful within this medical field. Therefore, the aim of this study was to systematically review the literature on generative AI applications in psychiatry and mental health. Methods: We conducted a systematic review following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The search was conducted across three databases, and the resulting articles were screened independently by two researchers. The content, themes, and findings of the articles were qualitatively assessed. Results: The search and screening process resulted in the inclusion of 40 studies. The median year of publication was 2023. The themes covered in the articles were mainly mental health and well-being in general – with less emphasis on specific mental disorders (substance use disorder being the most prevalent). The majority of studies were conducted as prompt experiments, with the remaining studies comprising surveys, pilot studies, and case reports. Most studies focused on models that generate language, ChatGPT in particular. Conclusions: Generative AI in psychiatry and mental health is a nascent but quickly expanding field. The literature mainly focuses on applications of ChatGPT, and finds that generative AI performs well, but notes that it is limited by significant safety and ethical concerns. Future research should strive to enhance transparency of methods, use experimental designs, ensure clinical relevance, and involve users/patients in the design phase.
- 1 October 2024
- Published by Wolters Kluwer Health in The Journal of ECT
- Vol. 41 (2) , e21-e22
Potential Pitfalls From Long-Acting Injectable Buprenorphine During ECT Treatment
- 1 October 2024
- Published by Wolters Kluwer Health in The Journal of ECT
- Vol. 41 (2) , e21-e22
Potential Pitfalls From Long-Acting Injectable Buprenorphine During ECT Treatment