Marcus Young
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Marcus Young has a total of 31 co-authors. Here are the top co-authors by number of shared publications:
- Rinaldo Bellomo
Department of Intensive Care, Austin Hospital, Melbourne, VIC, Australia
4 shared publications - Ary Serpa Neto
Department of Intensive Care, Austin Hospital, Heidelberg, Australia
3 shared publications - Andrew Casamento
Department of Intensive Care, Austin Hospital, Melbourne, VIC, Australia
2 shared publications Angajendra Ghosh
Department of Intensive Care, Northern Hospital, Melbourne, Australia; Department of Critical Care, University of Melbourne, Melbourne, Australia; Department of Medical Education, University of Melbourne, Melbourne, Australia
2 shared publicationsChristina Taplin
Department of Intensive Care, Northern Hospital, Melbourne, Australia
2 shared publicationsMervin Lawrence
Department of Intensive Care, Northern Hospital, Melbourne, Australia
2 shared publications
Latest Publications
- Open Access
- 6 September 2026
- Published by Health Libraries Australia in Journal of Health Information and Libraries Australasia
- Vol. 6 (2) , 12-21
The ELIZA Effect
Artificial Intelligence models can be informative; they can provide insight and accelerate scientific discovery. However, they can also hallucinate, reproduce biases from their training data, manufacture spurious associations, and they remain susceptible to malicious corruption. Critically, contemporary AI models lack the capacity to reliably and reproducibly distinguish such biased, malicious or counterfactual training data from legitimate data. Moreover, they do not have the capability to reliably validate their outputs. The curation of their training data and the validation of their outputs therefore remain the responsibility of their human operators.
- 6 September 2026
- Published by Health Libraries Australia in Journal of Health Information and Libraries Australasia
- Vol. 6 (2) , 12-21
Open AccessThe ELIZA Effect
Artificial Intelligence models can be informative; they can provide insight and accelerate scientific discovery. However, they can also hallucinate, reproduce biases from their training data, manufacture spurious associations, and they remain susceptible to malicious corruption. Critically, contemporary AI models lack the capacity to reliably and reproducibly distinguish such biased, malicious or counterfactual training data from legitimate data. Moreover, they do not have the capability to reliably validate their outputs. The curation of their training data and the validation of their outputs therefore remain the responsibility of their human operators.
- Open Access
- 5 May 2025
- Published by Springer Nature in Intensive Care Medicine
- Vol. 51 (5) , 977
Delirium in ICU and Macbeth’s dagger: author’s reply
- 5 May 2025
- Published by Springer Nature in Intensive Care Medicine
- Vol. 51 (5) , 977
Open AccessDelirium in ICU and Macbeth’s dagger: author’s reply
- Open Access
- 14 March 2025
- Published by Associação de Medicina Intensiva Brasileira in Critical Care Science
- Vol. 37, e20250010
A target trial emulation of dexmedetomidine to treat agitation in the intensive care unit
Objective Agitation is a major problem in the intensive care unit. However, no treatment has clearly emerged as effective and safe. Using target trial emulation, we aimed to test the hypothesis that early intervention with dexmedetomidine would accelerate agitation resolution. Methods We read clinical notes in an electronic medical records system with natural language […] ABSTRACT Objective Agitation is a major problem in the intensive care unit. However, no treatment has clearly emerged as effective and safe. Using target trial emulation, we aimed to test the hypothesis that early intervention with dexmedetomidine would accelerate agitation resolution. Methods We read clinical notes in an electronic medical records system with natural language […]
- 14 March 2025
- Published by Associação de Medicina Intensiva Brasileira in Critical Care Science
- Vol. 37, e20250010
Open AccessA target trial emulation of dexmedetomidine to treat agitation in the intensive care unit
Objective Agitation is a major problem in the intensive care unit. However, no treatment has clearly emerged as effective and safe. Using target trial emulation, we aimed to test the hypothesis that early intervention with dexmedetomidine would accelerate agitation resolution. Methods We read clinical notes in an electronic medical records system with natural language […] ABSTRACT Objective Agitation is a major problem in the intensive care unit. However, no treatment has clearly emerged as effective and safe. Using target trial emulation, we aimed to test the hypothesis that early intervention with dexmedetomidine would accelerate agitation resolution. Methods We read clinical notes in an electronic medical records system with natural language […]
- Open Access
- 4 November 2023
- Published by Springer Nature in Critical Care
- Vol. 27 (1) , 1-11
Natural language processing diagnosed behavioural disturbance phenotypes in the intensive care unit: characteristics, prevalence, trajectory, treatment, and outcomes
Background: Natural language processing (NLP) may help evaluate the characteristics, prevalence, trajectory, treatment, and outcomes of behavioural disturbance phenotypes in critically ill patients. Methods: We obtained electronic clinical notes, demographic information, outcomes, and treatment data from three medical-surgical ICUs. Using NLP, we screened for behavioural disturbance phenotypes based on words suggestive of an agitated state, a non-agitated state, or a combination of both. Results: We studied 2931 patients. Of these, 225 (7.7%) were NLP-Dx-BD positive for the agitated phenotype, 544 (18.6%) for the non-agitated phenotype and 667 (22.7%) for the combined phenotype. Patients with these phenotypes carried multiple clinical baseline differences. On time-dependent multivariable analysis to compensate for immortal time bias and after adjustment for key outcome predictors, agitated phenotype patients were more likely to receive antipsychotic medications (odds ratio [OR] 1.84, 1.35–2.51, p < 0.001) compared to non-agitated phenotype patients but not compared to combined phenotype patients (OR 1.27, 0.86–1.89, p = 0.229). Moreover, agitated phenotype patients were more likely to die than other phenotypes patients (OR 1.57, 1.10–2.25, p = 0.012 vs non-agitated phenotype; OR 4.61, 2.14–9.90, p < 0.001 vs. combined phenotype). This association was strongest in patients receiving mechanical ventilation when compared with the combined phenotype (OR 7.03, 2.07–23.79, p = 0.002). A similar increased risk was also seen for patients with the non-agitated phenotype compared with the combined phenotype (OR 6.10, 1.80–20.64, p = 0.004). Conclusions: NLP-Dx-BD screening enabled identification of three behavioural disturbance phenotypes with different characteristics, prevalence, trajectory, treatment, and outcome. Such phenotype identification appears relevant to prognostication and trial design. Graphical
- 4 November 2023
- Published by Springer Nature in Critical Care
- Vol. 27 (1) , 1-11
Open AccessNatural language processing diagnosed behavioural disturbance phenotypes in the intensive care unit: characteristics, prevalence, trajectory, treatment, and outcomes
Background: Natural language processing (NLP) may help evaluate the characteristics, prevalence, trajectory, treatment, and outcomes of behavioural disturbance phenotypes in critically ill patients. Methods: We obtained electronic clinical notes, demographic information, outcomes, and treatment data from three medical-surgical ICUs. Using NLP, we screened for behavioural disturbance phenotypes based on words suggestive of an agitated state, a non-agitated state, or a combination of both. Results: We studied 2931 patients. Of these, 225 (7.7%) were NLP-Dx-BD positive for the agitated phenotype, 544 (18.6%) for the non-agitated phenotype and 667 (22.7%) for the combined phenotype. Patients with these phenotypes carried multiple clinical baseline differences. On time-dependent multivariable analysis to compensate for immortal time bias and after adjustment for key outcome predictors, agitated phenotype patients were more likely to receive antipsychotic medications (odds ratio [OR] 1.84, 1.35–2.51, p < 0.001) compared to non-agitated phenotype patients but not compared to combined phenotype patients (OR 1.27, 0.86–1.89, p = 0.229). Moreover, agitated phenotype patients were more likely to die than other phenotypes patients (OR 1.57, 1.10–2.25, p = 0.012 vs non-agitated phenotype; OR 4.61, 2.14–9.90, p < 0.001 vs. combined phenotype). This association was strongest in patients receiving mechanical ventilation when compared with the combined phenotype (OR 7.03, 2.07–23.79, p = 0.002). A similar increased risk was also seen for patients with the non-agitated phenotype compared with the combined phenotype (OR 6.10, 1.80–20.64, p = 0.004). Conclusions: NLP-Dx-BD screening enabled identification of three behavioural disturbance phenotypes with different characteristics, prevalence, trajectory, treatment, and outcome. Such phenotype identification appears relevant to prognostication and trial design. Graphical
- 11 August 2023
- Published by Elsevier in Australian Critical Care
- Vol. 37 (2) , 236-243
The effect of age on clinical dose equivalency of fentanyl and morphine analgosedation in mechanically ventilated patients: Findings from the ANALGESIC trial
- 11 August 2023
- Published by Elsevier in Australian Critical Care
- Vol. 37 (2) , 236-243
The effect of age on clinical dose equivalency of fentanyl and morphine analgosedation in mechanically ventilated patients: Findings from the ANALGESIC trial
- 23 May 2022
- Published by Springer Nature in Intensive Care Medicine
- Vol. 48 (7) , 983-984
Confusion regarding the use of Natural Language Processing in ICU delirium assessment. Author’s reply
- 23 May 2022
- Published by Springer Nature in Intensive Care Medicine
- Vol. 48 (7) , 983-984
Confusion regarding the use of Natural Language Processing in ICU delirium assessment. Author’s reply
- 1 December 2021
- Published by Oxford University Press (OUP) in American Journal of Respiratory and Critical Care Medicine
- Vol. 204 (11) , 1286-1294
A Phase II Cluster-Crossover Randomized Trial of Fentanyl versus Morphine for Analgosedation in Mechanically Ventilated Patients
Rationale: The continuous infusion of fentanyl or morphine is often prescribed to assist with analgesia and sedation (analgosedation) during mechanical ventilation. Objectives: To compare the effect of fentanyl versus morphine on patient-centered outcomes in ventilated patients. Methods: We conducted a cluster-randomized, cluster-crossover trial between July 2019 and August 2020 in two adult ICUs. We compared two continuous infusion regimens (fentanyl versus morphine). One ICU was randomized to the fentanyl–morphine sequence and the other to the morphine–fentanyl sequence. The primary outcome was the number of ventilator-free days at Day 28. Secondary outcomes included, among others, duration of mechanical ventilation in survivors and ICU-free days at Day 28. Measurements and Main Results: Via cluster allocation, we randomized 737 patients. Of these, 56 were excluded because of the opt-out consent process, leaving 681 (344 to fentanyl and 337 to morphine) for primary analysis (median [interquartile range] age, 59 [44–69] years). Median ventilator-free days at Day 28 were 26.1 (20.7–27.3) in the fentanyl versus 25.3 (19.1–27.2) in the morphine group (median difference, 0.79 [95% confidence interval, 0.31 to 1.28], P = 0.001). ICU-free days were greater (P < 0.001) and length of stay in the ICU for survivors shorter (P < 0.001) in the fentanyl group. All other secondary outcomes were not statistically different by treatment group. Conclusions: Among adult patients requiring mechanical ventilation, compared with morphine, fentanyl infusion significantly increased the median number of ventilator-free days at Day 28. The choice of opioid infusion agent may affect clinical outcomes and requires further investigation.
- 1 December 2021
- Published by Oxford University Press (OUP) in American Journal of Respiratory and Critical Care Medicine
- Vol. 204 (11) , 1286-1294
A Phase II Cluster-Crossover Randomized Trial of Fentanyl versus Morphine for Analgosedation in Mechanically Ventilated Patients
Rationale: The continuous infusion of fentanyl or morphine is often prescribed to assist with analgesia and sedation (analgosedation) during mechanical ventilation. Objectives: To compare the effect of fentanyl versus morphine on patient-centered outcomes in ventilated patients. Methods: We conducted a cluster-randomized, cluster-crossover trial between July 2019 and August 2020 in two adult ICUs. We compared two continuous infusion regimens (fentanyl versus morphine). One ICU was randomized to the fentanyl–morphine sequence and the other to the morphine–fentanyl sequence. The primary outcome was the number of ventilator-free days at Day 28. Secondary outcomes included, among others, duration of mechanical ventilation in survivors and ICU-free days at Day 28. Measurements and Main Results: Via cluster allocation, we randomized 737 patients. Of these, 56 were excluded because of the opt-out consent process, leaving 681 (344 to fentanyl and 337 to morphine) for primary analysis (median [interquartile range] age, 59 [44–69] years). Median ventilator-free days at Day 28 were 26.1 (20.7–27.3) in the fentanyl versus 25.3 (19.1–27.2) in the morphine group (median difference, 0.79 [95% confidence interval, 0.31 to 1.28], P = 0.001). ICU-free days were greater (P < 0.001) and length of stay in the ICU for survivors shorter (P < 0.001) in the fentanyl group. All other secondary outcomes were not statistically different by treatment group. Conclusions: Among adult patients requiring mechanical ventilation, compared with morphine, fentanyl infusion significantly increased the median number of ventilator-free days at Day 28. The choice of opioid infusion agent may affect clinical outcomes and requires further investigation.