Joanne Tang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Joanne Tang has a total of 61 co-authors. Here are the top co-authors by number of shared publications:
- Martin Hong
Department of Medical Oncology, Liverpool Hospital, Liverpool, NSW 2170, Australia
5 shared publications - Udit Nindra
Illawarra Cancer Centre, Wollongong Hospital, Wollongong, Australia
4 shared publications Wei Chua
Department of Medical Oncology, Liverpool Hospital , Liverpool
4 shared publications- Abhijit Pal
Department of Medical Oncology, Liverpool Hospital, Liverpool, Australia
3 shared publications Adam Cooper
Department of Medical Oncology, Liverpool Hospital , Liverpool
3 shared publicationsAndrew Killen
Scientia Clinical Research , Randwick
3 shared publications
Latest Publications
- Open Access
- 27 October 2026
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 28, e100531
Development and Pilot Evaluation of a Standardized and Accessible Rubric (OMNI) for Assessing Generative AI in Health Care Contexts: Four-Stage Mixed Methods Study
Background: The rapid adoption of generative AI in the community and health care necessitates reliable, human-led evaluation. Objective: This study aimed to develop a standardized rubric to support health care professionals, administrators, and researchers in evaluating generative AI outputs in health contexts. Methods: A rubric was developed through a 4-stage mixed methods study: literature synthesis, expert refinement, pilot application, and post-pilot feedback. The rubric was informed by 69 publications and refined through iterative feedback from a multidisciplinary expert panel (n=23). The pilot rubric was applied to 4 health care scenarios: responses to health-related Reddit questions, generation of health-related images and videos, responses to health professional continuing education questions, and automated physician-to-physician clinical documentation generation. Via these scenarios, 15 clinicians (medical practitioners, pharmacists, and medical oncologists) evaluated 250 input-output pairs using domain-level ratings and qualitative comments. The evaluated outputs included 200 AI-generated responses across the 4 health care scenarios and 50 physician-tagged replies in the health-related Reddit questions scenario. Results: The final rubric, OMNI, comprised 10 evaluation domains (Accuracy, Safety, Recognizes Uncertainty, Completeness, Communicates Clearly, Empathetic, Context Aware, Referencing, Follows Instructions, and Avoids Coerciveness) assessed on 4-point Likert scales, supported by detailed instructional and example materials. Via the 4 pilot scenarios, assessors generated 9600 domain-level ratings and 4010 qualitative comments. The mean pairwise agreement for the 4-point Likert groups ranged from 30.8% to 53.4%, increasing to 59.3% to 79.6% when collapsed into broad agreement categories; Gwet’s AC2 ranged from 0.49 to 0.80. OMNI differentiated strengths and weaknesses between models within scenarios, while recurrent limitations in Referencing, Recognizes Uncertainty, Completeness, Empathetic, and Accuracy were identified across scenarios. Multimodal image and video outputs had the highest occurrence of negative rating domains. Assessors reported OMNI was usable and indicated they would reuse it. Conclusions: OMNI is a standardized, accessible rubric for expert-led evaluation of generative AI outputs in health contexts. Pilot application demonstrated that OMNI could differentiate AI performance across models and domains while generating structured feedback. Further work is needed to assess its applicability to purpose-built health AI tools, agentic systems, and organizational evaluation settings.
- 27 October 2026
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 28, e100531
Open AccessDevelopment and Pilot Evaluation of a Standardized and Accessible Rubric (OMNI) for Assessing Generative AI in Health Care Contexts: Four-Stage Mixed Methods Study
Background: The rapid adoption of generative AI in the community and health care necessitates reliable, human-led evaluation. Objective: This study aimed to develop a standardized rubric to support health care professionals, administrators, and researchers in evaluating generative AI outputs in health contexts. Methods: A rubric was developed through a 4-stage mixed methods study: literature synthesis, expert refinement, pilot application, and post-pilot feedback. The rubric was informed by 69 publications and refined through iterative feedback from a multidisciplinary expert panel (n=23). The pilot rubric was applied to 4 health care scenarios: responses to health-related Reddit questions, generation of health-related images and videos, responses to health professional continuing education questions, and automated physician-to-physician clinical documentation generation. Via these scenarios, 15 clinicians (medical practitioners, pharmacists, and medical oncologists) evaluated 250 input-output pairs using domain-level ratings and qualitative comments. The evaluated outputs included 200 AI-generated responses across the 4 health care scenarios and 50 physician-tagged replies in the health-related Reddit questions scenario. Results: The final rubric, OMNI, comprised 10 evaluation domains (Accuracy, Safety, Recognizes Uncertainty, Completeness, Communicates Clearly, Empathetic, Context Aware, Referencing, Follows Instructions, and Avoids Coerciveness) assessed on 4-point Likert scales, supported by detailed instructional and example materials. Via the 4 pilot scenarios, assessors generated 9600 domain-level ratings and 4010 qualitative comments. The mean pairwise agreement for the 4-point Likert groups ranged from 30.8% to 53.4%, increasing to 59.3% to 79.6% when collapsed into broad agreement categories; Gwet’s AC2 ranged from 0.49 to 0.80. OMNI differentiated strengths and weaknesses between models within scenarios, while recurrent limitations in Referencing, Recognizes Uncertainty, Completeness, Empathetic, and Accuracy were identified across scenarios. Multimodal image and video outputs had the highest occurrence of negative rating domains. Assessors reported OMNI was usable and indicated they would reuse it. Conclusions: OMNI is a standardized, accessible rubric for expert-led evaluation of generative AI outputs in health contexts. Pilot application demonstrated that OMNI could differentiate AI performance across models and domains while generating structured feedback. Further work is needed to assess its applicability to purpose-built health AI tools, agentic systems, and organizational evaluation settings.
- Open Access
- 3 November 2025
- Published by Oxford University Press (OUP) in JNCI Cancer Spectrum
- Vol. 9 (6)
Prospective assessment of health-related quality of life in early phase oncology clinical trials: PEARLER
Introduction: Health-related quality of life (HRQoL) is not routine in early phase clinical trials (EP-CTs), which focus on dose-limiting toxicities and safety. However, for clinicians, understanding the impact of such trials on HRQoL is fundamental to consent patients, especially when the benefits on tumor response may be unknown. Aims and Methods: The PEARLER (Patient diversity And experience in eaRLy phase cancEr clinical tRials) study was conducted with a key aim of focusing on assessing HRQoL in participants undergoing EP-CTs using a multi-center prospective cohort setting. All participants completed a baseline demographic survey on Cycle 1 Day 1 with EORTC-QLQ-C30 on Day 1 of Cycles 1 through 6 or end of treatment (EoT). Results: Overall, 122 participants were recruited with median age 62. Median baseline Global Health Status (GHS) was 67 and remained unchanged throughout EP-CT (P = .188). GHS deterioration occurred in 29/122 (24%) while improvement occurred in 16/122 (13%). Median baseline Physical Function Score (PFS) was 87. PFS deterioration occurred in 30/122 (25%) while improvement occurred in 6/122 (5%). Baseline median CFS was 84. Cognitive Function Score (CFS) deterioration occurred in 25/122 (20%) while improvement occurred in 20/122 (16%). Baseline median Emotional Function Score (EFS) was 77. EFS deterioration occurred in 14/122 (11%) while improvement occurred in 14/122 (11%). Presence of liver metastases was a negative predictive marker for GHS, CFS, and EFS over time (P = .01, P < .01, and P < .01). Conclusion: PEARLER is the first prospective cohort study investigating change in HRQoL over time in patients undergoing EP-CTs. Reassuringly, almost three-quarters of participants who undertake EP-CTs either sustain or improve their GHS or PFS. Presence of liver metastases appears to be a negative predictive marker of HRQoL.
- 3 November 2025
- Published by Oxford University Press (OUP) in JNCI Cancer Spectrum
- Vol. 9 (6)
Open AccessProspective assessment of health-related quality of life in early phase oncology clinical trials: PEARLER
Introduction: Health-related quality of life (HRQoL) is not routine in early phase clinical trials (EP-CTs), which focus on dose-limiting toxicities and safety. However, for clinicians, understanding the impact of such trials on HRQoL is fundamental to consent patients, especially when the benefits on tumor response may be unknown. Aims and Methods: The PEARLER (Patient diversity And experience in eaRLy phase cancEr clinical tRials) study was conducted with a key aim of focusing on assessing HRQoL in participants undergoing EP-CTs using a multi-center prospective cohort setting. All participants completed a baseline demographic survey on Cycle 1 Day 1 with EORTC-QLQ-C30 on Day 1 of Cycles 1 through 6 or end of treatment (EoT). Results: Overall, 122 participants were recruited with median age 62. Median baseline Global Health Status (GHS) was 67 and remained unchanged throughout EP-CT (P = .188). GHS deterioration occurred in 29/122 (24%) while improvement occurred in 16/122 (13%). Median baseline Physical Function Score (PFS) was 87. PFS deterioration occurred in 30/122 (25%) while improvement occurred in 6/122 (5%). Baseline median CFS was 84. Cognitive Function Score (CFS) deterioration occurred in 25/122 (20%) while improvement occurred in 20/122 (16%). Baseline median Emotional Function Score (EFS) was 77. EFS deterioration occurred in 14/122 (11%) while improvement occurred in 14/122 (11%). Presence of liver metastases was a negative predictive marker for GHS, CFS, and EFS over time (P = .01, P < .01, and P < .01). Conclusion: PEARLER is the first prospective cohort study investigating change in HRQoL over time in patients undergoing EP-CTs. Reassuringly, almost three-quarters of participants who undertake EP-CTs either sustain or improve their GHS or PFS. Presence of liver metastases appears to be a negative predictive marker of HRQoL.
- 3 November 2025
- Published by American Society of Clinical Oncology (ASCO) in JCO Oncology Practice
- Vol. 22 (9) , 1616-1623
Assessing Patient-Reported Financial, Social, and Time Toxicity in Early-Phase Cancer Clinical Trials in Australia Through a Prospective Multicenter Nonrandomized Cohort Study
PURPOSE: Early-phase clinical trials (EP-CTs) provide access to novel therapeutics for patients with limited standard treatment options. Beyond drug-related risks, participants may experience hidden toxicities, including time, financial, and social burdens that often go unmeasured. METHODS: The Patient Experience in Early-Phase Cancer Clinical Trials (PEARLER) study aimed to assess time, financial, and social toxicity (ST) in a prospective, multicenter cohort. Participants completed a time toxicity (TT) survey and the European Organisation for Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire C30 on day 1 of cycles 1-6. Financial toxicity (FT) was assessed using the EORTC scoring manual; ST was evaluated using the social and role functioning domains. Multilevel models examined trends over time. RESULTS: A total of 122 participants (median age 62 years, range, 25-83) were enrolled. Median objective TT was 26%. Only 16% (20/122) reported subjective TT, most commonly during cycles 1-2 (85%). FT was reported by 44% (54/122), but did not significantly change over time ( P = .136). Although 71% (87/122) reported some reduction in social function, social functioning scores remained stable throughout trial participation. Role functioning declined in participants age 60 years and younger but was maintained in those older than 60 years. CONCLUSION: To our knowledge, PEARLER is the first prospective study to assess TT, FT, and ST in Australian EP-CTs. Although objective time burden was measurable, few patients perceived it as problematic beyond early cycles. FT and ST were commonly reported but remained stable over time. These findings suggest that EP-CT participation may not negatively affect patient-reported quality-of-life domains and provide reassurance for both clinicians and patients considering enrollment.
- 3 November 2025
- Published by American Society of Clinical Oncology (ASCO) in JCO Oncology Practice
- Vol. 22 (9) , 1616-1623
Assessing Patient-Reported Financial, Social, and Time Toxicity in Early-Phase Cancer Clinical Trials in Australia Through a Prospective Multicenter Nonrandomized Cohort Study
PURPOSE: Early-phase clinical trials (EP-CTs) provide access to novel therapeutics for patients with limited standard treatment options. Beyond drug-related risks, participants may experience hidden toxicities, including time, financial, and social burdens that often go unmeasured. METHODS: The Patient Experience in Early-Phase Cancer Clinical Trials (PEARLER) study aimed to assess time, financial, and social toxicity (ST) in a prospective, multicenter cohort. Participants completed a time toxicity (TT) survey and the European Organisation for Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire C30 on day 1 of cycles 1-6. Financial toxicity (FT) was assessed using the EORTC scoring manual; ST was evaluated using the social and role functioning domains. Multilevel models examined trends over time. RESULTS: A total of 122 participants (median age 62 years, range, 25-83) were enrolled. Median objective TT was 26%. Only 16% (20/122) reported subjective TT, most commonly during cycles 1-2 (85%). FT was reported by 44% (54/122), but did not significantly change over time ( P = .136). Although 71% (87/122) reported some reduction in social function, social functioning scores remained stable throughout trial participation. Role functioning declined in participants age 60 years and younger but was maintained in those older than 60 years. CONCLUSION: To our knowledge, PEARLER is the first prospective study to assess TT, FT, and ST in Australian EP-CTs. Although objective time burden was measurable, few patients perceived it as problematic beyond early cycles. FT and ST were commonly reported but remained stable over time. These findings suggest that EP-CT participation may not negatively affect patient-reported quality-of-life domains and provide reassurance for both clinicians and patients considering enrollment.
- 1 May 2025
- Published by American Society of Clinical Oncology (ASCO) in JCO Clinical Cancer Informatics
- Vol. 9 (9) , e2400323
Application of Generative Artificial Intelligence for Physician and Patient Oncology Letters—AI-OncLetters
PURPOSE: Although large language models (LLMs) are increasingly used in clinical practice, formal assessments of their quality, accuracy, and effectiveness in medical oncology remain limited. We aimed to evaluate the ability of ChatGPT, an LLM, to generate physician and patient letters from clinical case notes. METHODS: Six oncologists created 29 (four training, 25 final) synthetic oncology case notes. Structured prompts for ChatGPT were iteratively developed using the four training cases; once finalized, 25 physician-directed and patient-directed letters were generated. These underwent evaluation by expert consumers and oncologists for accuracy, relevance, and readability using Likert scales. The patient letters were also assessed with the Patient Education Materials Assessment Tool for Print (PEMAT-P), Flesch Reading Ease, and Simple Measure of Gobbledygook index. RESULTS: Among physician-to-physician letters, 95% (119/125) of oncologists agreed they were accurate, comprehensive, and relevant, with no safety concerns noted. These letters demonstrated precise documentation of history, investigations, and treatment plans and were logically and concisely structured. Patient-directed letters achieved a mean Flesch Reading Ease score of 73.3 (seventh-grade reading level) and a PEMAT-P score above 80%, indicating high understandability. Consumer reviewers found them clear and appropriate for patient communication. Some omissions of details (eg, side effects), stylistic inconsistencies, and repetitive phrasing were identified, although no clinical safety issues emerged. Seventy-two percent (90/125) of consumers expressed willingness to receive artificial intelligence (AI)-generated patient letters. CONCLUSION: ChatGPT, when guided by structured prompts, can generate high-quality letters that align with clinical and patient communication standards. No clinical safety concerns were identified, although addressing occasional omissions and improving natural language flow could enhance their utility in practice. Further studies comparing AI-generated and human-written letters are recommended.
- 1 May 2025
- Published by American Society of Clinical Oncology (ASCO) in JCO Clinical Cancer Informatics
- Vol. 9 (9) , e2400323
Application of Generative Artificial Intelligence for Physician and Patient Oncology Letters—AI-OncLetters
PURPOSE: Although large language models (LLMs) are increasingly used in clinical practice, formal assessments of their quality, accuracy, and effectiveness in medical oncology remain limited. We aimed to evaluate the ability of ChatGPT, an LLM, to generate physician and patient letters from clinical case notes. METHODS: Six oncologists created 29 (four training, 25 final) synthetic oncology case notes. Structured prompts for ChatGPT were iteratively developed using the four training cases; once finalized, 25 physician-directed and patient-directed letters were generated. These underwent evaluation by expert consumers and oncologists for accuracy, relevance, and readability using Likert scales. The patient letters were also assessed with the Patient Education Materials Assessment Tool for Print (PEMAT-P), Flesch Reading Ease, and Simple Measure of Gobbledygook index. RESULTS: Among physician-to-physician letters, 95% (119/125) of oncologists agreed they were accurate, comprehensive, and relevant, with no safety concerns noted. These letters demonstrated precise documentation of history, investigations, and treatment plans and were logically and concisely structured. Patient-directed letters achieved a mean Flesch Reading Ease score of 73.3 (seventh-grade reading level) and a PEMAT-P score above 80%, indicating high understandability. Consumer reviewers found them clear and appropriate for patient communication. Some omissions of details (eg, side effects), stylistic inconsistencies, and repetitive phrasing were identified, although no clinical safety issues emerged. Seventy-two percent (90/125) of consumers expressed willingness to receive artificial intelligence (AI)-generated patient letters. CONCLUSION: ChatGPT, when guided by structured prompts, can generate high-quality letters that align with clinical and patient communication standards. No clinical safety concerns were identified, although addressing occasional omissions and improving natural language flow could enhance their utility in practice. Further studies comparing AI-generated and human-written letters are recommended.
- Open Access
- 3 March 2025
- Published by Oxford University Press (OUP) in JNCI Cancer Spectrum
- Vol. 9 (2)
Evaluating patient diversity in early phase clinical trials in Australia through a prospective multicenter nonrandomized cohort study
Background: Early phase clinical trials continue to have difficulty with enrolling real-world populations with many minorities being underrepresented. Reasons for this include patient or clinician perception as well as cultural, linguistic, or social barriers. In Australia, there is currently no prospective data in the early phase clinical trial space regarding recruitment of priority populations. Methods: Patient Diversity in Early Phase Clinical Trials was a multicenter, prospective, cohort study involving 2 major early phase clinical trial centers in Sydney, Australia. All participants who were consented to an early phase clinical trial between August 2023 and August 2024 were enrolled. Participants completed a baseline demographic survey, which included cultural and linguistic status, sexual orientation, socioeconomic status, and regional diversity. Results: A total of 114 participants were recruited. Median age was 63 years (range = 25-83 years) with predominance for female participants (52%). No participant reported a nonbinary gender. All participants reported their sexuality as heterosexual, with no LGBTQIA+ participants recruited. A total of 34 (30%) participants were identified as culturally diverse, while 28 (25%) were linguistically diverse. One patient identified as Indigenous Australian. Of the participants, 26% were born overseas, with 44% having at least 1 parent born overseas. The majority were living in households with family members, with 8% of participants living alone. Conclusion: Patient Diversity in Early Phase Clinical Trials is the first prospective study that provides granular description of social, cultural, linguistic, economic, and sexual diversity among early phase clinical trial participants. Certain subgroups are underrepresented, including those with sexual diversity, gender diversity, and Indigenous backgrounds. Ongoing efforts to monitor and promote inclusion of diverse populations in clinical trials are vital.
- 3 March 2025
- Published by Oxford University Press (OUP) in JNCI Cancer Spectrum
- Vol. 9 (2)
Open AccessEvaluating patient diversity in early phase clinical trials in Australia through a prospective multicenter nonrandomized cohort study
Background: Early phase clinical trials continue to have difficulty with enrolling real-world populations with many minorities being underrepresented. Reasons for this include patient or clinician perception as well as cultural, linguistic, or social barriers. In Australia, there is currently no prospective data in the early phase clinical trial space regarding recruitment of priority populations. Methods: Patient Diversity in Early Phase Clinical Trials was a multicenter, prospective, cohort study involving 2 major early phase clinical trial centers in Sydney, Australia. All participants who were consented to an early phase clinical trial between August 2023 and August 2024 were enrolled. Participants completed a baseline demographic survey, which included cultural and linguistic status, sexual orientation, socioeconomic status, and regional diversity. Results: A total of 114 participants were recruited. Median age was 63 years (range = 25-83 years) with predominance for female participants (52%). No participant reported a nonbinary gender. All participants reported their sexuality as heterosexual, with no LGBTQIA+ participants recruited. A total of 34 (30%) participants were identified as culturally diverse, while 28 (25%) were linguistically diverse. One patient identified as Indigenous Australian. Of the participants, 26% were born overseas, with 44% having at least 1 parent born overseas. The majority were living in households with family members, with 8% of participants living alone. Conclusion: Patient Diversity in Early Phase Clinical Trials is the first prospective study that provides granular description of social, cultural, linguistic, economic, and sexual diversity among early phase clinical trial participants. Certain subgroups are underrepresented, including those with sexual diversity, gender diversity, and Indigenous backgrounds. Ongoing efforts to monitor and promote inclusion of diverse populations in clinical trials are vital.