Overview
Initiative type
Service Improvement
Status
Deliver
Published
June 2026
Summary
This proof-of-concept project involved implementation of AI-scribe within Children's Health Queensland (CHQ) to reduce documentation burden, improve quality of documentation and patient experience and evaluate cost-effectiveness.
Dates: June 2024 - June 2025
Implementation sites: Queensland Children's Hospital
Aim
To evaluate the impact of an artificial intelligence medical scribe (AIMS) on clinical documentation efficiency, document quality, clinician-patient interaction and clinician well-being, consumer perspectives, overtime and transcription costs in a quaternary paediatric healthcare setting.
Outcomes
Significant reductions were observed in median time-to-finalise outpatient letters from 7.9 days [Interquartile range (IQR) 19 days] in pre-AIMS era to 16 minutes (IQR 6 days) (P < 0.001). (Reflections survey): participants reported improved: patient engagement (62/71, 87%), time-to-finalise documents (83%, 59/71), well-being (80%, 57/71) and work-life balance (66%, 47/71).
Document quality survey showed acceptable document quality with mean total score of 36.7 (out of 50), and median score 4, in 9 of the 10 survey domains (scale 1-to-5, n=77); Mean System Usability Score was 74 (n=77) suggesting above average usability. Transcription cost-savings of 16,903 AUD - no definite impact on overtime Patient feedback was highly positive >90% reporting improved clinician-patient interaction
Background
We are currently living in an era which is now being defined as the age of artificial intelligence (AI. AI is being incorporated into the workplace due to its capacity to save time, streamline tasks and improve efficiencies through rapidly evolving digital capabilities. Within healthcare, ambient AI scribes are being increasingly adopted by clinicians worldwide. Ambient AI scribes utilise advanced speech recognition and integrate powerful capabilities driven by generative AI and large language models (LLM’s) to convert dialogue from clinicians and patients into text alongside the capacity to produce various outputs supporting documentation requirements of clinical care, integrated seamlessly into efficient workflows. Leaders in healthcare are looking at this technology to improve clinical efficiency, enhance patient care, reduce clinician workload and improve quality of clinical data. Ambient AI scribes can allow clinicians to spend more time caring for patients instead of documenting interactions, thus returning clinicians to the human dimensions of healthcare which technology will never replace.
In the US private health care organisations have been steadily rolling out pilot studies of ambient AI scribes within their organisations in response to what many have described as a major crisis in healthcare due to unsustainable costs in delivering healthcare and increasing burnout amongst the clinician workforce. Wellbeing is fast becoming a priority and responsibility for healthcare organisations and clinicians are increasingly voicing difficulties in achieving work/life balance in the context of ever-increasing workloads that have emerged in modern complex healthcare environments. Early studies of this emergent technology have shown some promise of enhanced clinical efficiencies and workflow by reducing the administrative burden from note-taking/typing; improved patient care from increased availability for patient focus without the distraction of documentation; increased accuracy in documentation reducing risk of medical errors and promoting collaborative care between providers alongside supporting but not replacing better clinical decision-making.
Published studies to date of international AI scribe implementations have shown highest adoption in departments with the highest levels of documentation burden. There is limited evidence regarding efficiency outcomes, impact on health professionals, consumer perspectives and cost evaluation, especially in paediatric settings. Adding to the triple aims of healthcare which defined priorities of healthcare to improve population health, reduce healthcare costs and improve patient experience, this framework was later expanded to the quintuple aim which included reducing health inequities and improving the wellbeing of the healthcare workforce as important priorities for healthcare initiatives to address.
Ambient AI scribe technology is well positioned to achieve meaningful benefit in all sectors of this framework. The genesis of this project was in late 2023. As an early adopter of this technology in private practice, in early 2024 paediatrician Dr Catherine Skellern from the Queensland Children’s Hospital applied for, and was awarded, a grant from the Chief Executive Imagination Fund for a proof-of-concept project to evaluate the implementation of an AI scribe into Children’s Health Queensland across medical, allied health and nursing disciplines.
Initially undertaken as a quality improvement evaluation, this later transitioned into a retrospective-prospective research cohort study which is still ongoing.
Methods
Co-Leads Dr Vishal Kapoor and Pauline McGrath were appointed alongside Project Officer, Stephanie Brosnan from Data Analytics and Custom Solutions (DACS) in CHQ.
Using criteria from MDA National of AI scribe considerations the panel selected Heidi Health as the preferred AI vendor for the project following government tender/procurement procedures and market sweep of existing AI vendors with sovereign (on-shore Australian) servers. Experts from DACS led the Information Security and Risk Assessment Plan (ISRAP) and Privacy Impact Assessment (PIA) including defining requirements for the tool within CHQ systems. Around QCH and CHQ, Expressions of Interest invited clinicians from medical, nursing and allied health disciplines to join the project.
A phased implementation was undertaken including training and personalised template fine-tuning sessions with the AIMS vendor for the first two onboarding groups (designated Alpha and Beta), followed by the Gamma group, who had limited personalised training support and relied more on online training resources and peer support due to resource constraints. The phased onboarding took 12 weeks to ensure each cohort mastered using the system before the next cohort was onboarded. Participants provided data on baseline demographics including age, sex, professional stream, weekly work hours, outpatient appointments and filled online surveys using Microsoft® Forms. AIMS vendor provided data on the number of sessions, transcripts, documents and system outages. Data were extracted from hospital systems and surveys into a project database.
The project team identified from published literature validated tools used by other AIMS projects and designed customised surveys for the evaluation framework to capture metrics including utility and experiences of the AIMS from clinician, patient and organisational perspectives. These measures broadly map onto the Quintuple Aim of Healthcare framework and our original project aims. Once surveys/forms were developed, these were initially trialled with a small number of clinicians then finalised prior to distribution to participants.
Outcome evaluations included:
1. Workflow efficiencies: This relied on data of time taken to finalise outpatient correspondence within the hospital document management system (Dragon Medical Workflow Manager) over a 6 month period using comparison of a 6 month period in the preceding calendar year.
2. Patient experience: A survey was developed to capture patient experience with results grouped into the 3 clinical streams of medical, nursing and allied health appointments.
3. Clinician wellbeing measures: This used comparison measures within the Maslach Burnout Inventory related
to work of documentation demands.
4. Clinician perspectives “Reflections” survey: A custom designed survey capturing clinician perspectives about their experiences was completed 1-month post-implementation.
5. System Usability Scale: Completed at three months post-implementation, this provided clinician perspectives about ease of adoption of this new technology.
6. A Modified Physician Documentation Quality Instrument (M-PDQI): Initially adapted from its original published form, this clinician survey was trialled then completed at three months to evaluate note quality.
7. Focus group discussions were convened in May 2025 to allow clinicians to provide further feedback regarding their experiences with the AIMS using qualitative research methods (thematic and sentiment analyses).
Discussion
The current study is one of the first in-depth evaluations of AIMS implementation in Australia, within a public children’s HHS, viewed through the broad lens of Quintuple Aim of healthcare, capturing a broad range of clinicians including medical, nursing and allied health professionals showing improvement in documentation efficiency with acceptable document quality to clinicians, improvement in clinician-patient interactions and clinician well-being across all professional streams.
Our study captured patient experience with AIMS from a larger patient cohort than existing published studies (n=333) across a range of medical, allied health and nursing disciplines, with greater than 90% reporting improvement in clinician focus, communication and pace of discussions. Most consumers felt comfortable with the use of AIMS and 94.6% welcomed the use of AIMS in future appointments. Further work is underway to refine specialty specific templates and develop templates for First Nations people, focusing on culturally grounded co-design approach. Considering return of investment, our study showed the use of AIMS decreased transcription costs by 16,903 AUD for the study participants. An estimated 141 AIMS generated letters would lead to a cost-neutral argument for transcription costs for a single clinician based on current costs for licenses. Enablers of this project included the decision for the project to be embedded within CHQ Digital Health, who undertook the privacy/security risk assessment and ensured legal compliance with government tender/procurement procedures in bringing together CHQ and the AI vendor as a partner. This project was also successful in part because it coincided with emergence of the AI frontier, as AI scribes were increasingly being adopted in private healthcare settings leading to pressure from clinicians in hospitals seeking to similarly use this emergent technology.
The project team were able to leverage relationships with the clinician workforce within our organisation to access study participants. We also benefited from the expertise of an experienced researcher Dr Kapoor to ensure our project had a robust research
evaluation framework. Lessons learned included the finding that adoption of this technology was not influenced by age of clinician and that varied levels of support are needed by clinicians to enhance adoption of this technology into existing clinical workflows.
We needed to slow the onboarding process from what was initially planned to ensure adequate adoption was achieved in each cohort. Throughout the duration of this project there were frequent enquiries from clinicians and healthcare executives seeking advice regarding how to implement this technology within their own organisations, evidencing the word-of-mouth diffusion of our project’s success.
Our results are providing a strong foundation for broader implementation of AIMS across other Queensland Health sites and CHQ has now moved forward to a pilot study allowing >2000 clinicians to have access to the AIMS over the next 12 months. After this, it is expected that funding for ongoing access to an AIMS will be incorporated into business-as-usual funding given our
preliminary results on cost-effectiveness. Future efforts should focus on the impact of AIMS on quality of healthcare provision, clinical decision support, risk evaluation and specific clinical and administrative contexts where AIMS can provide efficiency
gains.
References
1. Balloch J, Sridharan S, Oldham G, et al. Use of an ambient artificial intelligence tool to improve quality of clinical documentation. Future Healthc J 2024; 11: 100157.
2. Clough RAJ, Sparkes WA, Clough OT, et al. Transforming healthcare documentation: harnessing the potential of AI to generate discharge summaries. BJGP Open 2024; 8.
3. Seth P, Carretas R, Rudzicz F. The Utility and Implications of Ambient Scribes in Primary Care. JMIR AI 2024; 3: e57673.
4. Bhattacharyya O, Agarwal P, Ha E, et al. Accelerating AI Adoption for Reducing Administrative Burden in Primary Care: Insights from
Evaluating AI Scribes. Healthc Pap 2025; 22: 63-8.
5. Melnick ER, West CP, Nath B, et al. The association between perceived electronic health record usability and professional burnout among US nurses. J Am Med Inform Assoc 2021; 28: 1632-41.
6. Robertson SL,
Robinson MD, Reid A. Electronic Health Record Effects on Work-Life Balance and Burnout Within the I(3) Population Collaborative. J Grad Med Educ 2017; 9: 479-84.
7. Babbott S, Manwell LB, Brown R, et al. Electronic medical records and physician stress in primary
care: results from the MEMO Study. J Am Med Inform Assoc 2014; 21: e100-6.
8. Frintner MP, Kaelber DC, Kirkendall ES, et al. The Effect of Electronic Health Record Burden on Pediatricians' Work-Life Balance and Career Satisfaction. Appl Clin Inform 2021; 12:
697-707.
9. Downing NL, Bates DW, Longhurst CA. Physician Burnout in the Electronic Health Record Era: Are We Ignoring the Real Cause? Ann Intern Med 2018; 169: 50-1.
10. Sinsky CA, Biddison LD, Mallick A, et al. Organizational Evidence-Based and Promising
Practices for Improving Clinician Well-Being. NAM Perspect 2020; 2020.
11. Itchhaporia D. The Evolution of the Quintuple Aim: Health Equity, Health Outcomes, and the Economy. J Am Coll Cardiol 2021; 78: 2262-4.
12. Tierney A, Gayre G, Hoberman B, et al. Ambient
Artificial Intelligence Scribes to Alleviate the Burden of Clinical Documentation. NEJM Catalyst 2024; 5.
13. Tierney A, Gayre G, Hoberman B, et al. Ambient Artificial Intelligence Scribes: Learnings after 1 Year and over 2.5 Million Uses. NEJM Catalyst 2025;
6.
14. Duggan MJ, Gervase J, Schoenbaum A, et al. Clinician Experiences With Ambient Scribe Technology to Assist With Documentation Burden and Efficiency. JAMA Netw Open 2025; 8: e2460637.
15. Shah SJ, Devon-Sand A, Ma SP, et al. Ambient artificial intelligence scribes: physician burnout and perspectives on usability and documentation burden. J Am Med Inform Assoc 2025; 32: 375-80.
16. Ma SP, Liang AS, Shah SJ, et al. Ambient artificial intelligence scribes: utilization and impact on documentation time. J Am Med
Inform Assoc 2025; 32: 381-5.
17. Albrecht M, Shanks D, Shah T, et al. Enhancing clinical documentation with ambient artificial intelligence: a quality improvement survey assessing clinician perspectives on work burden, burnout, and job satisfaction. JAMIA Open 2025; 8: ooaf013.
18. Shah SJ, Crowell T, Jeong Y, et al. Physician Perspectives on Ambient AI Scribes. JAMA Netw Open 2025; 8: e251904.
19. Evans K, Papinniemi A, Ploderer B, et al. Impact of using an AI scribe on clinical documentation and clinician-patient relationship.
Key contact
Dr Catherine Skellern
Senior Medical Officer
Child Protection and Forensic Medicine Service
Queensland Children's Hospital