Critical Intelligence Unit

AI: automating indirect clinical tasks and administration: Living evidence

Living evidence tables provide high level summaries of key studies and evidence on a particular topic, and links to sources. They are reviewed regularly and updated as new evidence and information is published.

This page is part of the Artificial Intelligence series.

This table describes areas where artificial intelligence is being used for indirect clinical care and administration in healthcare systems. Evidence includes examples of established tools and pilot studies, as well as expert commentary in cases where robust evidence is not yet available.

Regular checks are conducted for new content and any updates are highlighted.

To develop this living table:

  • Systematic reviews on artificial intelligence published from 2021 to May 2023 were screened in PubMed.
  • Since May 2023, a weekly PubMed search on artificial intelligence has been conducted to continually update the table.
  • Articles are screened for those that describe established uses of AI in indirect clinical care and administration, pilot studies of implementation, or expert commentary.
  • Grey literature sources are identified via Google or through publications and websites of well-known organisations with relevance to the Australian clinical context (e.g. World Health Organization, OECD, NICE, The King’s Fund, US Center for Disease Control and Prevention, etc).
Best bet – how can AI help Evidence and expert commentary Examples of adoption in healthcare systems

Automating routine indirect clinical tasks

Limited but emerging positive evidence that AI, such as AI scribes or speech recognition for clinical documentation, can automate tasks and free up clinician time.

Efficiency

  • Systematic reviews on LLM-based systems and scribes note that benefits can include:1, 2
    • increased efficiency including time taken to complete clinical notes and number of patients seen
  • However, for LLM-based systems that claim to shorten documentation time1, 2
    • studies often rely on small sample sizes or single-site studies, limiting generalisability
    • time savings can be negated by the required human-editing time

Capability and accuracy

  • generative AI can transcribe patient consults then summarise these into clinical notes3
    • this technology is commercially available4, 5
    • a systematic review noted that different systems have differing accuracy although some demonstrate impressive precision and recall1
  • large language models can draft replies to patient emails and outperform medical experts in analysing and summarising large amounts of clinical documentation6

Safety and quality

  • The evidence around safety remains contradictory1
    • AI-assisted documentation may capture more clinically relevant red flags than clinician-typed notes
    • However, errors or omissions may go unnoticed and pose risks to patient safety

Satisfaction

  • A systematic review noted that benefits can include increases in clinician wellness and burnout rates2
  • Small studies have noted patient satisfaction with the amount of time clinicians using AI scribes engage and look at them, however the evidence on patient satisfaction is limited2

Accessibility and health literacy

  • AI systems can also provide focused summaries of charts, summarise encounters for staff handovers, and create discharge instructions with an appropriate language and reading level3

In Australia, hospitals have started trialling the use of endorsed AI scribes

  • in outpatient settings in Queensland Health7
  • in South Australia health with a purpose-built system that integrates with the electronic health record8
  • in private hospitals in Queensland, with bespoke designed solutions9

Internationally, a number of hospitals have started trialling generative AI to write clinical documentation, although clinician checks of the documentation are still used10, 11

  • A trial of TORTUS software across nine NHS sites freed up clinicians to spend nearly a quarter more time with patients and reduce burnout12
  • Chelsea and Westminster NHS Trust are trialling an AI tool which drafts discharge   summaries by extracting key information from medical records13
  • Guy’s and St Thomas’ NHS has adopted an AI-powered speech recognition solution where a 90% adoption rate in one department led to an estimated time saving of 60 hours of clinician time per month by automating workflows14
  • ‘Heidi AI’ scribe has been endorsed for use within Health NZ, including in emergency departments15
Reducing errors via double checking mechanisms and risk alerts

Limited but emerging positive evidence that AI can support drug decisions and dispensing processes16and reduce near miss events including:17-19

  • reduce prescription errors20
  • identify harmful drug interactions20, 21
  • identify harmful drug-food interactions21
  • support drug dosing decisions for high-risk drugs20
 

Quality and process improvement

Limited but emerging positive evidence that AI can speed up and improve workflow:

  • large language models integrated into outpatient reception workflows can improve communication that benefits nurses and patients22

Limited but emerging positive evidence that AI can evaluate clinician skill and support training:

  • machine learning models can equal or outperform humans at evaluating neurosurgical methods and skill. This has implications for surgical training and improvement methods and may reduce the senior personnel hours required for supervision and training23-25
  • Mount Sinai has installed an AI platform that live records videos of surgery, adds automatic annotations and syncs these files to the patient EHR and clinical outcomes, for later   evaluation and in order to promote surgical quality improvement26

Patient-reported outcome data

Limited but emerging positive evidence that AI can evaluate patient-reported data:

  • natural language processing can analyse unstructured patient-reported outcome data to support risk stratification and adverse clinical outcomes in oncology27
 

Analysis and automation of patient triage and flow

Limited but emerging positive evidence that AI can automate patient prioritisation:

  • triaging patients in emergency departments or directing them to appropriate settings, including associated reduced waiting times3, 28
  • determining the necessity of hospitalisation for patients requiring urgent attention29
  • improved surgical scheduling and management of surgical waitlists30

Limited but emerging positive evidence that AI can predict and improve patient flow:31

  • in inpatient settings including improving bed capacity, predicting length of stay and achievement of discharge targets29, 32-34
  • in outpatient settings35
  • in surgical management to predict surgery length, recovery ward length of stay or to   predict surgery cancellation risks36
  • Three hospitals under Alfred Health in Western Australia have deployed an AI system that acts as a centralised ‘command centre’ to manage patient flow across their wards37

AI analysis of system resource allocation

Limited but emerging positive evidence that AI can suggest improvements in system organisation and resourcing:

  • during major emergencies:
    • allocating emergency services38
    • determining the optimum plan for resource allocation across multiple   hospitals39
  • predicting patient demand31
  • predicting staff needs and scheduling31
  • An NHS trust is piloting AI software to reduce missed hospital appointments. The software is projected to allow an additional 80-100,000 patients to be seen each year40
  • AI-driven resource allocation has been used by NGOs to assess diagnostic testing needs and simplify transport logistics41

Automation of administrative tasks (aka “back office” functions)42

  • coding
  • scheduling and patient reminders
  • billing
  • identifying fraud

AI adoption is expected to be rapid - administrative AI don’t require clinical or regulatory approval (like clinical AI does).43

Limited but emerging positive evidence that AI can improve scheduling:

  • Specifically in radiology departments, AI has enhanced workflows and clinical outcomes by improving appointment scheduling, triage, resource allocation and aftercare management44
  • For rostering, AI can very quickly create staff rosters which are fairer, include multiple combinations for redundancy and reduce staff burnout45
  • A hospital in Kansas City now uses AI software to automate scheduling and workflow in order to maximise operating theatre time and resources. The hospital  has been able to do 7% more surgical cases despite having to close 20% of its ORs at times46

References

  1. Sutherland E. Artificial intelligence in health: big opportunities, big risks. Paris: OECD; 2023 [cited 12 Feb 2024]. Available from: https://oecd.ai/en/wonk/artificial-intelligence-in-health-big-opportunities-big-risks
  2. Vize R. More Time to Care: Automation, Digitisation and the Workforce. Beamtree; 2023 [cited 12 Feb 2024]. Available from: https://beamtree.com.au/papers-publications/more-time-to-care/
  3. Canadian Federation of Independent Business. Patients before Paperwork. Toronto: Canadian Federation of Independent Business; 2023 [cited 12 Feb 2024]. Available from: https://www.cfib-fcei.ca/en/research-economic-analysis/patients-before-paperwork
  4. Dicuonzo G, Donofrio F, Fusco A, et al. Healthcare system: Moving forward with artificial intelligence. Technovation. 2023;120:102510. DOI: https://doi.org/10.1016/j.technovation.2022.102510
  5. Van Veen D, Van Uden C, Blankemeier L, et al. Adapted large language models can outperform medical experts in clinical text summarization. Nature Medicine. 2024. DOI: 10.1038/s41591-024-02855-5
  6. Garcia P, Ma SP, Shah S, et al. Artificial Intelligence–Generated Draft Replies to Patient Inbox Messages. JAMA Network Open. 2024;7(3):e243201-e. DOI: 10.1001/jamanetworkopen.2024.3201
  7. Kachman MM, Brennan I, Oskvarek JJ, et al. How artificial intelligence could transform emergency care. The American Journal of Emergency Medicine. 2024;81:40-6. DOI: https://doi.org/10.1016/j.ajem.2024.04.024
  8. Amazon.com Inc. AWS announces AWS HealthScribe, a new generative AI-powered service that automatically creates clinical documentation. New York: Amazon.com Inc,; 2023 [cited 25 Oct 2023]. Available from: https://press.aboutamazon.com/2023/7/aws-announces-aws-healthscribe-a-new-generative-ai-powered-service-that-automatically-creates-clinical-documentation#:~:text=With%20generative%20AI%20capabilities%20powered,to%20enter%20into%20the%20EHR.
  9. Pifer R. Amazon launches generative AI-based clinical documentation service. Washington, DC: Healthcare Dive; 2023 [cited 25 Oct 2023]. Available from: https://www.healthcaredive.com/news/amazon-generative-ai-clinical-documentation-healthscribe/688996/
  10. MIT Technology Review Insights. The AI Effect: How artificial intelligence is making health care more human. Cambridge, MA: MIT; 2019 [cited 14 Nov 2023]. Available from: https://www.gehealthcare.co.uk/-/jssmedia/61b7b6b1adc740e58d4b86eef1bb6604.pdf
  11. Tierney AA, Gayre G, Hoberman B, et al. Ambient Artificial Intelligence Scribes to Alleviate the Burden of Clinical Documentation. Catalyst non-issue content. 2024;5(1):CAT.23.0404. DOI: doi:10.1056/CAT.23.0404
  12. Willis M, Duckworth P, Coulter A, et al. Qualitative and quantitative approach to assess the potential for automating administrative tasks in general practice. BMJ Open. 2020;10(6):e032412. DOI: 10.1136/bmjopen-2019-032412
  13. Siwicki B. In pilot, generative AI expected to reduce clinical documentation time at Baptist Health. Healthcare IT News; 2023 [cited 25 Oct 2023]. Available from: https://www.healthcareitnews.com/news/generative-ai-reduces-clinical-documentation-time-baptist-health
  14. Southwick R. HIMSS23: Microsoft and Epic are bringing AI into electronic health records. Medical Economics; 2023 [cited 25 July 2024]. Available from: https://www.medicaleconomics.com/view/himss23-microsoft-and-epic-are-bringing-ai-into-electronic-health-records
  15. Buijs E, Maggioni E, Mazziotta F, et al. Clinical impact of AI in radiology department management: a systematic review. La radiologia medica. 2024;129(11):1656-66. DOI: 10.1007/s11547-024-01880-1
  16. Choudhury A, Asan O. Role of Artificial Intelligence in Patient Safety Outcomes: Systematic Literature Review. JMIR Med Inform. 2020;8(7):e18599. DOI: 10.2196/18599
  17. Pais C, Liu J, Voigt R, et al. Large language models for preventing medication direction errors in online pharmacies. Nature Medicine. 2024. DOI: 10.1038/s41591-024-02933-8
  18. Graafsma J, Murphy RM, van de Garde EMW, et al. The use of artificial intelligence to optimize medication alerts generated by clinical decision support systems: a scoping review. Journal of the American Medical Informatics Association. 2024;31(6):1411-22. DOI: 10.1093/jamia/ocae076
  19. Johns E, Alkanj A, Beck M, et al. Using machine learning or deep learning models in a hospital setting to detect inappropriate prescriptions: a systematic review. European Journal of Hospital Pharmacy. 2024;31(4):289-94. DOI: 10.1136/ejhpharm-2023-003857
  20. Chalasani SH, Syed J, Ramesh M, et al. Artificial intelligence in the field of pharmacy practice: A literature review. Exploratory Research in Clinical and Social Pharmacy. 2023;12:100346. DOI: https://doi.org/10.1016/j.rcsop.2023.100346
  21. Davenport L. AI Tool Reveals MS Drug Interactions, Offers Safer Options. Newark, NJ: Medscape Medical News; 2023 [cited 27 Nov 2023]. Available from: https://www.medscape.com/viewarticle/997331?ecd=mkm_ret_231126_mscpmrk-OUS_InFocus_etid6087426&uac=181280EV&impID=6087426
  22. Dykes PC, Adelman JS, Alfieri L, et al. The Fall TIPS (Tailoring Interventions for Patient Safety) Program: A Collaboration to End the Persistent Problem of Patient Falls. Nurse Leader. 2019;17(4):365-70. DOI: https://doi.org/10.1016/j.mnl.2018.11.006
  23. Dykes PC, Curtin-Bowen M, Lipsitz S, et al. Cost of Inpatient Falls and Cost-Benefit Analysis of Implementation of an Evidence-Based Fall Prevention Program. JAMA Health Forum. 2023;4(1):e225125-e. DOI: 10.1001/jamahealthforum.2022.5125
  24. Scardoni A, Balzarini F, Signorelli C, et al. Artificial intelligence-based tools to control healthcare associated infections: A systematic review of the literature. Journal of Infection and Public Health. 2020;13(8):1061-77. DOI: https://doi.org/10.1016/j.jiph.2020.06.006
  25. Salehinejad H, Meehan AM, Rahman PA, et al. Novel machine learning model to improve performance of an early warning system in hospitalized patients: a retrospective multisite cross-validation study. eClinicalMedicine. DOI: 10.1016/j.eclinm.2023.102312
  26. Landro L. How 4 hospitals are using technology to reduce medical errors. Online: Advisory Board; 2023 [cited 14 Nov 2023]. Available from: https://www.advisory.com/daily-briefing/2023/03/15/medical-errors
  27. Wan P, Huang Z, Tang W, et al. Outpatient reception via collaboration between nurses and a large language model: a randomized controlled trial. Nature Medicine. 2024. DOI: 10.1038/s41591-024-03148-7
  28. Titov O, Bykanov A, Pitskhelauri D. Neurosurgical skills analysis by machine learning models: systematic review. Neurosurgical Review. 2023;46(1):121. DOI: 10.1007/s10143-023-02028-x
  29. Lavanchy JL, Zindel J, Kirtac K, et al. Automation of surgical skill assessment using a three-stage machine learning algorithm. Scientific Reports. 2021;11(1):5197. DOI: 10.1038/s41598-021-84295-6
  30. Yilmaz R, Bakhaidar M, Alsayegh A, et al. Real-Time multifaceted artificial intelligence vs In-Person instruction in teaching surgical technical skills: a randomized controlled trial. Scientific Reports. 2024;14(1):15130. DOI: 10.1038/s41598-024-65716-8
  31. Siwicki B. AI helps Mount Sinai organize and unlock largely untapped surgical data. Chicago, IL: Healthcare IT News; 2023 [cited 14 Nov 2023]. Available from: https://www.healthcareitnews.com/news/ai-helps-mount-sinai-organize-and-unlock-largely-untapped-surgical-data
  32. Sim J-A, Huang X, Horan MR, et al. Using natural language processing to analyze unstructured patient-reported outcomes data derived from electronic health records for cancer populations: a systematic review. Expert Review of Pharmacoeconomics & Outcomes Research. 2024;24(4):467-75. DOI: 10.1080/14737167.2024.2322664
  33. Henderson AP, Van Schuyver PR, Economopoulos KJ, et al. The Use of Artificial Intelligence for Orthopedic Surgical Backlogs Such as the One Following the COVID-19 Pandemic: A Narrative Review. JBJS Open Access. 2024;9(3):e24.00100. DOI: 10.2106/jbjs.Oa.24.00100
  34. Bin Abdul Baten R. How are US hospitals adopting artificial intelligence? Early evidence from 2022. Health Affairs Scholar. 2024;2(10):qxae123. DOI: 10.1093/haschl/qxae123
  35. Spatharou A, Hieronimus S, Jenkins J. Transforming healthcare with AI: The impact on the workforce and organizations. New York: McKinsey & Company,; 2020 [cited 17 Jul 2023]. Available from: https://www.mckinsey.com/industries/healthcare/our-insights/transforming-healthcare-with-ai
  36. Jiandong Z, Andrew JB, David AC, et al. Improving patient flow through hospitals with machine learning based discharge prediction. medRxiv. 2023:2023.05.02.23289403. DOI: 10.1101/2023.05.02.23289403
  37. Li L, Diouf F, Gorkhali A. Managing outpatient flow via an artificial intelligence enabled solution. Systems Research and Behavioral Science. 2022;39(3):415-27. DOI: https://doi.org/10.1002/sres.2870
  38. Bellini V, Russo M, Domenichetti T, et al. Artificial Intelligence in Operating Room Management. Journal of Medical Systems. 2024;48(1):19. DOI: 10.1007/s10916-024-02038-2
  39. Hospital and Healthcare. Alfred Health rolls out new solutions to support patient flow. Online: Hospital and Healthcare; 2023 [cited 14 Nov 2023]. Available from: https://www.hospitalhealth.com.au/content/technology/news/alfred-health-rolls-out-new-solutions-to-support-patient-flow-826496502
  40. Du L. Medical Emergency Resource Allocation Model in Large-Scale Emergencies Based on Artificial Intelligence: Algorithm Development. JMIR Med Inform. 2020;8(6):e19202. DOI: 10.2196/19202
  41. Huang C-H, Batarseh FA, Boueiz A, et al. Measuring outcomes in healthcare economics using Artificial Intelligence: With application to resource management. Data & Policy. 2021;3:e30. DOI: 10.1017/dap.2021.29
  42. NHS England. NHS pilots artificial intelligence software to cut missed hospital appointments. London: NHS; 2023 [cited 15 Nov 2023]. Available from: https://www.england.nhs.uk/2023/02/nhs-pilots-artificial-untelligence-software-to-cut-missed-hospital-appointments/
  43. Purcell R. Coupa and FIND Recognized by Fast Company for a World Changing Idea. San Mateo, CA: Coupa; 2022 [cited 14 Nov 2023]. Available from: https://www.coupa.com/blog/coupa-news-culture/coupa-and-find-recognized-fast-company-for-world-changing-idea
  44. Organisation for Economic Co-operation and Development (OECD). Trustworthy AI in health. Paris: OECD; 2020 [cited 17 Jul 2023]. Available from: https://www.oecd.org/health/trustworthy-artificial-intelligence-in-health.pdf
  45. Davenport T, Bean R. Clinical AI Gets the Headlines, but Administrative AI May Be a Better Bet. Cambridge, MA: MIT Sloan Management Review; 2022 [cited 20 Nov 2023]. Available from: https://sloanreview.mit.edu/article/clinical-ai-gets-the-headlines-but-administrative-ai-may-be-a-better-bet/
  46. American Society of Anesthesiologists (ASA). Using AI to create work schedules significantly reduces physician burnout, study shows. Schaumburg, IL ASA; 2022 [cited 15 Nov 2023]. Available from: https://www.asahq.org/about-asa/newsroom/news-releases/2022/01/using-ai-to-create-work-schedules-significantly-reduces-physician-burnout
  47. Siwicki B. AI-powered OR scheduling tech brings big efficiencies for St. Luke's. Online: Healthcare IT News; 2022 [cited 20 Nov 2023]. Available from: https://www.healthcareitnews.com/news/ai-powered-or-scheduling-tech-brings-big-efficiencies-st-lukes

Living evidence tables are based on a simplified review method and may not be entirely exhaustive but aim to provide a balanced assessment of what is already known about a specific problem or issue. They include some links to low quality sources and an assessment of the original source has not been undertaken. Sources are monitored regularly but due to rapidly emerging information, tables may not always reflect the most current evidence. The tables are not peer reviewed, and inclusion does not imply official recommendation nor endorsement of NSW Health.

Last updated on 1 Dec 2025

Back to top