AIDH facebook
Select Page

Introduction
The June 2022 editorial [Impact of artificial intelligence and machine learning on nursing and midwifery]  indicated that artificial intelligence (AI) was continuing to grow at a rapid rate each year1, and introduced the concepts of virtual care, chatbots, robotics, data analytics, and clinical decision support. Yin identified that nurses and midwives are best positioned to measure the impact of AI and how AI affects patient outcomes, and urged them to be proactive in contributing to AI system development as subject matter experts and patient advocates1.

The new year has witnessed a continued increase in this technology in healthcare, with the Australian Alliance for Artificial Intelligence in Healthcare launching its national policy roadmap at the November 2023 AI.Care conference2. This document, whilst viewing AI as a pathway to generating a smarter, more adaptive health system, calls for a mature and coordinated national approach so that Australia can maximise the benefits of this technology2. As is the case with all digital technologies, nurses and midwives must be actively involved in the design, purchase, implementation, use, and refinement of AI technology. This can only be achieved by nurses and midwives acquiring a comprehensive understanding of AI and appreciating the impact it will likely to have on both the providers and recipients of care.

Benefits and challenges
Current nursing and midwifery technologies that leverage AI include virtual reality, voice assistive technology, natural language processing, image recognition, robotics, expert rule-based systems, and machine and deep learning3,4. Promised benefits of deploying AI into clinical nursing and midwifery practice are replete in the literature5-7, encompassing improved or enhanced:

  • care needs assessment
  • risk assessment
  • patient monitoring
  • care planning
  • medication management
  • evidence informed decision support
  • early detection of deterioration
  • end-of-life care
  • education and counselling with caregivers and patients
  • patient experience
  • clinician rostering and staff mix
  • administrative burden, and
  • alarm fatigue.

These benefits are largely gained through AI’s ability to quickly interrogate large volumes of data for patterns, trends and anomalies beyond human capacity through the use of algorithms, machine learning, and deep learning; accurately identify specific interventions and alerts for those patients at risk; automatically adjust decision making in response to newly added data; and automate routine, mundane, and time-consuming tasks that do not require specialised nursing or midwifery knowledge or skills, allowing nurses and midwives to perform tasks more efficiently8-10.

Despite providing advanced insights not possible in a paper-based healthcare environment, several challenges have also been identified7, 11-14, including:

  • data security
  • patient privacy
  • ethical considerations
  • unclear legal liability
  • impact on person-centred care
  • impact on workload and integration into clinical workflow
  • inferior infrastructure (high-speed internet with a stable connection)
  • data bias issues
  • lack of a common nursing and midwifery vocabulary
  • comprehensive training and support
  • incompatibilities with traditional nursing ideals, and
  • influence on compassionate and empathetic care.

Shaping the AI mission for nursing and midwifery
Given that nursing and midwifery are the largest workforce in healthcare, a substantial amount of patient data is recorded by nurses and midwives, and it is this data that is used to train AI tools10,15. Healthcare environments that have only recently moved to an electronic medical record system lack mature datasets to serve as a foundation for this training, leading to an AI-related skills gap6. Inadequate interoperability also limits this maturity.

As nursing and midwifery’s datasets expand, nurses and midwives must understand the relationship between the data they collect and the AI technologies they use10,16. Nursing and midwifery data needs to demonstrate structure, continuity, quantity, and quality, with many authors recommending using the International Classification of Nursing Practice [ICNP]6,17. Moreover, minimum competencies focusing on AI in nursing and midwifery must now be incorporated in all undergraduate programs. Knowledgeable nurses and midwives will be able to identify potential data collection biases resulting in embedded biases in AI tools10. They will not only be able to advocate for inclusion of equity and social justice considerations10 but will also be able to participate in the overall governance process of AI technology at the clinical level7.

In order to adapt to and accommodate this technology, nurses and midwives must be able to provide clinical insights, validate algorithms, and ensure that AI aligns with best practice nursing and midwifery7. The Canadian province of Quebec has established a competency framework for nurses in relation to AI16, outlining five areas that need to be addressed:

  1. Introduction to AI health technologies in nursing practice and clinical settings
  2. Knowledge of AI data and how the data are created and stored.
  3. Communication with healthcare professionals, patients, and families
  4. Ethical and social considerations, and
  5. Engagement in AI as a subject matter expert or end user.

Shaping analytics for nursing and midwifery
AI holds the potential to generate $150 billion in healthcare savings, with predictive analytics being a pivotal method through which healthcare can harness AI for these financial benefits3. The implementation of electronic health records and other healthcare-related data repositories has resulted in an exponential increase in data volume, offering boundless possibilities for enhancing patient outcomes. These systems rely heavily on nursing/midwifery input or knowledge of the data input process. Nurses and midwives must comprehend the impact of predictive analytics on accelerating innovation, improving decision making speed, and automating mundane tasks3.

McGrow3 identified three distinct types of analytics:

Clinical analytics – producing insights to improve treatment outcomes. Examples include clinical pathway prediction, disease progression prediction, predictive risk scoring, and disease diagnosis using medical imaging.

Operational analytics – enhancing the efficiency and effectiveness of systems providing and managing care processes. Examples encompass tracking safety metrics such as length of stay or risk management, monitoring the supply chain, and identifying fraud.

Behavioural analytics –  studying consumer behaviour to inform healthcare delivery. This includes targeting patient engagement, reducing readmissions, and predicting service needs.

To achieve optimal outcomes, predictive analytics relies on accurate and reliable data to generate statistics for machine learning to forecast future events3. Predictions are formulated by applying rules (algorithms) to reliable historical data. The ongoing retraining of algorithms using newer data ensures their continued effectiveness and reliability3. The adage garbage-in, garbage-out holds significant meaning in the realm of machine learning, emphasising the critical importance of data quality18. This underscores the necessity for nurses and midwives to comprehend AI and predictive analytics, including where data is obtained and the rules and processes governing data input and correction. Given their pivotal role in clinical systems and the outputs of predictive tools, nurses and midwives must actively engage in the development and implementation processes3. Greater involvement in these stages enables nurses and midwives to reduce the duplication of data entry, enhance the accuracy of predictive models, and consequently improve patient outcomes.

Conclusion
The integration of AI in nursing and midwifery practice, as underscored by Yin1 and further exemplified by the Australian Alliance for Artificial Intelligence in Healthcare’s national policy roadmap2, presents both promising benefits and persistent challenges. While AI technologies offer enhancements in care assessment, risk management, and overall efficiency, concerns surrounding data security, patient privacy, ethical considerations, and potential impacts on person-centered care persist. As nursing and midwifery play a crucial role in generating patient data used for AI training, it becomes imperative for professionals to actively engage in education, advocate for equity considerations, and contribute to AI governance. By fostering competencies and aligning AI practices with validated standards, nurses and midwives can ensure responsible and impactful integration, shaping a future where AI enhances patient outcomes while upholding ethical and equitable healthcare practices.

Dr Jen Bichel-Findlay FAIDH, CHIA
Alan Scanlon CHIA
6 February 2024

References

  1. Yin, C 2022, ‘Impact of artificial intelligence and machine learning on nursing and midwifery’, Australasian Institute of Digital Health [AIDH], web log post, 10 June, viewed 15 January 2024, https://digitalhealth.org.au/blog/impact-of-artificial-intelligence-and-machine-learning-on-nursing-and-midwifery/
  2. Australian Alliance for Artificial Intelligence in Healthcare [AAAiH] 2023 A national policy roadmap for artificial intelligence in healthcare, AAAiH, https://aihealthalliance.org/wp-content/uploads/2023/11/AAAiH_NationalPolicyRoadmap_FINAL.pdf
  3. McGrow, K 2019, ‘Artificial intelligence: essentials for nursing’, Nursing, vol. 49, no. 9, pp. 46-40, https://journals.lww.com/nursing/Fulltext/2019/09000/Artificial_intelligence__Essentials_for_nursing.12.aspx
  4. Risling, T 2018, ‘Why artificial intelligence needs nursing’, Policy Options, web log post, 5 February, viewed 15 January 2024, https://policyoptions.irpp.org/magazines/february-2018/why-ai-needs-nursing/#:~:text=Nurses%20should%20be%20involved%20in,to%20use%20the%20new%20technology.
  5. Buchanan, C, Howitt, ML, Wilson, R, et al. 2020, ‘Predicted influences of artificial intelligence on the domains of nursing: scoping review’, JMIR Nursing, vol. 3, no. 1, e23939, https://nursing.jmir.org/2020/1/e23939/
  6. Seibert, K, Domhoff, D, Fürstenau, D, et al. 2023, ‘Exploring needs and challenges for AI in nursing care – results of an explorative sequential mixed methods study’, BMC Digital Health, vol. 1, 13(1-17), https://bmcdigitalhealth.biomedcentral.com/articles/10.1186/s44247-023-00015-2#citeas
  7. Siwicki, B 2023,  ‘What nurse leaders need to consider when confronting AI’, Healthcare IT News, web log post, 18 October, viewed 15 January 2024, https://www.healthcareitnews.com/news/what-nurse-leaders-need-consider-when-confronting-ai
  8. Begum, A 2023, ‘The impact of artificial intelligence on nursing: revolutionizing patient care’, Nurses.co.uk, web log post, 8 August, viewed 15 January 2024, https://www.nurses.co.uk/blog/the-impact-of-artificial-intelligence-on-nursing-revolutionizing-patient-care/
  9. Douthit, BJ, Shaw, RJ, Lytle, KS, et al. 2022, ‘Artificial intelligence in nursing’, American Nurse, web log post, 11 January, viewed 15 January 2024, https://www.myamericannurse.com/ai-artificial-intelligence-in-nursing/
  10. Ronquillo, CE, Peltonen, L-M, Pruinelli, L, et al. 2021, ‘Artificial intelligence in nursing: priorities and opportunities from an international invitational think-tank of the Nursing and Artificial Intelligence Leadership Collaborative’, Journal of Advanced Nursing, vol. 77, no 9, pp. 3707-3717, https://onlinelibrary.wiley.com/doi/full/10.1111/jan.14855
  11. Booth, RG, Strudwick, G, McBride, S, et al. 2021 ‘How the nursing profession should adapt for a digital future’, BMJ, vol. 373, no. 1190, pp.1-5, https://www.bmj.com/content/373/bmj.n1190
  12. De Gagne, JC 2023, ‘The state of artificial intelligence in nursing education: past, present, and future directions’, International Journal of Environmental Research and Public Health, vol. 20, no. 6, 4884(1-4), https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10049425/
  13. Lambert, SI, Madi, M, Sopka, S, et al. 2023 ‘An integrative review on the acceptance of AI among healthcare professionals in hospitals’, npj digital medicine, vol. 6, 111(1-14), https://www.nature.com/articles/s41746-023-00852-5
  14. von Gerich, H, Moen, H, Block, LJ, et al. 2022, ‘Artificial intelligence-based technologies in nursing: a scoping literature review of the evidence’, International Journal of Nursing Studies, vol. 127, 104153(1-20), https://www.sciencedirect.com/science/article/pii/S0020748921002984
  15. Starr, B, Dickman, E & Watson, JL 2023 ‘Artificial intelligence: basics, impact, and how nurses can contribute’, Clinical Journal of Oncology Nursing, vol. 27, no. 6, pp. 595-601, https://www.ons.org/cjon/27/6/artificial-intelligence-basics-impact-and-how-nurses-can-contribute
  16. Lattuca, M, Maratta, D, Beffert, U, et  al. 2023 ‘Healthcare AI: a revised Quebec framework for nursing education’, Quality Advancement in Nursing Education, vol. 9, no. 3, Article 2(1-21), https://qane-afi.casn.ca/cgi/viewcontent.cgi?article=1408&context=journal
  17. International Council of Nurses [ICN] 2023, Digital health transformation and nursing practice: position statement, ICN, https://www.icn.ch/sites/default/files/2023-08/ICN%20Position%20Statement%20Digital%20Health%20FINAL%2030.06_EN.pdf
  18. Redman, TC 2018, ‘If your data is bad, your machine learning tools are useless’, Harvard Business Review, web log post, 2 April, viewed 28 January 2024, https://hbr.org/2018/04/if-your-data-is-bad-your-machine-learning-tools-are-useless

Pin It on Pinterest

Share This