Volume 28, Issue 1 (5-2025)
jha 2025, 28(1): 1-6 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Sheikhtaheri A, Mofarrahi M. Patient safety risks in digital health and artificial intelligence: a call for attention. jha 2025; 28 (1) :1-6
URL: http://jha.iums.ac.ir/article-1-4659-en.html
URL: http://jha.iums.ac.ir/article-1-4659-en.html
1- Health Management and Economics Research Centre, Health Management Research Institute, Iran University of Medical Sciences, Tehran, Iran. & Department of Health Information Management, School of Health Management and Information Sciences, Iran University of Medical Sciences, Tehran, Iran. , sheikhtaheri.a@iums.ac.ir
2- Student Research Committee, Iran University of Medical Sciences, Tehran, Iran.
2- Student Research Committee, Iran University of Medical Sciences, Tehran, Iran.
Keywords: Patient safety, Medical error, Health information technology, digital health, artificial intelligence
Full-Text [PDF 434 kb]
(332 Downloads)
| Abstract (HTML) (907 Views)
Full-Text: (258 Views)
The emergence of digital health technologies and artificial intelligence, such as electronic health records, telemedicine, electronic prescriptions, and clinical decision support systems has significantly improved the quality, efficiency, effectiveness, safety, and accessibility of healthcare services [1-3]. However, these technologies have sometimes compromised patient safety, leading to serious harm and even death, which necessitates special attention [4-6]. Patient safety refers to the prevention or reduction of avoidable harm during medical care [7].
In a 2012 report, 171 incidents related to health information technology were reviewed, among which 124 were classified as having a harmful impact. Of these incidents, eight (6.4%) resulted in actual patient harm. Three incidents were potentially linked to patient death, one required immediate resuscitation, one led to prolonged hospitalization, and three caused injuries that necessitated additional treatment. These errors often stemmed from software malfunctions, incorrect data input, or inadequate alerts [8]. Another report highlighted an increase in mortality from 2.8% to 6.6% following the implementation of a computerized physician order entry (CPOE) system in a pediatric care center. This rise was attributed to delays in order entry, a poor user interface, and system misalignment with clinical workflows [9]. Another study found that CPOE was associated with over 22 types of medication-related risks, including incorrect dosages, dangerous drug interactions, and accidental deletion of critical orders, sometimes resulting in hospitalization or emergency interventions [10]. These findings indicate that even systems designed to reduce errors may pose risks to patient safety due to design or implementation flaws.
Flaws in health information systems can also result in harmful consequences. Inadequate data elements in clinical forms and diagnostic errors caused by such systems may result in unnecessary imaging, treatment delays, and increased threats to patient safety [11]. In Sweden, investigations revealed that flaws in electronic prescription systems such as software functionality issues and user interface problems caused delays in care and posed serious risks to patient health [12]. According to a BBC report, failures in IT systems within the UK's NHS have been linked to delayed surgeries, patient deaths, and over 100 serious harm incidents. These problems were largely attributed to system outages, failures in delivering medical correspondence, and insufficient access to patient records [13].
Artificial intelligence also presents unique challenges in healthcare. For instance, a pneumonia detection system that performed well in two hospitals failed in a third due to data bias, leading to misdiagnoses and treatment delays [14]. A systematic review of the usability of AI in sepsis care found that twenty-two studies exhibited a high risk of bias or serious concerns regarding their applicability [15]. Additionally, in 2019, AI tools used in medical imaging provided incorrect diagnoses in 7% of cases, primarily due to algorithmic flaws and insufficient user training, leading to unnecessary imaging and increased risks for patients [5]. Common issues with AI tools include gender, racial, and geographical biases, as well as inadequate user training, factors that can result in erroneous decisions and compromise patient safety [14].
In Iran, a variety of digital health systems including hospital information systems, electronic health records, telehealth, and e-prescriptions are being implemented. However, most research has primarily focused on the benefits of these technologies, while their risks and adverse impacts on patient safety have received less attention [1-2, 16-19]. This oversight persists despite evidence indicating that such technologies can lead to adverse outcomes including unnecessary hospitalizations, permanent harm, or patient mortality.
Policymakers and healthcare professionals are also expected to revise standards and prioritize patient safety to ensure that technology remains a tool for improved patient safety. Additionally, it is essential for researchers and academic journals to pay more attention to the safety risks associated with these technologies. Awareness of these risks can lead to more careful system design, enhanced user training, and ongoing monitoring of technology performance and safety risks. Journal of Health Administration invites researchers to submit related high quality papers to this journal.
Declerations
Ethical considerations: Not applicable.
Funding: Not applicable.
Conflict of interest: Abbas sheikhtaheri is the Editor-in-Chief of the Journal of Health Administration. There are no other conflicts to declare.
Authors' contributions: A.S: Conceptualization, data collection, writing – review and editing, final approval; M.M: Conceptualization, data collection, writing – review and editing, final approval.
Consent for publication: Not applicable.
Data availability: Not applicable.
AI declaration: Not applicable.
Acknowledgments: Not applicable.
In a 2012 report, 171 incidents related to health information technology were reviewed, among which 124 were classified as having a harmful impact. Of these incidents, eight (6.4%) resulted in actual patient harm. Three incidents were potentially linked to patient death, one required immediate resuscitation, one led to prolonged hospitalization, and three caused injuries that necessitated additional treatment. These errors often stemmed from software malfunctions, incorrect data input, or inadequate alerts [8]. Another report highlighted an increase in mortality from 2.8% to 6.6% following the implementation of a computerized physician order entry (CPOE) system in a pediatric care center. This rise was attributed to delays in order entry, a poor user interface, and system misalignment with clinical workflows [9]. Another study found that CPOE was associated with over 22 types of medication-related risks, including incorrect dosages, dangerous drug interactions, and accidental deletion of critical orders, sometimes resulting in hospitalization or emergency interventions [10]. These findings indicate that even systems designed to reduce errors may pose risks to patient safety due to design or implementation flaws.
Flaws in health information systems can also result in harmful consequences. Inadequate data elements in clinical forms and diagnostic errors caused by such systems may result in unnecessary imaging, treatment delays, and increased threats to patient safety [11]. In Sweden, investigations revealed that flaws in electronic prescription systems such as software functionality issues and user interface problems caused delays in care and posed serious risks to patient health [12]. According to a BBC report, failures in IT systems within the UK's NHS have been linked to delayed surgeries, patient deaths, and over 100 serious harm incidents. These problems were largely attributed to system outages, failures in delivering medical correspondence, and insufficient access to patient records [13].
Artificial intelligence also presents unique challenges in healthcare. For instance, a pneumonia detection system that performed well in two hospitals failed in a third due to data bias, leading to misdiagnoses and treatment delays [14]. A systematic review of the usability of AI in sepsis care found that twenty-two studies exhibited a high risk of bias or serious concerns regarding their applicability [15]. Additionally, in 2019, AI tools used in medical imaging provided incorrect diagnoses in 7% of cases, primarily due to algorithmic flaws and insufficient user training, leading to unnecessary imaging and increased risks for patients [5]. Common issues with AI tools include gender, racial, and geographical biases, as well as inadequate user training, factors that can result in erroneous decisions and compromise patient safety [14].
In Iran, a variety of digital health systems including hospital information systems, electronic health records, telehealth, and e-prescriptions are being implemented. However, most research has primarily focused on the benefits of these technologies, while their risks and adverse impacts on patient safety have received less attention [1-2, 16-19]. This oversight persists despite evidence indicating that such technologies can lead to adverse outcomes including unnecessary hospitalizations, permanent harm, or patient mortality.
Policymakers and healthcare professionals are also expected to revise standards and prioritize patient safety to ensure that technology remains a tool for improved patient safety. Additionally, it is essential for researchers and academic journals to pay more attention to the safety risks associated with these technologies. Awareness of these risks can lead to more careful system design, enhanced user training, and ongoing monitoring of technology performance and safety risks. Journal of Health Administration invites researchers to submit related high quality papers to this journal.
Declerations
Ethical considerations: Not applicable.
Funding: Not applicable.
Conflict of interest: Abbas sheikhtaheri is the Editor-in-Chief of the Journal of Health Administration. There are no other conflicts to declare.
Authors' contributions: A.S: Conceptualization, data collection, writing – review and editing, final approval; M.M: Conceptualization, data collection, writing – review and editing, final approval.
Consent for publication: Not applicable.
Data availability: Not applicable.
AI declaration: Not applicable.
Acknowledgments: Not applicable.
Type of Study: Letter to Editor |
Subject:
Health Information Management
Received: 2025/05/18 | Accepted: 2025/08/5 | Published: 2025/08/31
Received: 2025/05/18 | Accepted: 2025/08/5 | Published: 2025/08/31
References
1. Najafi Sarband S, Amanzadeh M, Naimi R, Mohammad Shahi J, Mahdavi A. Benefits and challenges of electronic pharmaceutical prescriptions. Journal of Health Administration. 2024; 27 (3): 103-117. [In Persian] doi: http://doi.org/10.61186/jha.27.3.103 [DOI:10.61186/jha.27.3.103]
2. Amlashi SR, Leyli EK, Sheikhtaheri A. Physicians' and pharmacists' viewpoint on ambulatory electronic prescription system. Journal of Health Administration. 2022;25(3):108-14. [In Persian] doi: http://doi.org/10.22034/25.3.108
3. Koonin LM, Hoots B, Tsang CA, Leroy Z, Farris K, Jolly T, et al. Trends in the use of telehealth during the emergence of the COVID-19 pandemic-United States, January-March 2020. MMWR Morbidity and mortality weekly report. 2020;69(43):1595-9. doi: http://doi.org/10.15585/mmwr.mm6943a3 [DOI:10.15585/mmwr.mm6943a3]
4. Taheri Moghadam S, Hooman N, Sheikhtaheri A. Patient safety classifications for health information technology (HIT) and medical devices: a review on available systems. Studies in Health Technology and Informatics. 2022; 293:153-60. doi: http://doi.org/10.3233/SHTI220362 [DOI:10.3233/SHTI220362]
5. Jabin MSR. The need for a refined classification system and national incident reporting system for health information technology-related incidents. Frontiers in Digital Health. 2024;6:1422396. [DOI:10.3389/fdgth.2024.1422396]
6. Kim MO, Coiera E, Magrabi F. Problems with health information technology and their effects on care delivery and patient outcomes: a systematic review. Journal of American Medical Informatics Association. 2017 Mar 1;24(2):246-50. 28011595. doi: [DOI:10.1093/jamia/ocw154]
7. Taheri Moghadam S, Sheikhtaheri A, Hooman N. Patient safety classifications, taxonomies and ontologies, part 2: A systematic review on content coverage. Journal of Biomedical Informatics. 2023 (148):104549. doi: [DOI:10.1016/j.jbi.2023.104549]
8. Wergin J. Medical errors and health IT: what does the data say? [Internet] Pennsylvania (America): Emergency Care Research Institude (ECRI). 2023. Available from: https://home.ecri.org/blogs/ecri-blog/medical-errors-and-health-it-what-does-the-data-say/
9. Han YY, Carcillo JA, Venkataraman ST, Clark RS, Watson RS, Nguyen TC, et al. Unexpected increased mortality after implementation of a commercially sold computerized physician order entry system. Pediatrics. 2005;116(6):1506-12. doi: [DOI:10.1542/peds.2005-1287]
10. Koppel R, Metlay JP, Cohen A, Abaluck B, Localio AR, Kimmel SE, et al. Role of computerized physician order entry systems in facilitating medication errors. JAMA. 2005;293(10):1197-203. doi: [DOI:10.1001/jama.293.10.1197]
11. Hannaford N, Mandel C, Crock C, Buckley K, Magrabi F, Ong M, et al. Learning from incident reports in the Australian medical imaging setting: handover and communication errors. British Journal of Radiology. 2013;86(1022):20120336.. doi: [DOI:10.1259/bjr.20120336]
12. Rahman Jabin MS, Hammar T. Issues with the Swedish e-prescribing system - An analysis of health information technology-related incident reports using an existing classification system. Digital Health. 2022;8:20552076221131139. doi: [DOI:10.1177/20552076221131139]
13. Barbour S, Wright N, McNamee A, Roxby P. NHS computer issues linked to patient harm. 2024. Available from: https://www.bbc.com/news/articles/c4nn0vl2e78o
14. Muley A, Muzumdar P, Kurian G, Basyal GP. Risk of AI in healthcare: a comprehensive literature review and study framework. Asian Journal of Medicine and Health. 2023;21:276-91. doi:
https://doi.org/10.9734/ajmah/2023/v21i10903 [DOI:10.9734/AJMAH/2023/v21i10903]
15. Schinkel M, Paranjape K, Nannan Panday RS, Skyttberg N, Nanayakkara PWB. Clinical applications of artificial intelligence in sepsis: a narrative review. Computer in Biology and Medicine. 2019;115:103488. doi: [DOI:10.1016/j.compbiomed.2019.103488]
16. Abbasi H, Rahimi B, Jebraeily M, Nourani A. The extent of achieving the expected goals of the Ministry of Health in the implementation of the electronic health record system from the end-users' viewpoint in West Azerbaijan province. Journal of Health Administration. 2023;26(1):9-28 [In Persian] doi: [DOI:10.22034/26.1.11]
17. Kheradranjbar M, Khamseh A, Iranban Fard SJ. Evaluation of health services portfolio management based on information technology using adaptive neuro-fuzzy inference approach. Journal of Health Administration 2024; 27 (3) :54-69. [In Persian] doi: [DOI:10.61186/jha.27.3.54]
18. Sarbaz M, Mousavi Baigi SF, Salehzadea Z, Norouzi Aval R, Esmaeili M, Kimiaf K. Security barriers and facilitators in the use of mobile health applications from the perspective of paramedical students at Mashhad university of medical sciences: a descriptive cross-sectional study. Journal of Health Administration. 2024;23(3):1-16. [In Persian] [DOI:10.61186/jha.27.3.1]
19. Rahimloo L, Ebrahimi K, Mehrtak M, Mohammadnia A, Lotfnezhad H. Evaluating the integrated health system (SIB) from the users' perspectives. Journal of Health Administration. 2024;27(2):74-89. [In Persian] doi: [DOI:10.61186/jha.27.2.74]
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
Site Keywords
Health, Administration, Scientific Article, Iran University of Medical Sciences,
