The DOI matches its Crossref record; 28 of the 28 references it cites were found in a registry or an index; 12 of the 32 sources its claims cite were checked for support.
Considering Clinician Competencies for the Implementation of Artificial Intelligence–Based Tools in Health Care: Findings From a Scoping Review
Open accessAbstract
Background: The use of artificial intelligence (AI)–based tools in the care of individual patients and patient populations is rapidly expanding. Objective: The aim of this paper is to systematically identify research on provider competencies needed for the use of AI in clinical settings. Methods: A scoping review was conducted to identify articles published between January 1, 2009, and May 1, 2020, from MEDLINE, CINAHL, and the Cochrane Library databases, using search queries for terms related to health care professionals (eg, medical, nursing, and pharmacy) and their professional development in all phases of clinical education, AI-based tools in all settings of clinical practice, and professional education domains of competencies and performance. Limits were provided for English language, studies on humans with abstracts, and settings in the United States. Results: The searches identified 3476 records, of which 4 met the inclusion criteria. These studies described the use of AI in clinical practice and measured at least one aspect of clinician competence. While many studies measured the performance of the AI-based tool, only 4 measured clinician performance in terms of the knowledge, skills, or attitudes needed to understand and effectively use the new tools being tested. These 4 articles primarily focused on the ability of AI to enhance patient care and clinical decision-making by improving information flow and display, specifically for physicians. Conclusions: While many research studies were identified that investigate the potential effectiveness of using AI technologies in health care, very few address specific competencies that are needed by clinicians to use them effectively. This highlights a critical gap.
Overview
CoChat Summary
TL;DR
This scoping review found that, despite nearly 3,500 citations on AI in health care, only 4 studies met criteria for examining clinician competencies and performance related to AI use. The paper argues this is a major gap in the literature and that competencies for clinicians using AI-based tools remain poorly defined.
Key findings
- The searches returned 3476 unique citations, and only 4 articles met the inclusion criteria after full-text screening.
- The 4 included studies were all conducted in large, academic hospitals and all involved mature AI tools evaluated with practicing physicians, residents, fellows, or nurses.
- Across the 4 studies, AI tools were used to support processes such as MRI knee interpretation, diabetic retinopathy detection, EHR review before encounters, and ICU handovers; 3 studies showed moderate success and 1 had a neutral result.
- Although the studies measured user satisfaction, usability, and performance outcomes, none clearly described the specific AI-related competencies clinicians needed, and only 1 mentioned any form of training.
Methods
- This was a PRISMA-ScR–aligned scoping review with an a priori protocol.
- The authors searched MEDLINE, CINAHL, and the Cochrane Library for English-language, human studies in US settings published from January 1, 2009, to July 22, 2020.
- Screening and data abstraction used standardized forms and thematic synthesis; study quality was assessed with Oxford levels of evidence.
Limitations
- The paper states that limiting the review to US-only sites, English-language studies, and excluding gray literature may have reduced generalizability and missed relevant studies.
- It also notes that the search cutoff date may have omitted newer evidence, and that single screening of titles/abstracts is a limitation.
- Evidence was heterogeneous, so the authors say it was not possible to compare training approaches; this is stated in the paper.
- Evident from the text: the included evidence base was small (4 studies) and of moderate to low quality, with 2 studies at Level 4, which limits how strongly conclusions can be drawn.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures
![Figure 1 Results of literature search, the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) diagram [14]. Summary of articles identified by systematic search queries and tracking of articles that were included and excluded across the study screening phases with reasons for exclusion of full texts provided. AI: artificial intelligence.](https://figures.cochat.ai/paper-figures/ae7344f092964150b27e58d033f31949/considering-clinician-competencies-for-the-implementation-of-artificial-figure-1.jpg)
Verification
Claims the paper makes
8 Valid · 1 Not valid · 5 Not checked · 31 Own results and interpretations
At least one cited source doesn’t support this claim or points to a different paper, and none confirmed it — see the verdicts below.
In the paper
“While numerous authors have called for structured and intentional learning programs, to date, there has been no published framework to guide teaching, learning, and assessing health care students and practitioners in this emerging and transformative domain [7,9-12].”
§ Considering Clinician Competencies for the Implementation of Artificial Intelligence–Based Tools in Health Care: Findings From a Scoping Review > Introduction
What the sources say
- Not checked — no full text of the sourcecites [7] source
- Not checked — no full text of the sourcecites [9] source
- Not checked — no full text of the sourcecites [10] source
- Disputedcites [11] sourceExperimental — accuracy not yet measured
The Discussion’s Future Directions section explicitly proposes a framework for specialized AI training for medical, health informatics, and computer science students (Figure 5). Although the paper says AI curricula were not standardized and competencies had not been determined, its own published framework runs contrary to the categorical claim that no framework had been published. The proposed framework does not explicitly cover assessment or all practitioners. (LLM, conf=0.85)
Checked by a reasoning model — may be wrong.
- Not checked — no full text of the sourcecites [12] source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Using Oxford Levels of Evidence [19] to examine study quality to measure the extent that methodological safeguards (ie, internal study validity) against bias were implemented, 2 studies provided Level 2b evidence as modeling summarizations [20,23], and 2 studies provided Level 4 evidence [21,22].”
§ Considering Clinician Competencies for the Implementation of Artificial Intelligence–Based Tools in Health Care: Findings From a Scoping Review > Results > Study Quality
What the sources say
- Not checked — no full text of the sourcecites [20] source
- Partly supportedcites [23] sourceExperimental — accuracy not yet measured
The Methods (Dataset; page unavailable) describe retrospective collection of knee MRI exams to develop and validate a model, supporting the identification of this paper as a modeling study. Our copy does not assign it an Oxford Level 2b evidence grade, so that part of the citing authors’ classification cannot be verified from this source. (LLM, conf=0.81)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Real-world deployment of AI has caused harms due to data bias (eg, algorithms trained using biased or poor-quality data) and societal bias (eg, algorithmic output reflects societal biases of human developer) [6,26].”
§ Considering Clinician Competencies for the Implementation of Artificial Intelligence–Based Tools in Health Care: Findings From a Scoping Review > Discussion > Principal Findings
What the sources say
- Not checked — no full text of the sourcecites [6] source
- Partly supportedcites [26] Gerke, S, 2020, Artificial Intelligence in Healthcare sourceExperimental — accuracy not yet measured
The ‘Algorithmic fairness and biases’ section (page unavailable) reports real-world examples of algorithmic bias resulting in injustice and identifies unrepresentative datasets and data scientists’ choices as possible causes. It does not establish that the observed harms were caused by those factors, or specifically say that outputs reflected developers’ societal biases. (LLM, conf=0.83)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“All 4 studies were undertaken to characterize the performance of internally developed niche AI software systems when used by health care professionals in specific practice settings (Table 1) [20-23].”
§ Considering Clinician Competencies for the Implementation of Artificial Intelligence–Based Tools in Health Care: Findings From a Scoping Review > Results > Study Characteristics
What the sources say
- Not checked — no full text of the sourcecites [20] source
- Not checked — no full text of the sourcecites [21] source
- Supportedcites [22] source
The Objective section describes HARVEST as a system the authors developed and evaluated to support physicians reviewing patient information. The Materials and methods section describes its in-house natural-language-processing component and an evaluation with resident and fellow physicians using patient cases at NYPH. This confirms the cited paper as one example of the four studies; it does not establish the characterization of the other three. (LLM, conf=0.89)
- Partly supportedcites [23] sourceExperimental — accuracy not yet measured
The Introduction says the authors developed MRNet for knee MRI and evaluated clinicians’ diagnostic performance with its predictions; Methods > Evaluation describes radiologists and surgeons reviewing scans with and without assistance. But the Discussion calls this a retrospective experiment and says future studies must determine its effect in the clinical setting, so the paper does not establish use in an actual practice setting. This verdict addresses only MRNet, not the other three cited studies. (LLM, conf=0.85)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“Only 4 research studies were identified in our search [20-23] that addressed professional competencies observed by this new AI landscape; however, none of the identified studies described new AI-related clinical competencies that had to be developed.”
§ Considering Clinician Competencies for the Implementation of Artificial Intelligence–Based Tools in Health Care: Findings From a Scoping Review > Discussion > Principal Findings
What the sources say
- Not checked — no full text of the sourcecites [20] source
- Not checked — no full text of the sourcecites [21] source
- Supportedcites [22] source
This citation identifies HARVEST as one of the studies in the citing authors’ search; the count of four is the citing authors’ own search result, not a finding attributed to HARVEST. The Materials and methods section describes a study assessing physicians’ ability to extract, compare, synthesize, and recall clinical facts while using a natural-language-processing patient summarizer. The Discussion reports their task performance. No page number is available in our copy. (LLM, conf=0.82)
- Supportedcites [23] source
The Methods (Statistical methods) and Results (Clinical utility of model assistance) show that this is a research study of clinicians’ diagnostic performance with and without AI assistance, supporting its identification as a study relevant to professional competencies in an AI-assisted setting. The count of four studies is the citing authors’ account of their own search, not a finding attributed to this individual paper. (LLM, conf=0.85)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
23 more claims · 2 with their cited sources checked
Create a free account to see all 45 claims and every source verdict.
- Probably not a substantial claim. 14
- It may state two separate claims as one. 7
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“Additional factors might include insufficient training data for the LLM [17,35,36], context misinterpretation, lack of external validation, and overreliance on pattern recognition, all of which could contribute to variable degrees of hallucination, depending on the specific LLM and prompt structure.”
From Reference Hallucination Score for Medical Artificial Intelligence Chatbots: Development and Usability Study (2024) · cites this paper as [35]
Experimental — accuracy not yet measured
Citations
Cites
Preferred Reporting Items for Systematic Reviews and Meta-Analyses: the PRISMA statement
Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., & for the PRISMA Group (2009) · BMJ · doi:10.1136/bmj.b2535 · Cited by 83,823
Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement
Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., & The PRISMA Group (2009) · PLoS Medicine · doi:10.1371/journal.pmed.1000097 · Cited by 64,042 · cited 3×
“AI: artificial intelligence EHR: electronic health records MeSH: medical subject headings PICOST: population, intervention, control, outcomes, study design, and time frame PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses PRISMA-ScR: Preferred Reporting Items for Systematic Reviews and Meta-An…”
InfluentialWang, G., Ma, R., Qiao, G., Wada, K., Aizawa, Y., & Satoh, T. (2015) · Integrative Medicine International · doi:10.1159/000368408 · Cited by 18,657
A typology of reviews: an analysis of 14 review types and associated methodologies
Grant, M. J., & Booth, A. (2009) · Health Information & Libraries Journal · doi:10.1111/j.1471-1842.2009.00848.x · Cited by 10,817 · cited 2×
“AI: artificial intelligence EHR: electronic health records MeSH: medical subject headings PICOST: population, intervention, control, outcomes, study design, and time frame PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses PRISMA-ScR: Preferred Reporting Items for Systematic Reviews and Meta-An…”
MethodInfluentialHigh-performance medicine: the convergence of human and artificial intelligence
Topol, E. J. (2019) · Nature Medicine · doi:10.1038/s41591-018-0300-7 · Cited by 10,450 · cited 2×
“This powerful new generation of AI-based tools will have wide-ranging impacts on the entire health care ecosystem, but concerns about potentially serious technical and ethical liabilities have also emerged [6].”
BackgroundArtificial intelligence in healthcare
Yu, K.-H., Beam, A. L., & Kohane, I. S. (2018) · Nature Biomedical Engineering · doi:10.1038/s41551-018-0305-z · Cited by 2,817
The practical implementation of artificial intelligence technologies in medicine
He, J., Baxter, S. L., Xu, J., Xu, J., Zhou, X., & Zhang, K. (2019) · Nature Medicine · doi:10.1038/s41591-018-0307-0 · Cited by 2,246 · cited 1×
“The scarcity of competencies identified by this scoping review reiterates the need to develop and recommended professional competencies for the use of AI-based tools [27,28].”
BackgroundEthical and legal challenges of artificial intelligence-driven healthcare
Gerke, S., Minssen, T., & Cohen, G. (2020) · Artificial Intelligence in Healthcare · doi:10.1016/b978-0-12-818438-7.00012-5 · Cited by 1,170
Bien, N., Rajpurkar, P., Ball, R. L., Irvin, J., Park, A., Jones, E., Bereket, M., Patel, B. N., Yeom, K. W., Shpanskaya, K., et al. (2018) · PLOS Medicine · doi:10.1371/journal.pmed.1002699 · Cited by 734 · cited 10×
“Upon full-text screening, 4 articles met our inclusion criteria [20-23].”
BackgroundMethodResultInfluentialArtificial Intelligence in Health Care: A Report From the National Academy of Medicine
Matheny, M. E., Whicher, D., & Thadaney Israni, S. (2020) · JAMA · doi:10.1001/jama.2019.21579 · Cited by 628
Research works printed but not linked: 1
- Russell, LL, 2021, Health Professions Education Research Day
Cited by
Barriers and facilitators to utilizing digital health technologies by healthcare professionals
Borges do Nascimento, I. J., Abdulazeem, H., Vasanthan, L. T., Martinez, E. Z., Zucoloto, M. L., Østengaard, L., Azzopardi-Muscat, N., Zapata, T., & Novillo-Ortiz, D. (2023) · npj Digital Medicine · doi:10.1038/s41746-023-00899-4 · Cited by 673
Human-Centered Design to Address Biases in Artificial Intelligence
Chen, Y., Clayton, E. W., Novak, L. L., Anders, S., & Malin, B. (2023) · Journal of Medical Internet Research · doi:10.2196/43251 · Cited by 199
Self-citationAljamaan, F., Temsah, M.-H., Altamimi, I., Al-Eyadhy, A., Jamal, A., Alhasan, K., Mesallam, T. A., Farahat, M., & Malki, K. H. (2024) · JMIR Medical Informatics · doi:10.2196/54345 · Cited by 211
“Additional factors might include insufficient training data for the LLM [17,35,36], context misinterpretation, lack of external validation, and overreliance on pattern recognition, all of which could contribute to variable degrees of hallucination, depending on the specific LLM and prompt structure.”
BackgroundThe Potential of ChatGPT as a Self-Diagnostic Tool in Common Orthopedic Diseases: Exploratory Study
Kuroiwa, T., Sarcon, A. K., Ibara, T., Yamada, E., Yamamoto, A., Tsukamoto, K., & Fujita, K. (2023) · Journal of Medical Internet Research · doi:10.2196/47621 · Cited by 134
AI in the Health Sector: Systematic Review of Key Skills for Future Health Professionals
Gazquez-Garcia, J., Sánchez-Bocanegra, C. L., & Sevillano, J. L. (2025) · JMIR Medical Education · doi:10.2196/58161 · Cited by 140
Metadata
- Authors
- Garvey, K. V., Thomas Craig, K. J., Russell, R., Novak, L. L., Moore, D., & Miller, B. M.
- Affiliations
- Vanderbilt University (US)
- Venue
- JMIR Medical Informatics
- Publisher
- JMIR Publications Inc.
- Published
- 2022-11-16
- Volume
- 10
- Issue
- 11
- Pages
- e37478
- ISSN
- 2291-9694 (ISSN-L)
- PMC
- PMC9713618
- OpenAlex
- W4307973054
- Cited by
- 62
- References
- 28
- Citations by year
- 2026: 21 · 2025: 19 · 2024: 11 · 2023: 9 · 2022: 1
- FWCI
- 0.69
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by
- Locations
- JMIR Medical Informatics (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
- Topics
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Social Sciences › Education › Innovations in Medical Education
- Health Sciences › Health Professions › Health Information Management › Electronic Health Records Systems
- Keywords
- artificial intelligence in healthcare, clinician competence, clinical decision-making, professional development, healthcare workers, physician knowledge and skills, clinical implementation of AI
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-11, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-11 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.