CoChat

Healthcare professionals’ perspectives on artificial intelligence in patient care: a systematic review of hindering and facilitating factors on different levels

Open access

The DOI matches its Crossref record; 108 of the 110 references it cites were found in a registry or an index; 90 of the 118 sources its claims cite were checked for support.

Abstract

This systematic review synthesized 72 studies and identified 49 facilitating and 42 hindering factors affecting clinicians' perspectives on AI adoption.

Overview

CoChat Summary

TL;DR

This systematic review synthesized 72 studies on healthcare professionals’ perspectives on AI in patient care and organized 49 facilitating and 43 hindering factors using the socio-ecological model. Overall, professionals were generally positive about AI’s potential, but concerns centered on patient harm, loss of human touch, limited knowledge, and implementation barriers.

Key findings

  • The search identified 4,499 records; after screening, 72 studies were included. Across these studies, the authors extracted 49 facilitating factors and 43 hindering factors.
  • Among facilitating factors, the most were at the individual level (14) and institutional level (19); among hindering factors, most were at the individual level (11) and institutional level (13).
  • Common facilitating themes included improved diagnostic quality, better workflows and efficiency, enhanced communication and teamwork, evidence-based and transparent AI development, and system-level benefits such as improved access to care and relief of workforce pressure.
  • Common hindering themes included limited AI knowledge, fear of job replacement or dependency, lack of empathy/human touch, clinical error and patient-harm concerns, insufficient organizational readiness and funding, and legal/privacy/liability uncertainty.

Methods

  • Systematic review conducted according to PRISMA, using the socio-ecological model as the analytic framework.
  • Searched MEDLINE, PsycINFO, and Web of Science on February 14, 2024; included peer-reviewed English- and German-language primary studies from 2017 onward.
  • Included quantitative, qualitative, and mixed-methods studies on healthcare professionals in direct patient care; data were extracted by multiple reviewers, and factors were thematically coded and categorized by SEM level. No formal quality appraisal was performed.

Limitations

  • The paper states that no quality assessment of included studies was performed, which may have allowed lower-quality studies in.
  • The review was limited to three databases, peer-reviewed articles, and English/German studies, which the authors note could introduce publication and language bias; this is stated in the paper.
  • Most included studies were hypothetical rather than real-world AI use, and many were from radiology or unspecified general medicine, which the authors say may limit generalizability to other fields; this is stated in the paper.
  • The text also makes evident that the search ended in February 2024 and the review could not differentiate factors by AI type well because only a few studies fit specific AI categories; this is evident from the reported methods and results.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1186/s12913-025-12664-2

Cites

108 of 108 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Fig. 1 Socio-ecological framework for healthcare professionals’ perspectives on the facilitating and hindering factors to implementing and utilizing AI in healthcare
Fig. 1 Socio-ecological framework for healthcare professionals’ perspectives on the facilitating and hindering factors to implementing and utilizing AI in healthcareCC BY · PubMed Central
Fig. 2 Illustration of the analysis process of a facilitating factor
Fig. 2 Illustration of the analysis process of a facilitating factorCC BY · PubMed Central
Fig. 3 PRISMA flow diagram of studies in the review
Fig. 3 PRISMA flow diagram of studies in the reviewCC BY · PubMed Central
Fig. 4 Field of medicine (reported n > 1) stratified by “type of AI”. AI = artificial intelligence, ML = machine learning, NLPM = natural language processing model, RES = rule-based expert systems
Fig. 4 Field of medicine (reported n > 1) stratified by “type of AI”. AI = artificial intelligence, ML = machine learning, NLPM = natural language processing model, RES = rule-based expert systemsCC BY · PubMed Central

Verification

Claims the paper makes

24 Valid · 3 Not valid · 56 Not checked · 65 Own results and interpretations

71 more claims · 3 with their cited sources checked

Create a free account to see all 148 claims and to verify the 4 references still open.

See all claims — free account
  • It may state two separate claims as one. 40
  • Probably not a substantial claim. 41
  • The quote only partly states this claim. 1

Create a free account to see these claims.

Citations

Cites

Show all 108 references

Research works printed but not linked: 1

  1. Muftić F, Kadunić M, Mušinbegović A, Abd Almisreb A. Exploring Medical Breakthroughs: A Systematic Review of ChatGPT Applications in Healthcare. Southeast Eur J Soft Comp. 2023;12:13–41. https://doi.org/10.21533/scjournal.v12i1.252. · doi:10.21533/scjournal.v12i1.252

Cited by

See all 54

Metadata

Authors
Henzler, D., Schmidt, S., Koçar, A., Herdegen, S., Lindinger, G. L., Maris, M. T., Bak, M. A. R., Willems, D. L., Tan, H. L., Lauerer, M., Nagel, E., Hindricks, G., Dagres, N., & Konopka, M. J.
Affiliations
  • University of Bayreuth (DE)
  • Amsterdam University Medical Centers (NL)
  • Amsterdam UMC Location University of Amsterdam (NL)
  • Deutsches Herzzentrum der Charité (DE)
  • Maastricht University (NL)
Corresponding author
Henzler, Dennis
Venue
BMC Health Services Research
Published
2025-05-01
Volume
25
Issue
1
Pages
633-633
ISSN
1472-6963 (ISSN-L)
OpenAlex
W4410002143
Cited by
54
References
110
Citations by year
2026: 43 · 2025: 11
FWCI
3.19
Type
journal article
Language
en
Open access
Licence
Locations
  • BMC Health Services Research (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • EPub Bayreuth (University of Bayreuth) (acceptedVersion, other-oa)
  • Research Publications (Maastricht University) (submittedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Pure Amsterdam UMC (submittedVersion, cc-by)
  • EPub Bayreuth (University of Bayreuth) (cc-by)
Funders
Horizon 2020 (847999)
Topics
  • Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
  • Physical Sciences › Computer Science › Artificial Intelligence › Artificial Intelligence Applications
  • Health Sciences › Health Professions › Health Information Management › Electronic Health Records Systems
Keywords
artificial intelligence in healthcare, healthcare professionals' perspectives, artificial intelligence adoption, enabling factors, inhibiting factors, socio-ecological model, patient care
MeSH terms
Artificial Intelligence, Attitude of Health Personnel, Patient Care, Health Personnel, Humans

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 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol and qwen3.8-omni-flash).

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

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Healthcare professionals’ perspectives on artificial intelligence in patient care: a systematic review of hindering and facilitating factors on different levels. https://app.cochat.ai/papers/10.1186/s12913-025-12664-2
The paper
Henzler, D., Schmidt, S., Koçar, A., Herdegen, S., Lindinger, G. L., Maris, M. T., Bak, M. A. R., Willems, D. L., Tan, H. L., Lauerer, M., Nagel, E., Hindricks, G., Dagres, N., & Konopka, M. J. (2025). Healthcare professionals’ perspectives on artificial intelligence in patient care: a systematic review of hindering and facilitating factors on different levels. BMC Health Services Research. https://doi.org/10.1186/s12913-025-12664-2

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.