DSDavid Sundemo

David Sundemo

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


David Sundemo has a total of 110 co-authors. Here are the top co-authors by number of shared publications:

  • Kristian Samuelsson

    Department of Orthopaedics, Institute of Clinical Sciences, Gothenburg, Sweden

    13 shared publications
  • Eric Hamrin Senorski

    <!--2--> Department of Health and Rehabilitation, Institute of Neuroscience and Physiology, Sahlgrenska Academy University of Gothenburg Göteborg Sweden

    8 shared publications
  • Volker Musahl

    Department of Orthopaedic Surgery, UPMC Center for Sports Medicine, Pittsburgh, Pennsylvania, USA

    8 shared publications
  • Eduard Alentorn-Geli

    Fundacion Garcia Cugat

    7 shared publications
  • Eleonor Svantesson

    <!--1--> Department of Orthopaedics, Institute of Clinical Sciences, The Sahlgrenska Academy University of Gothenburg Gothenburg Sweden

    7 shared publications
  • Eric Hamrin Senorski

    Unit of Physiotherapy, Department of Health and Rehabilitation, Institute of Neuroscience and Physiology, Sahlgrenska Academy University of Gothenburg Gothenburg Sweden

    7 shared publications

Latest Publications

  • Open Access

    Generative AI in Primary Care: A qualitative study of UK General Practitioners’ Views (Preprint)

    by Charlotte Blease, Anna Kharko, Carolina Garcia Sanchez, Jens Gaab, Stephanie Kumpunen, David Sundemo, John Torous

    BACKGROUND Generative artificial intelligence (GenAI) is increasingly entering primary care, yet evidence has focused largely on adoption rates rather than how clinicians interpret the role, risks, and professional implications of these technologies. Quantitative findings from the third wave of the GPAI-UK survey suggest rapidly increasing use among UK general practitioners (GPs), alongside uncertainty about how AI may reshape clinical work. Qualitative insight is needed to understand how GPs are making sense of this change. OBJECTIVE This study aimed to explore how UK GPs conceptualise the role of GenAI in general practice, including perceived benefits, risks, and implications for professional identity and the future of the GP role. METHODS We conducted an exploratory qualitative analysis of free-text responses from the 2026 GPAI-UK survey, an annual online, convenience sample survey administered via Doctors.net.uk as part of a scheduled omnibus. The survey included a single optional open-ended question inviting comments on AI and general practice. Responses were analysed using an inductive, descriptive thematic approach. RESULTS A total of 232 GPs provided free-text comments. Four overarching themes were identified. First, GPs largely framed GenAI as a pragmatic administrative tool, particularly for documentation, rather than a substitute for clinical judgment. Second, respondents expressed concrete concerns about risks and harms, including errors, hallucinations, deskilling, medico-legal accountability and environmental concerns. Third, many situated AI within broader political and economic narratives, expressing distrust toward institutional and commercial drivers of adoption. Finally, GPs anticipated substantial transformation of general practice while expressing ambivalence and uncertainty about its direction, with some respondents feeling ill-equipped to offer predictions. CONCLUSIONS Although based on brief, self-selected free-text responses from a convenience sample, this qualitative study offers original insight into how UK GPs are making sense of GenAI as it becomes embedded in primary care. Situated within the third consecutive year of a national survey program, the findings capture evolving clinician attitudes and experiences, including how GPs conceptualize the appropriate role of AI, articulate emerging risks, and anticipate future professional change.

  • Open Access

    Acceptable accuracy for medical AI: a survey of physicians and the general population in Sweden

    by Rasmus Arvidsson, Jonathan Widén, Lina Al-Naasan, Ronny Kent K Gunnarsson, Peter Nymberg, Charlotte R Blease, Anna Moberg, Pär-Daniel Sundvall, Carl Wikberg, David Sundemo

    Objectives: To identify the lowest sensitivity and specificity that physicians and the general population consider acceptable for medical artificial intelligence (AI), relative to current human performance. Methods: In a nationwide, cross-sectional survey in Sweden, 2025, random samples of 500 physicians and 500 adults from the general population were mailed a questionnaire presenting three vignettes (chest pain triage, sore throat triage, ECG myocardial infarction detection) with the corresponding human performance. Participants reported the maximum number of cases an AI should be allowed to miss or over-refer. Results: Response rates were 45% among physicians and 31% in the general population. Both groups demanded higher AI accuracy than the human benchmark for all cases. In the chest pain triage vignette, the nurse correctly referred 84 of 100 true emergencies; physicians required the AI to correctly refer 11 additional patients (95% sensitivity) and the general population demanded referral of 16 additional patients (100% sensitivity) (p<0.001 for both groups). Among 100 patients not requiring referral, the nurse would mistakenly refer 66. Both groups required the AI to reduce unnecessary referrals by 16 (50% specificity) (p<0.001). A similar pattern was observed in the other vignettes. Discussion: The accuracy thresholds required by the respondents exceed the performance of many existing systems, although emerging AI research shows promise in narrowing the gap. Conclusion: Physicians and the general population require medical AI systems to outperform human clinicians. When implementing AI in healthcare settings, early engagement with both groups may be necessary to align expectations with real-world system performance.

  • Open Access

    Patients are not waiting for permission: the rise of the AI-empowered patient

    by Sara Riggare, David Sundemo, Marcus Lewis, Charlotte Blease


  • Open Access

    Exploring Clinician Perspectives on Artificial Intelligence in Primary Care: Qualitative Systematic Review and Meta-Synthesis

    by Robin Bogdanffy, Alisa Mundzic, Peter Nymberg, David Sundemo, Anna Moberg, Carl Wikberg, Ronny Kent Gunnarsson, Jonathan Widén, Pär-Daniel Sundvall, Artin Entezarjou

    Background: Recent advances have highlighted the potential of artificial intelligence (AI) systems to assist clinicians with administrative and clinical tasks, but concerns regarding biases, lack of regulation, and potential technical issues pose significant challenges. The lack of a clear definition of AI, combined with limited focus on qualitative research exploring clinicians' perspectives, has limited the understanding of perspectives on AI in primary health care settings. Objective: This review aims to synthesize current qualitative research on the perspectives of clinicians on AI in primary care settings. Methods: A systematic search was conducted in MEDLINE (PubMed), Scopus, Web of Science, and CINAHL (EBSCOhost) databases for publications from inception to February 5, 2024. The search strategy was designed using the Sample, Phenomenon of Interest, Design, Evaluation, and Research type (SPIDER) framework. Studies were eligible if they were published in English, peer-reviewed, and provided qualitative analyses of clinician perspectives on AI in primary health care. Studies were excluded if they were gray literature, used questionnaires, surveys, or similar methods for data collection, or if the perspectives of clinicians were not distinguishable from those of nonclinicians. A qualitative systematic review and thematic synthesis were performed. The Grading of Recommendations Assessment, Development and Evaluation-Confidence in Evidence from Reviews of Qualitative Research (GRADE-CERQual) approach was used to assess confidence in the findings. The CASP (Critical Appraisal Skills Program) checklist for qualitative research was used for risk-of-bias and quality appraisal. Results: A total of 1492 records were identified, of which 13 studies from 6 countries were included, representing qualitative data from 238 primary care physicians, nurses, physiotherapists, and other health care professionals providing direct patient care. Eight descriptive themes were identified and synthesized into 3 analytical themes using thematic synthesis: (1) the human-machine relationship, describing clinicians’ thoughts on AI assistance in administration and clinical work, interactions between clinicians, patients, and AI, and resistance and skepticism toward AI; (2) the technologically enhanced clinic, highlighting the effects of AI on the workplace, fear of errors, and desired features; and (3) the societal impact of AI, reflecting concerns about data privacy, medicolegal liability, and bias. GRADE-CERQual assessment rated confidence as high in 15 findings, moderate in 5 findings, and low in 1 finding. Conclusions: Clinicians view AI as a technology that can both enhance and complicate primary health care. While AI can provide substantial support, its integration into health care requires careful consideration of ethical implications, technical reliability, and the maintenance of human oversight. Interpretation is constrained by heterogeneity in qualitative methods and the diversity of AI technologies examined across studies. More in-depth qualitative research on the effects of AI on clinicians’ careers and autonomy could prove helpful for the future development of AI systems.


  • Open Access

    Ethical risks of AI-enabled remote patient monitoring for COPD: a multi-dimensional use case analysis

    by D Behdadi, Lowie E. G. W. Vanfleteren, David Sundemo

    Artificial intelligence (AI)-enabled remote patient monitoring (RPM) is promoted as a solution to rising pressures in health care, including personnel shortages and the growing burden associated with population aging and chronic disease management. Yet, the ethical implications of deploying adaptive systems in routine care remain underexamined at the level of specific, situated use cases. This article examines the ethical risks of MonitAir, an AI-enabled RPM system for chronic obstructive pulmonary disease (COPD) in a Swedish health care setting. Drawing on the Three Domains, Six Levels (3D6L) framework, we identify epistemic, normative and traceability-related risks across six levels of abstraction. The article offers, to our knowledge, the first operationalization of the 3D6L as an analytic tool for screening ethical risks. We argue that screening AI-enabled health care technology with 3D6L clarifies how ethical risks manifest across levels, from individual patients and patient–clinician relationships to organizational and sectoral contexts. In addition, the framework’s minimal normativity allows alignment with context-sensitive principles and guidelines. Through this analysis, we identify ethical risks related to data bias, intelligibility of outputs, uneven access and blurred responsibility, including redistributive and role-shifting effects. While MonitAir may support earlier detection of exacerbations, its implementation in Swedish COPD care may also reproduce and amplify existing health disparities or overburden patients without sufficient support. We demonstrate how structured ethical screening makes visible concerns typically overlooked by the optimization discourse in digital health. Finally, we argue that open-ended evaluation of ethical risks of AI-enabled DHT provides a valuable early phase that complements ethical assessment, without collapsing into checklist compliance.


  • Open Access

    Exploring Primary Care Patients’ Perspectives on Artificial Intelligence: Systematic Literature Review and Qualitative Meta-Synthesis

    by Alisa Mundzic, Robin Bogdanffy, David Sundemo, Pär-Daniel Sundvall, Jonathan Widén, Peter Nymberg, Carl Wikberg, Anna Moberg, Ronny Gunnarsson, Artin Entezarjou

    Background: The introduction of artificial intelligence (AI) in health care holds great promise, offering the potential to alleviate physicians’ workloads and allocate more time for patient interactions. After the emergence of large language models (LLMs), interest in AI has surged in the health care sector, including within primary care. However, patients have expressed concerns about the ethical implications and use of AI in primary care. Understanding patients’ perspectives on using AI in primary care is crucial for its effective integration. Despite this, few studies have addressed patients’ perspectives on using AI in primary care. Objective: This study aimed to synthesize qualitative research on primary care patients’ perspectives regarding the use of AI, including LLMs, in primary care. Methods: A qualitative systematic review, using thematic analysis, was performed in accordance with PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. Databases, including PubMed, Scopus, Web of Science, CINAHL, and PsycINFO, were searched from inception to February 5, 2024. Eligible studies (1) used a qualitative interview research design, (2) explored primary care patients’ perspectives on the use of AI in primary care, (3) were written in English, and (4) were published in peer-reviewed scientific journals. Quantitative studies, gray literature, surveys, and studies lacking depth in qualitative analysis were excluded. The Critical Appraisal Skills Program (CASP) checklist was used for quality assessment. Results: Of 1004 studies screened, 6 were included, comprising 170 patients aged 13-91 years from 3 countries. Three themes emerged: “The Relationship with and Actions of AI Systems,” “Implementing AI responsibly,” and “Training Physicians and Artificial Minds.” Patients acknowledged AI’s potential benefits but advocated for clinician oversight, safety frameworks, and the preservation of patient autonomy. Conclusions: This systematic review provides an understanding of patients’ perspectives on AI in primary care. We identified heterogeneity in AI definitions across studies. Further research is needed on patients’ perspectives across different countries. Notably, our synthesis revealed a significant research gap, as none of the included studies particularly explored patients’ perspectives on LLMs, highlighting an important area for future research.


  • Open Access

    General practitioners’ opinions of generative artificial intelligence in the UK: An online survey

    by Anna Kharko, Carolina Garcia Sanchez, Josefin Hagström, Jens Gaab, Cosima Locher, Brian McMillan, David Sundemo, Charlotte Blease

    Background: The rapid rise of large language model-based generative artificial intelligence (AI) tools, such as chatbots and AI scribes, has prompted interest in their clinical applicability. This study examined the perspectives of general practitioners (GPs) in the UK regarding the impact of generative AI on clinical practice. Methods: In January 2025, a national online survey of 1005 GPs, stratified by region, was administered via the UK's largest clinical marketing platform. The survey gathered data on attitudes towards generative AI, its perceived effects on clinical tasks, care delivery and respondent demographics. Group differences and correlations between opinions and respondent characteristics were analysed. Results: Among the 1005 respondents, 50% were male, 54% aged 46 or older, and 83% practised in England. Only 11% were encouraged by their employer to use generative AI tools, and 5% had received training. Most agreed that AI could improve documentation (69%) and patient information gathering (57%), and 59% expected increased efficiency in care. A similar proportion (59%) anticipated that patients may rely on AI instead of seeking medical care. The majority (79%) agreed that training is needed, while 71% disagreed that AI could enhance empathic communication. Male GPs were more likely to view AI as beneficial across tasks, while female GPs more often emphasised training needs. Older GPs expressed concern about potential inequities stemming from AI use. Conclusions: While GPs were cautiously optimistic about generative AI – particularly for documentation and data collection – scepticism persisted. In particular, around empathy and equity, highlighting the need for guidance on AI integration into primary care.


  • Open Access

    Exploring Primary Care Patients’ Perspectives on Artificial Intelligence: A systematic literature review and qualitative meta-synthesis (Preprint)

    by Alisa Mundzic, Robin Bogdanffy, Artin Entezarjou, David Sundemo, Pär-Daniel Sundvall, Jonathan Widén, Peter Nymberg, Carl Wikberg, Anna Moberg, Ronny Gunnarsson

    BACKGROUND The progression of Artificial Intelligence (AI) in healthcare holds great promise, offering the potential to alleviate physicians’ workloads and allocate more time for patient interactions. However, implementing AI in primary care also carries inherent risks and ethical obligations that must be carefully considered. Understanding patients’ perspectives on using AI in primary care is crucial for its effective integration. OBJECTIVE The primary objective of this qualitative meta-synthesis is to synthesize current qualitative research on primary care patients’ perspectives regarding the use of AI and Large Language Models (LLMs) in primary care. METHODS A qualitative meta-synthesis, using thematic analysis, was performed in accordance with PRISMA guidelines. PubMed and Scopus were searched on February 5th, 2024. Quality assessment was carried out using the Critical Appraisal Skills Program (CASP) checklist for qualitative research. RESULTS A total of 861 studies were screened for title and abstract, and six studies were included. Three themes emerged: “The Relationship with and Actions of a Robot”, “Implementing AI responsibly”, and “Training Physicians and Artificial Minds”. Findings indicate that while patients acknowledge the potential benefits of AI, concerns persist regarding the loss of human interaction in medical care, data security, and the ethical implications of AI implementation. CONCLUSIONS Patients recognized the advantages of AI in alleviating administrative tasks and enhancing physician effectiveness, yet raised substantial concerns that should be addressed prior to its adoption in primary care. Patients emphasize the importance of maintaining human interaction and advocate for strong clinician oversight, robust safety frameworks, and preserving patient choice and autonomy. Addressing patient concerns and understanding associated risks are key to ensuring AI's safe and effective use in healthcare.

  • Open Access

    Exploring Clinician Perspectives on Artificial Intelligence in Primary Care: a qualitative systematic review and meta synthesis (Preprint)

    by Robin Bogdanffy, Alisa Mundzic, Peter Nymberg, David Sundemo, Anna Moberg, Carl Wikberg, Ronny Kent Gunnarsson, Jonathan Widén, Pär-Daniel Sundvall, Artin Entezarjou

    BACKGROUND Recent advances have highlighted the potential of artificial intelligence (AI) systems in assisting clinicians with administrative and clinical tasks, but concerns regarding biases, lack of regulation, and potential technical issues pose significant challenges. The lack of a clear definition of AI, combined with a limited focus on qualitative research exploring clinicians' perspectives has limited the understanding of perspectives on AI in primary healthcare settings. OBJECTIVE This review aims to synthesize the current qualitative research on the perspectives of clinicians on artificial intelligence in primary care settings. METHODS A qualitative systematic review and meta synthesis using thematic analysis was performed, adhering to PRISMA guidelines. Searches were conducted in PubMed and Scopus. The Critical Appraisal Skills Program checklist for qualitative research was used for quality appraisal. RESULTS Twelve articles were included. Three themes emerged: the human-machine relationship, the technologically enhanced clinic, and the societal effects of AI. CONCLUSIONS Clinicians view AI as a technology that either could enhance or complicate primary healthcare. While AI can provide substantial support, its integration into healthcare requires careful consideration of ethical implications, technical reliability, and the maintenance of human oversight. More in-depth qualitative research on the effects of AI on clinicians’ careers and autonomy could prove helpful for the future development of AI systems.

  • Open Access

    ChatGPT (GPT-4) versus doctors on complex cases of the Swedish family medicine specialist examination: an observational comparative study

    by Rasmus Arvidsson, Ronny Gunnarsson, Artin Entezarjou, David Sundemo, Carl Wikberg

    Background Recent breakthroughs in artificial intelligence research include the development of generative pretrained transformers (GPT). ChatGPT has been shown to perform well when answering several sets of medical multiple-choice questions. However, it has not been tested for writing free-text assessments of complex cases in primary care. Objectives To compare the performance of ChatGPT, version GPT-4, with that of real doctors. Design and setting A blinded observational comparative study conducted in the Swedish primary care setting. Responses from GPT-4 and real doctors to cases from the Swedish family medicine specialist examination were scored by blinded reviewers, and the scores were compared. Participants Anonymous responses from the Swedish family medicine specialist examination 2017–2022 were used. Outcome measures Primary: the mean difference in scores between GPT-4’s responses and randomly selected responses by human doctors, as well as between GPT-4’s responses and top-tier responses by human doctors. Secondary: the correlation between differences in response length and response score; the intraclass correlation coefficient between reviewers; and the percentage of maximum score achieved by each group in different subject categories. Results The mean scores were 6.0, 7.2 and 4.5 for randomly selected doctor responses, top-tier doctor responses and GPT-4 responses, respectively, on a 10-point scale. The scores for the random doctor responses were, on average, 1.6 points higher than those of GPT-4 (pConclusion In complex primary care cases, GPT-4 performs worse than human doctors taking the family medicine specialist examination. Future GPT-based chatbots may perform better, but comprehensive evaluations are needed before implementing chatbots for medical decision support in primary care.