Bibliographic record
Abstract
Drawing on different types of knowledge—from clinical research to the experiential knowledge of patients and healthcare professionals—holds significant potential to lead to better quality and relevant healthcare. However, systematically integrating such knowledge has been a long-standing challenge within the framework of Evidence-Based Medicine (EBM) and, more specifically, in the development of clinical and public health guidelines. This highlights the need to better understand and facilitate the integration of diverse forms of knowledge, particularly experiential knowledge, into EBM and guideline development. This thesis explores the potential of digital methods, especially AI-based methods, as a different and potentially innovative approach to supporting this integration. It thereby also examines the insights that experimenting with these methods can offer regarding the inclusion and marginalization of experiential knowledge in healthcare standards. To explore the potential and relevance of these methods in a reflexive and differentiated way, it is argued that they must be situated within the long-standing body of work in sociology and science and technology studies (STS) on knowledge exclusion and inclusion in the field of health. I adopt a transdisciplinary and STS Making & Doing approach, generating practical and theoretical insights into knowledge integration and the role of digital methods through experimenting with these methods in collaboration with various actors during the development of the Dutch public health guidelines on COVID-19 vaccination, scabies, and transgender care. In each case, AI-based methods, particularly from the field of natural language processing, were developed and applied to identify and analyze experiential knowledge shared online by patients, health professionals, and citizens—rendering this knowledge accessible for guideline development. These were complemented by various qualitative methods, including interviews, participant observation, and autoethnography, to better understand the dynamics of knowledge integration and exclusion. The findings demonstrate that AI-based methods are effective in gaining valuable insights into the experiential knowledge of patients, healthcare professionals, and citizens—insights that would otherwise be difficult to obtain. However, their meaningful application requires careful consideration of the dynamics of knowledge inclusion and exclusion. The analyses identify some of the multifaceted mechanisms through which experiential knowledge is marginalized, as well as strategies to support its inclusion—demonstrating that technological solutions alone, including AI-based methods, are insufficient to address the challenge of integrating diverse knowledge in health contexts. Based on these findings, the thesis offers three key lessons for integrating experiential knowledge in epistemically charged healthcare settings: (1) digital methods must be accompanied by dialogical engagement and collaborative knowledge work among the various actors involved; (2) knowledge integration should be conceptualized more broadly, particularly embracing approaches that allow differences to remain unresolved rather than forcing reconciliation and consensus; and (3) strategies that subtly deconstruct or bypass rigid knowledge categories can be as effective as those that explicitly foreground difference. Overall, this thesis seeks to advance conceptual and practical insights into how diverse forms of knowledge can be systematically and meaningfully integrated into medical knowledge production and healthcare practice, as well as the opportunities that innovative digital methods may offer to further support this integration.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".