Repurposing electronic health data for clinical research and beyond:Experiences from a specialised neurorehabilitation clinic treating patients with acquired brain injury
Bibliographic record
Abstract
Introduction: In Denmark, rehabilitation after acquired brain injury (ABI) is defined in a national guideline [1]. The provision of rehabilitation according to this guideline covers patients with e. g. stroke, traumatic brain injury, encephalitis or encephalopathy. In addition, the complexity of rehabilitation is defined on three service levels (highly specialised, specialised and basic). Across the five administrative regions (figure 1), two inpatient rehabilitation facilities provide comprehensive rehabilitation for severe acquired brain injury at the highly specialised service level. In addition, 14 facilities provide rehabilitation at the specialised service level (moderate to severe injury) [2]. Hammel Neurorehabilitation Centre – University Research Clinic (HNC) is one of two tertiary care inpatient rehabilitation facilities and provides post-acute rehabilitation both on the highly specialised and specialised service levels. Currently, HNC provides 113 beds across 11 wards, including a paediatric and neuro-intensive step-down ward. Annually, circa 750 patients are admitted, with a median length of stay of 49 days (IQR 29–71) in 2022. An electronic healthcare record-system was implemented at HNC in late 2011 when the Central Denmark Region rolled out the corresponding software to all hospitals. Since then, routinely gathered health data have been increasingly repurposed from its primary treatment use to secondary data work e.g. aiding quality assurance, research, and organisational management. Objectives: The objective of this abstract is to present our approach to circular data work. This work will be exemplified with some of the ongoing research, clinical data support and quality assurance projects. Methods: HNC maintains a nightly updated local database containing demographic and clinical information on all former and current patients. After approval from relevant authorities [3], these data can be extracted and used for research and quality assurance purposes. The vision underlying our data work is a circular approach, denominated the Health Data Cycle. Establishing and maintaining the cycle entails support of clinical staff in meaningful documentation to enable repurposing for secondary use, managing the local database, conducting or supporting research projects, and finally collaborating with the clinical staff in the contextual interpretation and preparation for potential implementation. Results: Our local database entails information on approximately 7500 rehabilitation patients with acquired brain injury (figure 2). Currently, three research projects use the database for prognosis research (function at discharge, tube feeding, decannulation) according to the Prognosis Research Strategy Framework [4]. Other research projects use the database to investigate e. g. incidence of and risk factors for in-hospital urinary tract infections or psychotropic drug prescription practices. Conclusion: The repurposing of routinely gathered electronic health data has yielded valuable insights into the rehabilitation practice at our highly specialised inpatient rehabilitation facility. References: 1. Danish Health Authority. Organisation of rehabilitation for adults with acquired brain injury (in Danish). Danish Health Authority 2011. 2. Schmidt M, Schmidt SAJ, Adelborg K et al. The Danish health care system and epi- demiological research: from health care contacts to database records. Clin Epide- miol 2019;11:563–591. doi: 10.2147/CLEP.S179083 3. The Health Act (in Danish) 2022. 4. Steyerberg EW, Moons KGM, van der Windt DA et al. Prognosis Research Strategy (PROGRESS) 3: Prognostic Model Research. PLoS Med 201310(2):e1001381. doi: 10.1371/journal.pmed.1001381
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.047 | 0.142 |
| Meta-epidemiology (narrow) | 0.000 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.004 | 0.004 |
| Scholarly communication | 0.008 | 0.007 |
| Open science | 0.003 | 0.011 |
| Research integrity | 0.003 | 0.004 |
| Insufficient payload (model declined to judge) | 0.005 | 0.002 |
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 source (direct Gemma or distilled Codex), 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".