Essential Requirements for Establishing and Operating Data Trusts
Bibliographic record
Abstract
INTRODUCTION: Increasingly, the label "data trust" is being applied to repeatable mechanisms or approaches to sharing data in a timely, fair, safe, and equitable way. However, there is an absence of practical guidance regarding how to establish and operate a data trust. AIM AND APPROACH: In December 2019, the Canadian Institute for Health Information and the Vector Institute for Artificial Intelligence convened a working meeting of 19 people representing 15 Canadian organizations/initiatives involved in data sharing, most of which focus on public sector health data. The objective was to identify essential requirements for the establishment and operation of data trusts in the Canadian context. Preliminary requirements were discussed during the meeting and then refined as authors contributed to this manuscript. RESULTS: Twelve minimum specification requirements ("min specs") for data trusts were identified. The foundational min spec is that data trusts must meet all legal requirements, including legal authority to collect, hold or share data. In addition, there was agreement that data trusts must have (i) an accountable governing body to ensure that the data trust achieves its stated purpose and is transparent, (ii) comprehensive data management including clear processes and qualified individuals responsible for the collection, storage, access, disclosure and use of data, (iii) training and accountability requirements for all data users and (iv) ongoing public and stakeholder engagement. CONCLUSIONS: Practical guidance for the establishment and operation of data trusts was articulated in the form of 12 min specs requirements. The 12 min specs are a starting point. Future work to refine and strengthen them with members of the public, companies, and additional research data stakeholders from within and outside of Canada, is recommended.
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 | 0.132 | 0.241 |
| Meta-epidemiology (narrow) | 0.001 | 0.002 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.004 | 0.003 |
| Science and technology studies | 0.009 | 0.011 |
| Scholarly communication | 0.016 | 0.013 |
| Open science | 0.005 | 0.010 |
| Research integrity | 0.008 | 0.007 |
| Insufficient payload (model declined to judge) | 0.006 | 0.004 |
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".