Health Data for Linguistic Minority Group Research in Canada: Proof-of-Concept Centralized Health Care Metadata Repository Development and Usability Study (Preprint)
Bibliographic record
Abstract
BACKGROUND Language barriers between Canadian patients and health care providers are associated with poorer health outcomes, including decreased patient safety and quality of care, misdiagnosis and longer treatment initiation times, and increased mortality. However, research exploring language as a social determinant of health is limited, as Canadian health data are scattered across many jurisdictions, each with its own policies and procedures. This fragmentation makes it difficult for researchers to identify, locate, and use existing data. This paper presents the results of a pilot study that attempts to address this gap by creating a metadata repository (MDR) to act as a central source of information about what data are available at which data holdings across Canada. OBJECTIVE This project aimed to (1) create a proof-of-concept MDR for Canadian health data at the variable level; (2) identify and label language-related variables existing within the MDR data; and (3) develop an interactive, public-facing web application to let users browse and search the MDR. METHODS Metadata were collected from 5 Canadian health data sources, including 4 provincial data holdings and 1 national survey, and pooled to create a data repository. Then, we performed bottom-up labeling of language-related variables within the pooled metadata by first using a search string algorithm across all variable labels, names, and definitions and then consensus screening these variables using a derived, standardized definition of language or linguistic variables. Using the Shiny web framework in R, we then developed an openly accessible web application to allow users to search the proof-of-concept MDR. RESULTS A total of 850,343 variables were collected and included in the repository, with most coming from Ontario (n=712,037, 83.7%) and Manitoba (n=97,051, 11.4%) provincial data holdings. Among all variables in the repository, 213,696 (25.1%) were confirmed to be language related. CONCLUSIONS Developing a national MDR would be a transformative opportunity for Canadian researchers to leverage the full scope of Canadian health administrative data. Although a top-down approach with consistent engagement of and collaboration between provincial data holdings and federal data agencies is ideal to develop a national MDR, this study demonstrates the feasibility of a bottom-up approach in contributing to this overarching goal.
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 | 0.082 | 0.127 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.004 |
| Science and technology studies | 0.007 | 0.003 |
| Scholarly communication | 0.005 | 0.003 |
| Open science | 0.004 | 0.006 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
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".