Data variability across Canadian administrative health databases: Differences in content, coding, and completeness
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
PURPOSE: The Canadian Network for Observational Drug Effect Studies (CNODES) is a network of Canadian research centres using administrative data to conduct distributed drug safety and effectiveness studies. In this study, we compare the provincial administrative databases and illustrate the potential impact of database differences on a CNODES study about domperidone and the risk of ventricular tachyarrhythmia and sudden cardiac death (VT/SCD). METHODS: We assessed the impact of varying versions and precision of the International Classification of Diseases coding system in physician claims data, and the content and completeness of hospital discharge abstracts across CNODES sites, as these variations can introduce differences in the study cohorts formed and affect study results. RESULTS: In our study of 214 962 patients, hospital diagnosis type (such as most responsible, admitting, or secondary diagnosis) was missing in some provinces, resulting in misclassification of the outcome and variation in rates and risk estimates. Incidence rates of VT/SCD ranged from 19.8 (95% confidence interval [CI] 17.7-22.2) per 10 000 person-years in British Columbia to 53.4 (95% CI 50.3-56.5) in Quebec. While most provinces reported an increased risk of VT/SCD, a null effect was observed in Quebec (rate ratio 1.06; 95% CI 0.79-1.41). CONCLUSIONS: Distributed analyses allow for rapid responses to drug safety signals. However, variation in characteristics of the administrative data across research centres can influence study results. By identifying the sources of database heterogeneity, one can evaluate the potential biases these differences may introduce, highlighting the importance of considering such variation in distributed networks.
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.
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 | 0.020 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| 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 it