Understanding data provenance when using electronic medical records for research: Lessons learned from the Deliver Primary Healthcare Information (DELPHI) database
Bibliographic record
Abstract
Introduction: We set out to assess the impact of Choosing Wisely Canada recommendations (2014) on reducing unnecessary health investigations and interventions in primary care across Southwestern Ontario. Methods: We used the Deliver Primary Healthcare Information (DELPHI) database, which stores deidentified electronic medical records (EMR) of nearly 65,000 primary care patients across Southwestern Ontario. When conducting research using EMR data, data provenance (i.e., how the data came to be) should first be established. We first considered DELPHI data provenance in relation to longitudinal analyses, flagging a change in EMR software that occurred during 2012 and 2013. We attempted to link records between EMR databases produced by different software using probabilistic linkage and inspected 10 years of data in the DELPHI database (2009 to 2019) for data quality issues, including comparability over time. Results: We encountered several issues resulting from this change in EMR software. These included limited linkage of records between software without a common identifier; data migration issues that distorted procedure dates; and unusual changes in laboratory test and medication prescription volumes. Conclusion: This study reinforces the necessity of assessing data provenance and quality for new research projects. By understanding data provenance, we can anticipate related data quality issues such as changes in EMR data over time-which represent a growing concern as longitudinal data analyses increase in feasibility and popularity.
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 | 0.644 | 0.662 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.005 | 0.006 |
| Science and technology studies | 0.009 | 0.014 |
| Scholarly communication | 0.015 | 0.017 |
| Open science | 0.006 | 0.021 |
| Research integrity | 0.004 | 0.007 |
| 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; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
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".