LinkMD: Linking Medical and Dental Records with 4 Linking Algorithms
Bibliographic record
Abstract
Despite well-established connections between oral and systemic health, electronic health records (EHRs) and electronic dental records (EDRs) remain largely siloed due to infrastructural and interoperability challenges. This separation limits interdisciplinary care and data-driven research to generate practice-based evidence. We developed and validated 4 algorithmic frameworks specifically designed to link EHR with EDR across nonintegrated systems. Using data from more than 1.7 million medical records and 222,480 dental records spanning a 10-y period at Temple University, we evaluated 4 linkage strategies: (1) direct Social Security number matching, (2) unweighted similarity scoring, (3) weighted average similarity scoring, and (4) a probabilistic expectation-conditional maximization classification model. We compared these approaches using expert-reviewed validation of 1,000 candidate record pairs and selected optimal similarity thresholds for high-fidelity linkages. Our weighted average similarity algorithm demonstrated the best performance with 100% specificity (correctly avoiding false matches), 99% sensitivity (correctly identifying all true matches), and 99% accuracy (proportion of all correct linkages out of total comparisons) at the threshold of 0.82 for successfully linking 121,771 unique patients and 144,229 patients' linkage with 96% sensitivity, 78% specificity, and 89% accuracy. After linking the datasets, the completeness of key patient demographic information significantly improved, with missing race data reduced from 79% to 11% and missing ethnicity data from 82% to 17%. We designed the algorithm to be transparent and vendor neutral, making it potentially adaptable to any institution or practice regardless of their existing EHR/EDR systems. This provides a foundation for developing a clinical decision support systems that facilitate real-time health information exchange, supporting safer dental procedures, timely medical referrals, and integrative research. Our findings provide a critical bridge between medicine and dentistry, which have remained largely divorced from each other. Future work will focus on multi-institutional validation, implementation, and integration into routine clinical workflows.
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 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 | 0.006 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.003 |
| 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, 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".