Automatic Estimation of the Most Likely Drug Combination in Electronic Health Records Using the Smooth Algorithm: Development and Validation Study
Bibliographic record
Abstract
BACKGROUND: Since the use of electronic health records (EHRs) in an automated way, pharmacovigilance or pharmacoepidemiology studies have been used to characterize the therapy using different algorithms. Although progress has been made in this area for monotherapy, with combinations of 2 or more drugs the challenge to characterize the treatment increases significantly, and more research is needed. OBJECTIVE: The goal of the research was to develop and describe a novel algorithm that automatically returns the most likely therapy of one drug or combinations of 2 or more drugs over time. METHODS: We used the Information System for Research in Primary Care as our reference EHR platform for the smooth algorithm development. The algorithm was inspired by statistical methods based on moving averages and depends on a parameter Wt, a flexible window that determines the level of smoothing. The effect of Wt was evaluated in a simulation study on the same data set with different window lengths. To understand the algorithm performance in a clinical or pharmacological perspective, we conducted a validation study. We designed 4 pharmacological scenarios and asked 4 independent professionals to compare a traditional method against the smooth algorithm. Data from the simulation and validation studies were then analyzed. RESULTS: The Wt parameter had an impact over the raw data. As we increased the window length, more patient were modified and the number of smoothed patients augmented, although we rarely observed changes of more than 5% of the total data. In the validation study, significant differences were obtained in the performance of the smooth algorithm over the traditional method. These differences were consistent across pharmacological scenarios. CONCLUSIONS: The smooth algorithm is an automated approach that standardizes, simplifies, and improves data processing in drug exposition studies using EHRs. This algorithm can be generalized to almost any pharmacological medication and model the drug exposure to facilitate the detection of treatment switches, discontinuations, and terminations throughout the study period.
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 | 0.006 | 0.019 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 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".