Identifying Patterns of Clinical Interest in Clinicians’ Treatment Preferences: Hypothesis-free Data Science Approach to Prioritizing Prescribing Outliers for Clinical Review
Bibliographic record
Abstract
BACKGROUND: Data analysis is used to identify signals suggestive of variation in treatment choice or clinical outcome. Analyses to date have generally focused on a hypothesis-driven approach. OBJECTIVE: This study aimed to develop a hypothesis-free approach to identify unusual prescribing behavior in primary care data. We aimed to apply this methodology to a national data set in a cross-sectional study to identify chemicals with significant variation in use across Clinical Commissioning Groups (CCGs) for further clinical review, thereby demonstrating proof of concept for prioritization approaches. METHODS: Here we report a new data-driven approach to identify unusual prescribing behaviour in primary care data. This approach first applies a set of filtering steps to identify chemicals with prescribing rate distributions likely to contain outliers, then applies two ranking approaches to identify the most extreme outliers amongst those candidates. This methodology has been applied to three months of national prescribing data (June-August 2017). RESULTS: Our methodology provides rankings for all chemicals by administrative region. We provide illustrative results for 2 antipsychotic drugs of particular clinical interest: promazine hydrochloride and pericyazine, which rank highly by outlier metrics. Specifically, our method identifies that, while promazine hydrochloride and pericyazine are barely used by most clinicians (with national prescribing rates of 11.1 and 6.2 per 1000 antipsychotic prescriptions, respectively), they make up a substantial proportion of antipsychotic prescribing in 2 small geographic regions in England during the study period (with maximum regional prescribing rates of 298.7 and 241.1 per 1000 antipsychotic prescriptions, respectively). CONCLUSIONS: Our hypothesis-free approach is able to identify candidates for audit and review in clinical practice. To illustrate this, we provide 2 examples of 2 very unusual antipsychotics used disproportionately in 2 small geographic areas of England.
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 | 0.222 | 0.462 |
| Meta-epidemiology (narrow) | 0.002 | 0.002 |
| Meta-epidemiology (broad) | 0.003 | 0.007 |
| Bibliometrics | 0.017 | 0.012 |
| Science and technology studies | 0.002 | 0.004 |
| Scholarly communication | 0.007 | 0.004 |
| Open science | 0.005 | 0.004 |
| Research integrity | 0.004 | 0.005 |
| Insufficient payload (model declined to judge) | 0.003 | 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".