Lost in Translation: Exposure Misclassification when Relying on Days Supply in Pharmacy Claims Data
Bibliographic record
Abstract
Administrative pharmacy claims data are frequently utilized in pharmacoepidemiology. Days supply values are the most commonly used to estimate drug exposure. This research investigated the potential for exposure misclassification when relying on days supply values to quantify drug adherence and estimate drug effectiveness. With scheduled long-dose intervals, osteoporosis drugs provided a unique case example to examine the potential for misclassified days supply values. Using Ontario administrative claims data, three independent, yet related studies were completed. First, a cross-sectional study of all osteoporosis medications dispensed in Ontario identified potential inaccuracies in days supply values, particularly in long-term care (LTC), where only 59% of days supply values matched pre-defined expected values. In comparison, 90% of community prescriptions matched the expected. Next, two cohort studies were completed to investigate the potential impact of the noted variation in days supply reporting on measures of medication adherence (Study Two) and drug effectiveness (Study Three). To adjust for misclassification, dose-specific cleaning algorithms were developed based on the identification of logical typos and refill patterns, resulting in two values that could be compared; the observed and cleaned days supply. Measures of compliance and persistence were used to identify patient adherence, and were calculated using the observed and cleaned days supply. Results in Study Two identified that data cleaning significantly increased estimates of drug adherence, particularly among LTC residents, where mean compliance increased from 59% to 83% and proportion persisting with therapy increased from 62% to 78%. In the third study, Cox proportional hazard models were used to estimate the relationship between compliance and hip fractures. Results identified important differences in effect estimates following data cleaning, particularly in LTC, where a significant 35% (HRobserved=0.99 to HRcleaned=0.65) change in hazard ratio estimates was observed for the effect of high compliance on fracture risk. Overall, results identified larger differences in LTC settings where exposure was most likely to be misclassified; however, important differences were identified when all patients were combined. Cumulatively, the findings of this thesis have important methodological implications for pharmacoepidemiologic research, and will inform best practices when using days supply values.
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 | 0.140 | 0.447 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.004 | 0.008 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.004 | 0.003 |
| Open science | 0.003 | 0.004 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 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".