Abstract 015: Importance of Balanced Follow-up Time and Other Study Design Considerations When Evaluating Adherence Using Two Novel Oral Anticoagulants
Bibliographic record
Abstract
Background: Medication adherence rates decline over time, especially after the first dispensing. Comparing adherence rates for medications that have been on the market for differing period of time may distort real differences in medication adherence. Other analysis factors such as minimum number of dispensing criteria and Pharmacy Quality Alliance (PQA) adherence measures can also affect adherence measurement. Objectives: To use one real world example (rivaroxaban vs apixaban) in non-valvular atrial fibrillation (NVAF) patients to quantify the impact of adjusting for imbalances in follow-up periods, minimum number of dispensing, and use of the PQA adherence measure. Methods: Using IMS Health Real-World Data Adjudicated Claims and Truven MarketScan claims databases, we included adult patients with ≥1 rivaroxaban or apixaban dispensing (index date), ≥1 year of pre-index eligibility, ≥1 AF diagnosis pre-index, newly initiated on oral anticoagulant therapy, and no valvular involvement. Adherence was evaluated using proportion of days covered (PDC) ≥0.8 for cohorts with (1) unbalanced follow-up (2) balanced follow-up (by matching on month and year of follow-up since fill-date) (3) ≥2 rivaroxaban or apixaban dispensings and a balanced follow-up, and using (4) the PQA adherence measure. Results: Rivaroxaban users had significantly longer mean (SD) follow-up than apixaban (408 [300] versus 254 [196] days, respectively). While apixaban users appeared to be more adherent in unadjusted analyses, this finding was reversed after 1) adjusting for unbalanced follow-up 2) excluding single-time users; and 3) applying the PQA-endorsed adherence measure (Figure). Similar results were found using the Truven databases. Conclusion: Comparisons of the adherence rates among medications need to account for the period of time each have been on the market, number of dispensing and PQA measures. Retrospective analyses of adherence that do not adjust for such differences could produce spurious findings.
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 | 0.003 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| 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".