Immortal time bias in observational studies of drug effects
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
PURPOSE: Recent observational studies suggest that various drugs are remarkably effective at reducing morbidity and mortality. These cohort studies used a flawed approach to design and data analysis which can lead to immortal time bias. We describe the bias from 20 of these studies and illustrate it by showing that unrelated drugs can be made to appear effective at treating cardiovascular disease (CVD). METHODS: The illustration used a cohort of 3315 patients, with chronic obstructive pulmonary disease (COPD), identified from the Saskatchewan Health databases, hospitalised for CVD and followed for up to a year. We used the biased approach to assess the effect of two medications, namely gastrointestinal drugs (GID) and inhaled beta-agonists (IBA), both unknown to be effective in CVD, on the risk of all-cause mortality. We also estimated these effects using the proper person-time approach. RESULTS: Using the inappropriate approach, the rates ratios of all-cause death were 0.73 (95%CI: 0.57-0.93), with IBA and 0.78 (95%CI: 0.61-0.99), with GID. These rate ratios became 0.98 (95%CI: 0.77-1.25) and 0.94 (95%CI: 0.73-1.20), respectively, with the proper person-time analysis. CONCLUSIONS: Several recent observational studies used a flawed approach to design and data analysis, leading to immortal time bias, which can generate an illusion of treatment effectiveness. Observational studies, with surprising beneficial drug effects should be re-assessed to account for this source of bias.
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.
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 | 0.008 | 0.007 |
| 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 it