A computational method for analyzing interval-censored time-to-event data in the presence of informative examination times
Bibliographic record
Abstract
In statistical analysis of time-to-event data, there are situations where the event of interest is only known to have occurred within an interval of time. Such data are considered interval-censored. In studying the human immunodeficiency (HIV) epidemic, interval-censored time-to-event data arise naturally, since the time an individual becomes HIV positive is not exactly known. In addition, individuals may undergo multiple tests before an event (HIV infection) is observed. When these tests are requested instead of pre-scheduled, their visit pattern may reflect individual risk behavior for which covariate information is not available. Therefore, the test visit times are said to be informative for the risk of HIV infection. Statistical methods that model the dependence between the event of interest and examination times are required for the analysis of such data. Farrington and Gay (1999, Statistics in Medicine 18: 1235-1248) analyzed interval-censored data with informative examination times using random effects which capture the correlation between individual's risk and visit rate. However, their proposed methodology is based on an approximation of the marginal log-likelihood and relies on large number of visits per individual. We developed a method for analyzing interval-censored time-to-event data with informative examination times, which addresses the limitations of the Farrington and Gay's method. Our method avoids the likelihood function approximation embedded in their method, thus leading to less biased parameter estimates and more accurate standard errors. Expectation-Maximization (EM) algorithm is used for parameter estimation. We employed multiple imputation of event times via Sampling/Importance Resampling method (Rubin, 1987, JASA 82). This reduced the likelihood of interval-censored time-to-event data to that of right-censored data. We evaluated the bias of parameter estimates of the proposed method and the accuracy of their standard by simulations. Under most simulation scenarios the new approach resulted in reduced bias compared to the Farrington and Gay's method. We applied the proposed method to the analysis of the Polaris HIV repeat testers data in Ontario.
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 | 0.010 | 0.019 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| 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 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".