Estimating Disease Prevalence in Administrative Data
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: Disease prevalence estimates from population-based administrative databases are often biased due to measurement (misclassification) errors. The purpose of this article is to review the methodology for estimating disease prevalence in administrative data, with a focus on bias correction. SOURCE: Several approaches to bias correction in administrative data were reviewed and application of these methods was demonstrated using an example from the literature: physician claims and hospitalization data were employed to estimate diabetes prevalence in Ontario, Canada. FINDINGS: Misclassification bias in prevalence estimates from administrative data can be reduced by developing and selecting an optimal algorithm for case identification, applying a bias correction formula, or using statistical modelling. An algorithm for which sensitivity equals positive predictive value provides an unbiased estimate of prevalence. Bias reduction methods generally require information about the measurement properties of the algorithm, such as sensitivity, specificity, or predictive value. These properties depend on disease type, prevalence, algorithm definition (including the observation window), and may vary by population and time. Prevalence estimates can be improved by applying multivariable disease prediction models. CONCLUSION: Frequency of a positive case identification algorithm in administrative data is generally not equivalent to disease prevalence. Although prevalence estimates can be corrected for bias using known measurement properties of the algorithm, these properties may be difficult to estimate accurately; therefore, disease prevalence estimates based on administrative data must be treated with caution.
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 | 0.004 | 0.114 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.003 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.008 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| 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 it