The accuracy of using administrative healthcare data to identify epilepsy cases: A systematic review of validation studies
Bibliographic record
Abstract
Our objective was to undertake a systematic review ascertaining the accuracy of using administrative healthcare data to identify epilepsy cases. We searched MEDLINE and Embase from 01/01/1975 to 03/07/2018 for studies evaluating the diagnostic accuracy of routinely collected healthcare data in identifying epilepsy cases. Any disease coding system in use since the International Classification of Diseases, Ninth Revision (ICD-9) was permissible. Two authors independently screened studies, extracted data, and quality-assessed studies. We assessed positive predictive value (PPV), sensitivity, negative predictive value (NPV), and specificity. The primary analysis was a narrative synthesis of review findings. Thirty studies were included, published between 1989 and 2018. Risks of bias were low, high, and unclear in 4, 14, and 12 studies, respectively. Coding systems included ICD-9, ICD-10, and Read Codes, with or without antiepileptic drugs (AEDs). PPVs included ranges of 5.2%-100% (Canada), 32.7%-96.0% (USA), 47.0%-100% (UK), and 37.0%-88.0% (Norway). Sensitivities included ranges of 22.2%-99.7% (Canada), 12.2%-97.3% (USA), and 79.0%-94.0% (UK). Nineteen studies contained at least one algorithm with a PPV >80%. Sixteen studies contained at least one algorithm with a sensitivity >80%. PPV was highest in algorithms consisting of disease codes (ICD-10 G40-41, ICD-9 345) in combination with one or more AEDs. The addition of symptom codes to this (ICD-10 R56; ICD-9 780.3, 780.39) lowered PPV. Sensitivity was highest in algorithms consisting of symptom codes with one or more AEDs. Although using AEDs alone achieved high sensitivities, the associated PPVs were low. Most NPVs and specificities were >90%. We conclude that it is reasonable to use administrative data to identify people with epilepsy (PWE) in epidemiological research. Studies prioritizing high PPVs should focus on combining disease codes with AEDs. Studies prioritizing high sensitivities should focus on combining symptom codes with AEDs. We caution against the use of AEDs alone to identify PWE.
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 | 0.002 | 0.022 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.006 | 0.001 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| 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".