Comparison of Vital Statistics Definitions of Suicide against a Coroner Reference Standard: A Population-Based Linkage Study
Bibliographic record
Abstract
OBJECTIVE: We sought to determine the utility of health administrative databases for population-based suicide surveillance, as these data are generally more accessible and more integrated with other data sources compared to coroners' records. METHOD: In this retrospective validation study, we identified all coroner-confirmed suicides between 2003 and 2012 in Ontario residents aged 21 and over and linked this information to Statistics Canada's vital statistics data set. We examined the overlap between the underlying cause of death field and secondary causes of death using ICD-9 and ICD-10 codes for deliberate self-harm (i.e., suicide) and examined the sociodemographic and clinical characteristics of misclassified records. RESULTS: Among 10,153 linked deaths, there was a very high degree of overlap between records coded as deliberate self-harm in the vital statistics data set and coroner-confirmed suicides using both ICD-9 and ICD-10 definitions (96.88% and 96.84% sensitivity, respectively). This alignment steadily increased throughout the study period (from 95.9% to 98.8%). Other vital statistics diagnoses in primary fields included uncategorised signs and symptoms. Vital statistics records that were misclassified did not differ from valid records in terms of sociodemographic characteristics but were more likely to have had an unspecified place of injury on the death certificate ( P < 0.001), more likely to have died at a health care facility ( P < 0.001), to have had an autopsy ( P = 0.002), and to have been admitted to a psychiatric hospital in the year preceding death ( P = 0.03). CONCLUSIONS: A high degree of concordance between vital statistics and coroner classification of suicide deaths suggests that health administrative data can reliably be used to identify suicide deaths.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.035 | 0.103 |
| Meta-epidemiology (narrow) | 0.000 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.005 | 0.007 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.001 | 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), 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".