Clinical Decision Support for Suicide Risk Assessment: Exploring the Opportunity of Predictive Analytics and the Need to Build Different Approaches
Bibliographic record
Abstract
Suicide is a large public health concern accounting for 4000 deaths per year in Canada. Previous research has focused on identifying patient risk factors and developing risk assessment tools to better recognize high risk patients, but there has been limited supporting evidence for the use of these tools. Through five interrelated studies, this dissertation is focused on how electronic health records (EHRs) and Computerized Decision Support Systems can aid healthcare professionals in suicide risk assessment (SRA). First, the feasibility of incorporating EHR-based algorithms for SRA is explored. A systematic mapping review on EHR data used in suicide prediction found promising predictive performance of such algorithms, however a majority remained in the realm of research and were not validated or implemented within clinical care. Next, a retrospective validation of such a predictive algorithm (developed externally) was performed within a mental health hospital. Low sensitivity ceased implementation and clinical adoption, with data availability and quality acting as the main barriers. Following that, other targets of clinical decision-support were explored, focusing on understanding healthcare professional-level behaviour instead of relying on patient-specific algorithms. Personal and contextual factors that affect a healthcare professional’s ability to assess suicide risk were catalogued, finding a range of system-level (e.g. an ill-equipped system), organizational-level (e.g. documentation requirements, team dynamics) and individual-level (e.g. ability to connect with a patient, cultural issues) factors through means of a scoping review. Then, through a concurrent mixed methods study (79 healthcare professionals surveyed and 21 interviewed), the decision-making processes, current barriers and facilitators were explored. It was identified that a health professional’s clinical designation and mental health experience were significantly associated with their decision and confidence in SRA. Most SRAs were conducted conversationally, without the use of a structured tool. Barriers within this process included gathering appropriate collateral or past history of a patient and effectively detecting changes from baseline for patients who are chronically suicidal, among many other barriers identified. Based on this needs assessment, customized clinical decision support systems are recommended– one that moves beyond patient algorithms. This work also has implications for SRA training and education.
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 | 0.081 | 0.216 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.003 |
| Bibliometrics | 0.009 | 0.006 |
| Science and technology studies | 0.002 | 0.004 |
| Scholarly communication | 0.018 | 0.015 |
| Open science | 0.003 | 0.008 |
| Research integrity | 0.003 | 0.007 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
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".