Essays on Child Mental Health: Estimating Diagnostic Errors, Latent Risk, and Long-term Trajectories
Bibliographic record
Abstract
Understanding diagnostic decision-making for child mental illness is the central focus of my dissertation research. I combine multiple quasi-experimental methods and machine learning tools with novel individual-level tax and survey data linkages to explore these issues. Over three papers, I address several sources of diagnostic and treatment errors in child mental illness, stemming from school and home environments. In the first paper, I study how a child's school starting age generates over-diagnosis of ADHD in young-for-grade and male students and missed diagnoses in relatively older students, especially females. Importantly, I demonstrate that teacher special education training mitigates these age-based assessment errors. The second paper investigates multiple sources of under-and over-utilization of mental health care by combining a machine learning-derived mental health risk prediction with multiple exogenous changes in diagnosis and treatment likelihood. These changes include the effect of school starting age on diagnosis and the role of Quebec's pharmaceutical insurance plan expansion on treatment uptake. I identify that over-diagnosis may be driven by over-weighting externalized behaviours, which are more common in males. At the same time, I additionally find that removing out-of-pocket costs increases treatment rates, specifically for high-risk children. The final paper expands upon this work to understand the long-term trajectories of children with mental health issues and heterogeneous treatment effects at the population level. I find that measured gaps in educational attainment, social assistance receipt and income due to child mental illness persist well into adulthood. I find evidence that diagnosis and treatment worsen human capital outcomes like educational attainment and unemployment for low-risk individuals. Nevertheless, treatment can help close mental health-related socioeconomic inequalities, but only for very high-risk children. Together these papers outline the growing issue of mental illness in childhood and its long-term implications. They highlight the potential for diagnostic mistakes or misallocation of care and the role of new data-driven tools to alleviate some of these medical errors. In my concluding chapter, I highlight some of the strengths and limitations of machine learning tools for improving mental illness detection and treatment going forward and point towards critical work required to make these tools useful for clinical practice.
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 | 0.024 | 0.155 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.002 | 0.004 |
| Scholarly communication | 0.004 | 0.004 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.002 | 0.005 |
| Insufficient payload (model declined to judge) | 0.006 | 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".