Clinical and Biological Stratification in 121,560 Antidepressant Prescription Trajectories using Unsupervised Modelling and Clustering
Bibliographic record
Abstract
Abstract Major depressive disorder is a complex condition with diverse presentations and polygenic underpinnings. Leveraging large biobanks linked to primary care prescription data, we developed a data-driven approach based on antidepressant prescription trajectories for patient stratification and novel phenotype identification. We extracted quantitative prescription trajectories for 56,951 UK Biobank (UKB) and 64,609 Danish National Biobank (CHB+DBDS) individuals. Using Hidden Markov Models and K-means clustering, we identified five and six patient clusters, respectively. Multinomial logistic regression and non-parametric association tests, using clinical information, enabled patient group characterization. We consistently identified three common patient groups across cohorts: first, a majority group of individuals with mild to moderate depression; second, those with severe mental illness (i.e., a group with a higher likelihood of psychiatric diagnoses, such as bipolar depression, with odds ratios: OR UKB = 1.87 [95% CI = 1.48, 2.35], p = 2.7e-6; OR CHB+DBDS = 1.69 [95% CI = 1.41, 2.02], p = 2.3e-7); and third, patients with less severe forms of depression or receiving treatment for conditions other than depression (i.e., a group with a lower likelihood of depression diagnosis: OR UKB = 0.80 [95% CI = 0.74, 0.85], p = 3e-10; OR CHB+DBDS = 0.77 [95% CI = 0.73, 0.82], p < 1e-10). Genome-wide association studies (GWAS) revealed 14 significant loci, including USP4 and BCHE on chromosome 3, as well as a locus associated with the drug metabolising enzyme CYP2D6 . These findings, and the reproducibility across cohorts, demonstrate the power of unsupervised phenotyping from primary care prescriptions for patient stratification and pharmacogenetics research.
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 | 0.002 | 0.009 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 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".