86Segmenting persistently high-cost individuals into actionable groups
Bibliographic record
Abstract
Abstract Background Despite growing evidence describing high cost patients, decision-makers struggle to implement effective strategies to improve care and curb spending in this population. Using a multi-phased approach, we aimed to classify high cost patients into homogeneous subgroups amenable to targeted interventions. Methods We linked population-level administrative health data in Alberta, Canada from 2012-2017. We defined “persistently high-cost” as those in the top 1% of cumulative inpatient, outpatient and medication cost in at least two consecutive years. We used latent class analysis to separate this persistent high-cost population into potentially actionable subgroups. Results Of the 3,795,067 adults residing in Alberta, 21,361 were ‘persistently high-cost’. Latent class models identified 10 high-cost subgroups: individuals with CKD (19.3% of persistent high-cost individuals), those undergoing joint surgery/replacement and rehabilitation (18.6%), individuals with IBD (11.6%), patients receiving biologics for autoimmune conditions (11.3%), patients receiving high cost drugs for other conditions (11.1%), community-dwelling individuals with multimorbid chronic conditions (9.0%), individuals with schizophrenia (6.8%), individuals with other mental health issues (6.2%), rural individuals with COPD (3.4%), and frail elderly in institutional settings (2.7%). Conclusions Latent class analysis was able to identify 10 persistently high-cost groups based on meaningful differences in health care spending, demographics, and clinical diagnoses. Key messages This taxonomy will inform the identification of interventions shown to improve care and reduce cost for each subgroup in addition to consultation with key stakeholders to identify and reflect on key barriers and facilitators to implementing identified interventions within the local context.
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 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 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".