MétaCan
Menu
Back to cohort
Record W4409746284 · doi:10.1002/asmb.70012

Assessing Latent Risk Based on Joint Modelling of Multiple Health Insurance Outcomes of Mixed Types

2025· article· en· W4409746284 on OpenAlex
Xingde Duan, Renjun Ma

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.
fundA Canadian funder is recorded on the work.

Bibliographic record

VenueApplied Stochastic Models in Business and Industry · 2025
Typearticle
Languageen
FieldMathematics
TopicStatistical Methods and Bayesian Inference
Canadian institutionsUniversity of New Brunswick
FundersNatural Sciences and Engineering Research Council of CanadaNational Natural Science Foundation of China
KeywordsActuarial scienceRelevance (law)Health insuranceDiseaseLatent class modelHealth careEconometricsMedicineComputer scienceBusinessEconomicsMachine learning

Abstract

fetched live from OpenAlex

ABSTRACT Our research is motivated by an insurance study involving 788 insurance subscribers who made claims resulting from ischemic heart disease. Four different types of health services used by these subscribers as well as the corresponding total cost were observed for two years. Health care utilizations vary a lot even for subscribers of the same personal characteristics. The research question of primary interest is how to capture patient‐specific latent risks beyond what can be explained by known personal characteristics. In this study, we characterize unobserved latent risks by random effects in our joint Tweedie mixed models for multiple health outcomes of mixed types. An optimal estimation of our model has been developed using the orthodox best linear unbiased predictors of random effects. Our approach is illustrated with the analysis of the health insurance study of 788 ischemic heart disease patients. Applying cluster analysis to the patient‐specific latent risks predicted by our model, we were able to classify patients into a high risk group of 36 patients, a medium risk group of 256 patients and a low risk group of 496 patients. The finding is of important policy relevance since the losses suffered by a few are known to be spread over many in an insurance system. Grouping of patients and prioritization of specific groups based on subject‐specific latent risks facilitates resource allocation and pricing.

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.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.508
Threshold uncertainty score0.611

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0010.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0010.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

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.

Opus teacher head0.168
GPT teacher head0.352
Teacher spread0.185 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it