{"id":"W4415495055","doi":"10.1093/aje/kwaf239","title":"Evaluating treatment benefit predictors using observational data: contending with identification and confounding bias","year":2025,"lang":"en","type":"article","venue":"American Journal of Epidemiology","topic":"Advanced Causal Inference Techniques","field":"Mathematics","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Observational study; Confounding; Causal inference; Identification (biology); Intuition; Calibration; Measure (data warehouse); Instrumental variable","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.2116155,0.001209744,0.002542403,0.003260941,0.001169145,0.004270764,0.002861727,0.003873094,0.001306649],"category_scores_gemma":[0.5407571,0.001009325,0.002605972,0.004435949,0.007624074,0.007625611,0.005706603,0.006207493,0.0002633809],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002025377,"about_ca_system_score_gemma":0.003776409,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003331982,"about_ca_topic_score_gemma":0.001696011,"domain_scores_codex":[0.8250595,0.1529848,0.005957786,0.005659013,0.00938863,0.0009503776],"domain_scores_gemma":[0.3312455,0.6197416,0.02112501,0.02296206,0.004285036,0.0006408568],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0005654276,0.0002160502,0.1117442,0.001979337,0.003318792,0.0005141136,0.002182984,0.1153273,0.0008684964,0.5344766,0.002678006,0.2261286],"study_design_scores_gemma":[0.00009506143,0.0002024288,0.01080464,0.0006260979,0.0003832993,0.0002977339,0.0002125661,0.1284112,0.001271571,0.8536463,0.003973984,0.00007503809],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02819072,0.003881714,0.9590212,0.006392449,0.0001292437,0.0002670311,0.0003536491,0.0001172487,0.001646815],"genre_scores_gemma":[0.5601853,0.003689187,0.4300092,0.002626628,0.0007174357,0.001202066,0.0004735313,0.0001255072,0.0009712387],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.7883845,"threshold_uncertainty_score":0.9722185,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.7092128010368095,"score_gpt":0.577015850385838,"score_spread":0.1321969506509715,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}