{"id":"W4308636435","doi":"10.1002/pds.5566","title":"High‐dimensional propensity scores for empirical covariate selection in secondary database studies: Planning, implementation, and reporting","year":2022,"lang":"en","type":"article","venue":"Pharmacoepidemiology and Drug Safety","topic":"Advanced Causal Inference Techniques","field":"Mathematics","cited_by":49,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; McGill University Health Centre","funders":"National Institute of Diabetes and Digestive and Kidney Diseases; International Society for Pharmacoepidemiology","keywords":"Covariate; Medicine; Propensity score matching; Pharmacoepidemiology; Selection (genetic algorithm); Statistics; Econometrics; Internal medicine; Pharmacology; Computer science; Machine learning","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.008450756,0.0001652278,0.000557493,0.0000992773,0.0005096261,0.000004205971,0.00006055466,0.00003653158,0.00009667542],"category_scores_gemma":[0.003667156,0.0001491819,0.00003115089,0.0001168997,0.0001129278,0.0001845844,0.000323883,0.0004387573,1.996651e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001118468,"about_ca_system_score_gemma":0.00009718626,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001193063,"about_ca_topic_score_gemma":0.00009593996,"domain_scores_codex":[0.9971504,0.0008799467,0.00115625,0.0004395128,0.00008833926,0.0002855347],"domain_scores_gemma":[0.9948435,0.004051924,0.0008730636,0.00008743873,0.00008459593,0.00005942184],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.002746897,0.0003120643,0.7490204,0.0008148542,0.0004489981,0.0001094186,0.005317327,0.001221112,0.008588118,0.06720819,0.1486271,0.01558555],"study_design_scores_gemma":[0.005988381,0.0005358721,0.1001063,0.0001379822,0.0002851719,0.0005444043,0.003160872,0.007876035,0.008749208,0.8621148,0.009544991,0.0009559088],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9665351,0.0008224193,0.02839371,0.002653648,0.0001687934,0.001071634,0.0001721581,0.0001555675,0.00002699697],"genre_scores_gemma":[0.9245417,0.0001634737,0.07213134,0.002367856,0.00007806925,0.0004783325,0.0001708688,0.00002042224,0.00004798306],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7949067,"threshold_uncertainty_score":0.6083461,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.44729527161018,"score_gpt":0.5604829685782081,"score_spread":0.1131876969680281,"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."}}