{"id":"W4200057099","doi":"10.1002/pds.5403","title":"A comparison of confounder selection and adjustment methods for estimating causal effects using large healthcare databases","year":2021,"lang":"en","type":"article","venue":"Pharmacoepidemiology and Drug Safety","topic":"Advanced Causal Inference Techniques","field":"Mathematics","cited_by":17,"is_retracted":false,"has_abstract":true,"ca_institutions":"Jewish General Hospital; McMaster University; Université Laval; Université de Montréal; Programs for Assessment of Technology in Health Research Institute; McMaster University Medical Centre; Impact; McGill University; Centre Hospitalier de l’Université de Montréal; The Quebec Population Health Research Network","funders":"Fonds de Recherche du Québec - Santé; Natural Sciences and Engineering Research Council of Canada; Canadian Institutes of Health Research; Université Laval","keywords":"Confounding; Propensity score matching; Lasso (programming language); Observational study; Medicine; Statistics; Bayesian probability; Selection bias; Collinearity; Machine learning; Computer science; Econometrics; Data mining; Artificial intelligence; Mathematics","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":[],"category_scores_codex":[0.07365565,0.001095925,0.001589858,0.001719951,0.0005613621,0.001315077,0.002082424,0.001565293,0.001709341],"category_scores_gemma":[0.1935411,0.0006840497,0.002522889,0.00215053,0.001063295,0.002024122,0.00202004,0.001631937,0.0001500832],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001536058,"about_ca_system_score_gemma":0.003713918,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007516032,"about_ca_topic_score_gemma":0.004059152,"domain_scores_codex":[0.9533505,0.04085349,0.001187123,0.00195338,0.002372013,0.0002834736],"domain_scores_gemma":[0.7898147,0.1906926,0.006990847,0.008070346,0.003589376,0.0008421049],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.005908085,0.0005716845,0.06402383,0.001874781,0.006552602,0.0002747277,0.0006642892,0.5136467,0.000892813,0.04699074,0.003991914,0.3546079],"study_design_scores_gemma":[0.002012184,0.0008101691,0.01525741,0.0004821788,0.0009240562,0.0002168706,0.0001694319,0.9343908,0.0008527433,0.04158991,0.003196256,0.00009798588],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1244256,0.00596317,0.862541,0.002577075,0.0001514801,0.001438516,0.0007217401,0.000500634,0.00168075],"genre_scores_gemma":[0.6214262,0.002169159,0.3725004,0.0005534566,0.00009785825,0.001952125,0.000835486,0.00008223722,0.0003830148],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9263443,"threshold_uncertainty_score":0.389533,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3429585724946843,"score_gpt":0.6076724006262337,"score_spread":0.2647138281315495,"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."}}