{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003016219,0.0001846896,0.0007582941,0.00004994067,0.0001986428,0.000003211693,0.00004353549,0.00008220509,0.0000143211],"category_scores_gemma":[0.004677681,0.0001722391,0.00004742636,0.00009486938,0.00009097086,0.0001283765,0.0001192773,0.0002450109,1.075706e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005719208,"about_ca_system_score_gemma":0.00007109951,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006058427,"about_ca_topic_score_gemma":0.0000442786,"domain_scores_codex":[0.9970203,0.001665062,0.0006120817,0.0003375953,0.00005101856,0.0003139548],"domain_scores_gemma":[0.9847146,0.01453997,0.0003612287,0.0001276005,0.0001633923,0.00009317564],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005863608,0.0005587818,0.04908534,0.008224471,0.0004856757,0.000006033526,0.002114347,0.0004281125,0.1045822,0.6702894,0.00200632,0.1616329],"study_design_scores_gemma":[0.001984973,0.0001745991,0.00218701,0.0004641064,0.0005071327,0.0001011984,0.0005285213,0.6462712,0.1549907,0.1908567,0.001503044,0.0004308983],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02223119,0.004635881,0.9719091,0.0003462547,0.0002073489,0.000506067,0.00005237462,0.00009287153,0.00001890158],"genre_scores_gemma":[0.1869214,0.0002690559,0.8121974,0.0004257355,0.00007301614,0.00004901636,0.00004083055,0.00001589183,0.000007666325],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.645843,"threshold_uncertainty_score":0.7023708,"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."}}