{"id":"W3210133931","doi":"10.1097/ede.0000000000001432","title":"On the Nature of Informative Presence Bias in Analyses of Electronic Health Records","year":2021,"lang":"en","type":"article","venue":"Epidemiology","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":44,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"National Institute of Environmental Health Sciences","keywords":"Health records; Confounding; Electronic health record; Scope (computer science); Non-response bias; Cohort; Information bias; Medical record","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.004287656,0.00009139824,0.0004868664,0.000123762,0.00004159001,0.000002411786,0.0006410318,0.0001263846,0.00002724363],"category_scores_gemma":[0.02024631,0.00006024481,0.00007045081,0.0007629059,0.0000805555,0.00008004461,0.0001893294,0.001035476,0.000003948263],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007222804,"about_ca_system_score_gemma":0.0005940381,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001286916,"about_ca_topic_score_gemma":0.000599077,"domain_scores_codex":[0.9947664,0.003774898,0.0007171005,0.0002385335,0.0001205209,0.0003825907],"domain_scores_gemma":[0.9861816,0.01240498,0.0006071154,0.0006401652,0.0001248136,0.00004128256],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"observational","study_design_scores_codex":[0.00001680321,0.00003724288,0.08932006,0.0001159441,0.000025418,0.000002470856,0.00185681,0.01171666,0.00002600833,0.884215,0.002797448,0.009870113],"study_design_scores_gemma":[0.0003867254,0.001466804,0.461901,0.0004021456,0.000004404566,0.00003962727,0.0003782879,0.2844855,0.002105626,0.2445433,0.00404762,0.0002390161],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7411366,0.009541574,0.05699478,0.1880843,0.0006503305,0.0004923578,0.00001015692,0.00006670775,0.003023308],"genre_scores_gemma":[0.9881424,0.000260641,0.005297167,0.006226773,0.00001739887,0.000009341823,0.000004013209,0.000003082181,0.00003918209],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6396717,"threshold_uncertainty_score":0.9880066,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1350699049802592,"score_gpt":0.4530477399609975,"score_spread":0.3179778349807383,"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."}}