{"id":"W2972679677","doi":"10.1002/pds.4889","title":"Data variability across Canadian administrative health databases: Differences in content, coding, and completeness","year":2019,"lang":"en","type":"article","venue":"Pharmacoepidemiology and Drug Safety","topic":"Medical Coding and Health Information","field":"Health Professions","cited_by":45,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University; McGill University; Institute for Clinical Evaluative Sciences; University of Calgary; University of Manitoba; Jewish General Hospital; Manitoba Health; University of Toronto","funders":"Canadian Institutes of Health Research","keywords":"Medicine; Pharmacoepidemiology; Confidence interval; Observational study; Database; Coding (social sciences); Demography; Statistics; Internal medicine","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"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":[0.1267293,0.0008427367,0.001739802,0.0105207,0.003569578,0.005670833,0.006355925,0.0008900075,0.001262802],"category_scores_gemma":[0.3417695,0.001006249,0.001845208,0.02766198,0.002746764,0.00149544,0.003668837,0.0012699,0.0001948083],"about_ca_system_candidate":true,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.02503761,"about_ca_system_score_gemma":0.04102952,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.8523905,"about_ca_topic_score_gemma":0.8208104,"domain_scores_codex":[0.7307053,0.1080055,0.03594425,0.01885604,0.101167,0.005321923],"domain_scores_gemma":[0.5940795,0.209446,0.06101795,0.0508958,0.08178638,0.002774388],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0003729253,0.00007587319,0.9171261,0.001787785,0.003767323,0.0001770604,0.004954374,0.003808003,0.000495933,0.006384614,0.01124757,0.04980254],"study_design_scores_gemma":[0.0001048987,0.00005610189,0.9703547,0.001218417,0.0008041105,0.0002436923,0.002042671,0.004969253,0.0007435055,0.003012356,0.01627703,0.000173125],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7121421,0.01979619,0.1047706,0.01943908,0.0009707707,0.003488302,0.1175982,0.0006384443,0.02115632],"genre_scores_gemma":[0.9398667,0.00239208,0.02564947,0.002484988,0.0001593851,0.001244316,0.02744829,0.0001494369,0.0006053007],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9749624,"threshold_uncertainty_score":0.6702168,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.6711014377746589,"score_gpt":0.5662363753812069,"score_spread":0.1048650623934521,"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."}}