{"id":"W4404528517","doi":"10.23889/ijpds.v9i1.2407","title":"Secondary use of routinely collected administrative health data for epidemiologic research: Answering research questions using data collected for a different purpose","year":2024,"lang":"en","type":"article","venue":"International Journal for Population Data Science","topic":"Medical Coding and Health Information","field":"Health Professions","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia; Simon Fraser University; AIDS Vancouver","funders":"","keywords":"Data science; Computer science","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[{"model":"gemma","categories":["metaresearch"],"domain":"methods","study_design":"theoretical_or_conceptual","genre":"methods","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"},{"model":"gpt","categories":[],"domain":null,"study_design":"design_other","genre":"methods","about_ca_system":false,"about_ca_topic":false,"confidence":"medium","status":"direct model label, unvalidated"}],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.3364265,0.001081192,0.003519734,0.01206262,0.005674601,0.01780628,0.004267856,0.004638214,0.002214344],"category_scores_gemma":[0.6205288,0.001837294,0.00238389,0.02311188,0.0164789,0.01893263,0.009911479,0.007192357,0.001071709],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.007040386,"about_ca_system_score_gemma":0.03067953,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01280404,"about_ca_topic_score_gemma":0.01080716,"domain_scores_codex":[0.4389062,0.4625074,0.04357406,0.01085906,0.04142137,0.002731973],"domain_scores_gemma":[0.2204045,0.5876941,0.04988023,0.08411799,0.05481742,0.003085753],"domain_codex":"methods","domain_gemma":"methods","domain_candidate":"methods","domain_consensus":"methods","study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0003934008,0.0006823076,0.2177114,0.02001981,0.002095957,0.0009989113,0.1165787,0.002219635,0.002027057,0.2427073,0.04267021,0.3518953],"study_design_scores_gemma":[0.0002252695,0.00123355,0.1060365,0.04977865,0.001040689,0.001858453,0.1191164,0.007700007,0.003696194,0.4529151,0.2557346,0.0006645359],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1480013,0.03734382,0.499591,0.2365756,0.008952487,0.008341763,0.01204394,0.0006912271,0.04845891],"genre_scores_gemma":[0.6008949,0.01686804,0.3216722,0.03676516,0.00552114,0.01079215,0.005789718,0.0003801812,0.001316519],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.6635735,"threshold_uncertainty_score":0.8183043,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.9660910023644246,"score_gpt":0.746661212221578,"score_spread":0.2194297901428466,"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."}}