{"id":"W4410964242","doi":"10.1186/s12911-025-03028-1","title":"Using self-generated identification codes to match anonymous longitudinal data in a sexual health study of secondary school students: a cohort study","year":2025,"lang":"en","type":"article","venue":"BMC Medical Informatics and Decision Making","topic":"Survey Methodology and Nonresponse","field":"Social Sciences","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia, Okanagan Campus","funders":"Health and Medical Research Fund; Food and Health Bureau","keywords":"Health informatics; Identification (biology); Longitudinal study; Cohort; Longitudinal data; Cohort study; Computer science; Medical education; Medicine; Psychology; Public health; Family medicine; Data mining; Nursing","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":["metaresearch"],"category_scores_codex":[0.06692087,0.000128411,0.0004801702,0.0004749046,0.0004395761,0.0001683816,0.0007866606,0.0001274433,0.00006395998],"category_scores_gemma":[0.01328359,0.000115172,0.00001387567,0.001090128,0.0000889548,0.0003390253,0.0007728128,0.0002728785,0.000003991148],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001195806,"about_ca_system_score_gemma":0.001606128,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001976366,"about_ca_topic_score_gemma":0.03219969,"domain_scores_codex":[0.9908342,0.005169527,0.00167844,0.0003498255,0.001640495,0.0003275497],"domain_scores_gemma":[0.9915822,0.007178901,0.0003195609,0.0005424863,0.0001710971,0.0002057307],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0007993514,0.0009083819,0.9547051,0.00005474019,0.00006590426,0.000008427655,0.01799999,0.000128399,0.000002121733,0.00002760735,0.000209668,0.02509032],"study_design_scores_gemma":[0.001337051,0.0002113732,0.8544365,0.0001929501,0.00003479098,0.000002834942,0.126932,0.01661132,7.922665e-7,0.0001055829,0.00002805475,0.0001068039],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8765544,0.00009229216,0.1217451,0.00001757267,0.0002641825,0.001240495,0.00001178243,0.00002538637,0.00004871441],"genre_scores_gemma":[0.9560442,0.00002853362,0.04359048,0.0002577707,0.00002845273,0.00002190564,0.000008483738,0.00000494666,0.00001521556],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.108932,"threshold_uncertainty_score":0.995028,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3231162113409023,"score_gpt":0.5521945060174723,"score_spread":0.22907829467657,"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."}}