{"id":"W3124915543","doi":"10.1101/2021.01.21.21250243","title":"Problems with Evidence Assessment in COVID-19 Health Policy Impact Evaluation: A systematic review of study design and evidence strength","year":2021,"lang":"en","type":"review","venue":"medRxiv","topic":"Health Policy Implementation Science","field":"Health Professions","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"National Institute on Drug Abuse; Knut och Alice Wallenbergs Stiftelse; National Institutes of Health; Laura and John Arnold Foundation","keywords":"Impact assessment; Coronavirus disease 2019 (COVID-19); Health impact assessment; Impact factor; Systematic review; Actuarial science; Sample (material); Medicine; Set (abstract data type); MEDLINE; Psychology; Computer science; Political science; Business; Public health; Nursing; Disease","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","metaepi_narrow"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.09040727,0.0006938013,0.005965208,0.0009629593,0.0005441642,0.000052207,0.0008685697,0.0001841833,0.0003596038],"category_scores_gemma":[0.05905394,0.000448842,0.0001962358,0.004764479,0.0001409025,0.0005822401,0.0003201113,0.001254864,0.00002369366],"about_ca_system_candidate":true,"about_ca_system_consensus":true,"about_ca_system_score_codex":0.006340455,"about_ca_system_score_gemma":0.1270142,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004305741,"about_ca_topic_score_gemma":0.001509319,"domain_scores_codex":[0.9163256,0.07125857,0.006786099,0.001321657,0.003036293,0.001271793],"domain_scores_gemma":[0.959969,0.02939607,0.006628409,0.00180256,0.001054321,0.001149668],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"systematic_review","study_design_gemma":"systematic_review","study_design_scores_codex":[0.000004059436,0.00009154167,0.00279092,0.9853793,0.00008532272,0.000009474706,0.004650297,0.0000164103,3.204335e-8,0.00003511237,0.0003609799,0.006576586],"study_design_scores_gemma":[0.0005929975,0.0009135072,0.0005606249,0.9914672,0.0008411177,0.00003417825,0.001507741,0.00008467142,7.839938e-9,0.000009381535,0.003670869,0.0003177462],"study_design_candidate":"systematic_review","study_design_consensus":"systematic_review","genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.00005815303,0.9383615,0.001161189,0.006739576,0.00008462946,0.05350753,0.00003753558,0.00003874784,0.00001119824],"genre_scores_gemma":[0.0006286683,0.9763891,0.001080148,0.005951227,0.00007426552,0.01576833,0.00001355986,0.00005745368,0.00003722315],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.1206738,"threshold_uncertainty_score":0.9997963,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.8374160443260061,"score_gpt":0.7610109532821004,"score_spread":0.07640509104390569,"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."}}