{"id":"https://openalex.org/W7225313555","doi":"https://doi.org/10.3352/jeehp.2026.23.36","title":"Item-writing flaws and answer-position bias in large language model-generated medical multiple-choice questions: a benchmark study of 15 models with an open corpus and a rule-based detector","display_name":"Item-writing flaws and answer-position bias in large language model-generated medical multiple-choice questions: a benchmark study of 15 models with an open corpus and a rule-based detector","publication_year":2026,"publication_date":"2026-11-05","ids":{"openalex":"https://openalex.org/W7225313555","doi":"https://doi.org/10.3352/jeehp.2026.23.36","pmid":"https://pubmed.ncbi.nlm.nih.gov/42984146"},"language":"en","primary_location":{"id":"doi:10.3352/jeehp.2026.23.36","doi":"https://doi.org/10.3352/jeehp.2026.23.36","is_oa":true,"landing_page_url":"https://doi.org/10.3352/jeehp.2026.23.36","pdf_url":"https://www.jeehp.org/upload/pdf/jeehp-23-36.pdf","source":{"id":"https://openalex.org/S30317932","display_name":"Journal of Educational Evaluation for Health Professions","issn_l":"1738-1339","issn":["1738-1339","1975-5937"],"is_oa":true,"is_in_doaj":true,"is_core":true,"listed_in":["cwts-core","doaj","doyens","jufo-1","kci-registered","ki-jl-1","medline","poland-20","russia-white-list-2"],"host_organization":"https://openalex.org/P4310319416","host_organization_name":"Korea Health Insurance Licensing Examination Institute","host_organization_lineage":["https://openalex.org/P4310319416"],"host_organization_lineage_names":["Korea Health Insurance Licensing Examination Institute"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of Educational Evaluation for Health Professions","raw_type":"journal-article"},"volume":null,"type":"article","is_traditional":true,"indexed_in":["crossref","doaj","pubmed"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://www.jeehp.org/upload/pdf/jeehp-23-36.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5152512349","display_name":"Rohit Gundam","orcid":"https://orcid.org/0009-0001-3018-9663","observed_orcids":["https://orcid.org/0009-0001-3018-9663"]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Rohit Gundam","raw_affiliation_strings":["Independent researcher, Cleveland, OH, USA. rgundam@icloud.com"],"raw_orcid":"https://orcid.org/0009-0001-3018-9663","affiliations":[{"raw_affiliation_string":"Independent researcher, Cleveland, OH, USA. rgundam@icloud.com","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5152512349"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.0,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"23","issue":null,"first_page":"36","last_page":"36"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.8511000275611877,"subfield":{"id":"https://openalex.org/subfields/2718","display_name":"Health Informatics"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},"topics":[{"id":"https://openalex.org/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.8511000275611877,"subfield":{"id":"https://openalex.org/subfields/2718","display_name":"Health Informatics"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T10254","display_name":"Innovations in Medical Education","score":0.07760000228881836,"subfield":{"id":"https://openalex.org/subfields/3304","display_name":"Education"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T14414","display_name":"Artificial Intelligence in Education","score":0.014100000262260437,"subfield":{"id":"https://openalex.org/subfields/3304","display_name":"Education"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.8259999752044678},{"id":"https://openalex.org/keywords/item-writing-flaws","display_name":"item-writing flaws","score":0.7160000205039978},{"id":"https://openalex.org/keywords/automated-item-generation","display_name":"automated item generation","score":0.6470000147819519},{"id":"https://openalex.org/keywords/prompt-engineering","display_name":"prompt engineering","score":0.5849999785423279},{"id":"https://openalex.org/keywords/item-writing-guidelines","display_name":"item writing guidelines","score":0.7490000128746033},{"id":"https://openalex.org/keywords/llm-as-a-judge","display_name":"LLM-as-a-judge","score":0.49000000953674316},{"id":"https://openalex.org/keywords/content-validity","display_name":"content validity","score":0.41999998688697815}],"concepts":[{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.713100016117096},{"id":"https://openalex.org/C9719361","wikidata":"https://www.wikidata.org/wiki/Q7928967","display_name":"Vignette","level":2,"score":0.6467999815940857},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6144999861717224},{"id":"https://openalex.org/C94915269","wikidata":"https://www.wikidata.org/wiki/Q1834857","display_name":"Detector","level":2,"score":0.5076000094413757},{"id":"https://openalex.org/C143095724","wikidata":"https://www.wikidata.org/wiki/Q515895","display_name":"Odds","level":3,"score":0.48660001158714294},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4846999943256378},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.46219998598098755},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.4050999879837036},{"id":"https://openalex.org/C19527891","wikidata":"https://www.wikidata.org/wiki/Q1120908","display_name":"Medical physics","level":1,"score":0.39890000224113464},{"id":"https://openalex.org/C156957248","wikidata":"https://www.wikidata.org/wiki/Q1862216","display_name":"Odds ratio","level":2,"score":0.3930000066757202},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.38260000944137573},{"id":"https://openalex.org/C198082294","wikidata":"https://www.wikidata.org/wiki/Q3399648","display_name":"Position (finance)","level":2,"score":0.38029998540878296},{"id":"https://openalex.org/C44249647","wikidata":"https://www.wikidata.org/wiki/Q208498","display_name":"Confidence interval","level":2,"score":0.3732999861240387},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.36230000853538513},{"id":"https://openalex.org/C2779473830","wikidata":"https://www.wikidata.org/wiki/Q1540899","display_name":"MEDLINE","level":2,"score":0.35409998893737793},{"id":"https://openalex.org/C71924100","wikidata":"https://www.wikidata.org/wiki/Q11190","display_name":"Medicine","level":0,"score":0.3433000147342682},{"id":"https://openalex.org/C3018949938","wikidata":"https://www.wikidata.org/wiki/Q17166101","display_name":"Text messaging","level":2,"score":0.3411000072956085},{"id":"https://openalex.org/C509550671","wikidata":"https://www.wikidata.org/wiki/Q126945","display_name":"Medical education","level":1,"score":0.33820000290870667},{"id":"https://openalex.org/C22156102","wikidata":"https://www.wikidata.org/wiki/Q5341294","display_name":"Educational measurement","level":3,"score":0.32659998536109924},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.3138999938964844},{"id":"https://openalex.org/C31601959","wikidata":"https://www.wikidata.org/wiki/Q931309","display_name":"Medical imaging","level":2,"score":0.29280000925064087},{"id":"https://openalex.org/C2777548347","wikidata":"https://www.wikidata.org/wiki/Q5456937","display_name":"Flagging","level":2,"score":0.2921999990940094},{"id":"https://openalex.org/C64869954","wikidata":"https://www.wikidata.org/wiki/Q1859747","display_name":"False positive paradox","level":2,"score":0.2727000117301941},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.26429998874664307},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.25859999656677246}],"mesh":[{"descriptor_ui":"D000098342","descriptor_name":"Large Language Models","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D004521","descriptor_name":"Educational Measurement","qualifier_ui":"Q000379","qualifier_name":"methods","is_major_topic":true},{"descriptor_ui":"D004521","descriptor_name":"Educational Measurement","qualifier_ui":"Q000592","qualifier_name":"standards","is_major_topic":true},{"descriptor_ui":"D006801","descriptor_name":"Humans","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D014956","descriptor_name":"Writing","qualifier_ui":"Q000592","qualifier_name":"standards","is_major_topic":true},{"descriptor_ui":"D015982","descriptor_name":"Bias","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D019985","descriptor_name":"Benchmarking","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true}],"locations_count":2,"locations":[{"id":"doi:10.3352/jeehp.2026.23.36","doi":"https://doi.org/10.3352/jeehp.2026.23.36","is_oa":true,"landing_page_url":"https://doi.org/10.3352/jeehp.2026.23.36","pdf_url":"https://www.jeehp.org/upload/pdf/jeehp-23-36.pdf","source":{"id":"https://openalex.org/S30317932","display_name":"Journal of Educational Evaluation for Health Professions","issn_l":"1738-1339","issn":["1738-1339","1975-5937"],"is_oa":true,"is_in_doaj":true,"is_core":true,"listed_in":["cwts-core","doaj","doyens","jufo-1","kci-registered","ki-jl-1","medline","poland-20","russia-white-list-2"],"host_organization":"https://openalex.org/P4310319416","host_organization_name":"Korea Health Insurance Licensing Examination Institute","host_organization_lineage":["https://openalex.org/P4310319416"],"host_organization_lineage_names":["Korea Health Insurance Licensing Examination Institute"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of Educational Evaluation for Health Professions","raw_type":"journal-article"},{"id":"pmid:42984146","doi":null,"is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/42984146","pdf_url":null,"source":{"id":"https://openalex.org/S4306525036","display_name":"PubMed","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of educational evaluation for health professions","raw_type":"Journal Article"}],"best_oa_location":{"id":"doi:10.3352/jeehp.2026.23.36","doi":"https://doi.org/10.3352/jeehp.2026.23.36","is_oa":true,"landing_page_url":"https://doi.org/10.3352/jeehp.2026.23.36","pdf_url":"https://www.jeehp.org/upload/pdf/jeehp-23-36.pdf","source":{"id":"https://openalex.org/S30317932","display_name":"Journal of Educational Evaluation for Health Professions","issn_l":"1738-1339","issn":["1738-1339","1975-5937"],"is_oa":true,"is_in_doaj":true,"is_core":true,"listed_in":["cwts-core","doaj","doyens","jufo-1","kci-registered","ki-jl-1","medline","poland-20","russia-white-list-2"],"host_organization":"https://openalex.org/P4310319416","host_organization_name":"Korea Health Insurance Licensing Examination Institute","host_organization_lineage":["https://openalex.org/P4310319416"],"host_organization_lineage_names":["Korea Health Insurance Licensing Examination Institute"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of Educational Evaluation for Health Professions","raw_type":"journal-article"},"sustainable_development_goals":[{"score":0.4043000042438507,"id":"https://metadata.un.org/sdg/3","display_name":"Good health and well-being"}],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7225313555.pdf","grobid_xml":"https://content.openalex.org/works/W7225313555.grobid-xml"},"referenced_works_count":14,"referenced_works":["https://openalex.org/W2891855722","https://openalex.org/W4391254266","https://openalex.org/W4400210547","https://openalex.org/W4410859334","https://openalex.org/W4411303115","https://openalex.org/W4416040306","https://openalex.org/W4417441013","https://openalex.org/W7117988202","https://openalex.org/W7160325330","https://openalex.org/W7161774373","https://openalex.org/W7162399887","https://openalex.org/W7166582255","https://openalex.org/W7166798950","https://openalex.org/W7167805024"],"related_works":["https://openalex.org/W7117988202","https://openalex.org/W4393318738","https://openalex.org/W4416693324","https://openalex.org/W7202163425","https://openalex.org/W7128498154","https://openalex.org/W7168598869","https://openalex.org/W4412945732","https://openalex.org/W7203742123","https://openalex.org/W7167805024","https://openalex.org/W4415583550"],"abstract_inverted_index":{"PURPOSE:":[0],"Item-writing":[1],"flaws":[2,235,253],"in":[3,14,72,185,207],"large":[4],"language":[5],"model":[6],"(LLM)-generated":[7],"medical":[8],"multiple-choice":[9],"questions":[10],"have":[11],"been":[12],"assessed":[13,112],"small":[15],"samples.":[16],"This":[17],"study":[18],"developed":[19],"an":[20],"open":[21],"benchmark":[22],"of":[23,39,87,128,138,159,187,209,218],"such":[24],"items":[25,54,71,114,130,210,232],"labeled":[26,79],"for":[27],"structural":[28,133],"flaws,":[29,125],"compared":[30,135],"15":[31],"models":[32,178],"and":[33,42,62,82,100,115,240],"2":[34,101],"prompts,":[35],"traced":[36],"the":[37,80,143,147,157,174,180,190,195,221,247,252],"source":[38],"answer-position":[40],"bias,":[41],"evaluated":[43],"LLM":[44,110,214],"judges":[45,111],"against":[46,95],"physicians.":[47],"METHODS:":[48],"Fifteen":[49],"LLMs":[50],"generated":[51],"12,000":[52],"vignette":[53],"across":[55],"8":[56],"specialties":[57],"using":[58],"a":[59,63,132,160,204,227,241],"plain":[60,175],"prompt":[61,64],"with":[65,136,220],"10":[66],"item-writing":[67],"guidelines,":[68],"plus":[69],"15,800":[70],"5":[73],"experimental":[74],"arms.":[75],"A":[76,184],"14-rule":[77],"detector":[78,121],"corpus":[81],"667":[83],"retired":[84],"National":[85],"Board":[86],"Medical":[88],"Examiners":[89],"(NBME)":[90],"items;":[91,108,188],"it":[92],"was":[93],"validated":[94],"3,737":[96],"external":[97],"human":[98],"labels":[99],"independent":[102],"readers.":[103],"Two":[104],"physicians":[105,222],"rated":[106],"100":[107],"4":[109],"those":[113],"216":[116],"human-labeled":[117],"items.":[118,141],"RESULTS:":[119],"The":[120,212],"found":[122,203],"no":[123],"format":[124,234],"but":[126,236],"23.4%":[127],"plain-prompt":[129],"had":[131],"flaw,":[134],"17.3%":[137],"NBME":[139],"5-option":[140],"Excluding":[142],"longest-option":[144],"rule":[145],"reduced":[146,194,251],"gap":[148],"from":[149,198,246],"6.1":[150],"to":[151,171,200],"1.2":[152],"percentage":[153],"points.":[154],"Guidelines":[155,250],"halved":[156],"odds":[158],"flaw":[161],"(odds":[162],"ratio,":[163],"0.50;":[164],"95%":[165],"confidence":[166],"interval,":[167],"0.45-0.56),":[168],"reducing":[169],"prevalence":[170],"13.6%.":[172],"Under":[173],"prompt,":[176],"7":[177],"placed":[179],"key":[181],"at":[182],"position":[183],"61%-94%":[186],"removing":[189],"template's":[191],"example":[192],"letter":[193],"pooled":[196],"proportion":[197],"54%":[199],"23%.":[201],"Physicians":[202],"content":[205],"defect":[206],"12%-13%":[208],"(\u03ba=0.59).":[211],"best":[213],"judge":[215],"achieved":[216],"agreement":[217],"\u03ba=0.53-0.56":[219],"on":[223],"differently":[224],"defined":[225],"composites,":[226],"preliminary":[228],"result.":[229],"CONCLUSION:":[230],"LLM-generated":[231],"avoided":[233],"contained":[237],"answer":[238],"cues":[239],"position-A":[242],"bias":[243],"arising":[244],"mostly":[245],"output":[248],"template.":[249],"they":[254],"specified":[255],"without":[256],"demonstrably":[257],"improving":[258],"content,":[259],"which":[260],"needs":[261],"physician":[262],"review.":[263]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-11-06T00:00:00"}