Alison S Rustagi, Marzieh Vali, Francis J Graham, Emily N Lum, Christopher G Slatore, Salomeh Keyhani
{"title":"A Novel Automated Algorithm to Identify Lung Cancer Screening from Free Text of Radiology Orders.","authors":"Alison S Rustagi, Marzieh Vali, Francis J Graham, Emily N Lum, Christopher G Slatore, Salomeh Keyhani","doi":"10.1007/s11606-025-09429-2","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Lung cancer screening (LCS) is recommended for asymptomatic patients. Administrative codes for LCS may capture tests prompted by signs/symptoms.</p><p><strong>Objective: </strong>To validate an automated algorithm that identifies LCS among asymptomatic patients.</p><p><strong>Design: </strong>In this cross-sectional study, an algorithm was iteratively developed to identify outpatient low-dose chest CT scans via Current Procedural Terminology (CPT) codes, search free text of radiology orders for screening terms and signs/symptoms (e.g., cough), and classify scans as screening or not.</p><p><strong>Participants: </strong>National population-based sample of 4503 adults ages 65-80 in Veterans Health Affairs primary care, with detailed smoking history to identify LCS-eligible individuals (30 + pack-years, current tobacco use, or quit < 15 years prior).</p><p><strong>Main measures: </strong>Algorithm specificity, sensitivity, positive predictive value (PPV), and negative predictive value (NPV) relative to manual chart review (gold standard) on 100% of screening scans and > 10% random sample of non-screening scans.</p><p><strong>Key results: </strong>Chart review was conducted on n = 335 scans. The final algorithm could not classify 22% of scans, of which 73% were non-screening; these were excluded from primary analyses. Among 842 LCS-eligible individuals, the algorithm demonstrated 97% sensitivity (95%CI 91-99%) and 79% specificity (58-93%). Only 69% (61-77%) of scans classified as LCS via administrative codes were truly screening, compared to 95% of those classified as screening via the algorithm (p < 0.001). Algorithm performance was similar regardless of LCS eligibility, with 90% PPV (84-94%) and 93% NPV (86-97%) in the overall population regardless of tobacco cigarette history.</p><p><strong>Conclusions: </strong>An automated algorithm can accurately identify screening versus diagnostic chest imaging, a necessary step to unbiased analyses of LCS in non-randomized settings. Studies should assess the accuracy of administrative codes for LCS in other health systems.</p>","PeriodicalId":15860,"journal":{"name":"Journal of General Internal Medicine","volume":" ","pages":""},"PeriodicalIF":4.3000,"publicationDate":"2025-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of General Internal Medicine","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1007/s11606-025-09429-2","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"HEALTH CARE SCIENCES & SERVICES","Score":null,"Total":0}
引用次数: 0
Abstract
Background: Lung cancer screening (LCS) is recommended for asymptomatic patients. Administrative codes for LCS may capture tests prompted by signs/symptoms.
Objective: To validate an automated algorithm that identifies LCS among asymptomatic patients.
Design: In this cross-sectional study, an algorithm was iteratively developed to identify outpatient low-dose chest CT scans via Current Procedural Terminology (CPT) codes, search free text of radiology orders for screening terms and signs/symptoms (e.g., cough), and classify scans as screening or not.
Participants: National population-based sample of 4503 adults ages 65-80 in Veterans Health Affairs primary care, with detailed smoking history to identify LCS-eligible individuals (30 + pack-years, current tobacco use, or quit < 15 years prior).
Main measures: Algorithm specificity, sensitivity, positive predictive value (PPV), and negative predictive value (NPV) relative to manual chart review (gold standard) on 100% of screening scans and > 10% random sample of non-screening scans.
Key results: Chart review was conducted on n = 335 scans. The final algorithm could not classify 22% of scans, of which 73% were non-screening; these were excluded from primary analyses. Among 842 LCS-eligible individuals, the algorithm demonstrated 97% sensitivity (95%CI 91-99%) and 79% specificity (58-93%). Only 69% (61-77%) of scans classified as LCS via administrative codes were truly screening, compared to 95% of those classified as screening via the algorithm (p < 0.001). Algorithm performance was similar regardless of LCS eligibility, with 90% PPV (84-94%) and 93% NPV (86-97%) in the overall population regardless of tobacco cigarette history.
Conclusions: An automated algorithm can accurately identify screening versus diagnostic chest imaging, a necessary step to unbiased analyses of LCS in non-randomized settings. Studies should assess the accuracy of administrative codes for LCS in other health systems.
期刊介绍：
The Journal of General Internal Medicine is the official journal of the Society of General Internal Medicine. It promotes improved patient care, research, and education in primary care, general internal medicine, and hospital medicine. Its articles focus on topics such as clinical medicine, epidemiology, prevention, health care delivery, curriculum development, and numerous other non-traditional themes, in addition to classic clinical research on problems in internal medicine.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
