{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,9,24]],"date-time":"2025-09-24T20:40:25Z","timestamp":1758746425561,"version":"3.44.0"},"reference-count":22,"publisher":"Oxford University Press (OUP)","issue":"9-10","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,9,1]]},"abstract":"<jats:title>ABSTRACT<\/jats:title>\n               <jats:sec>\n                  <jats:title>Introduction<\/jats:title>\n                  <jats:p>Approximately 1 in 4 Navy recruits will not complete their obligated service. Identifying factors associated with Navy attrition is essential to retaining a skilled and motivated fighting force. This study utilized machine learning and variable importance measures (VIM) extracted from random forest (RF) and extreme gradient boosting (XGBoost) decision trees to identify factors that contribute to Navy attrition. These methods are more robust and sensitive when assessing hundreds of predictor variables, offering superior performance relative to regression analyses. Leveraging advanced analytics is crucial to better understand these factors to maximize military retention.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Materials and Methods<\/jats:title>\n                  <jats:p>The study population included Navy personnel with an accession date in 2016. The outcome of interest was a discharge date before the end of obligated service (i.e., attrition). The analysis incorporated 542 independent variables, including demographic factors, medical outpatient visits, and outpatient pharmacy medications. RF, XGBoost, and logistic regression models were trained with different subsets of features to determine which set of variables best predicted attrition. Model performance metrics were generated describing the sensitivity, specificity, positive and negative predictive values, area under the curve, and classification accuracy for the validation data.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>Overall, there were 39,866 Navy personnel with accessions in 2016 and, of these, 28.15% (n\u2009=\u200911,177) did not complete their obligated service. The RF model provided the best accuracy for predicting attrition (81.7%) and area under the curve (90.0%). Both the RF and XGBoost models outperformed the logistic regression model. VIMs were then assessed, including mean decrease accuracy, mean decrease Gini impurity, gain, and cover. The resulting VIM identified five groups described as: mental health, occupations, demographics\/sex-related issues, pain management, and medical compliance. The top 3 VIM and their overall relative risk on attrition were adjustment disorders (RR\u2009=\u20091.39 higher), Seaman Specialists (RR\u2009=\u20093.01 higher), and Electronic Equipment Repairers (RR\u2009=\u20090.44 lower).<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Conclusions<\/jats:title>\n                  <jats:p>Five groups of variables were found to be predictive of Navy attrition: Mental health (MH) disorder, alcohol-related problems, occupations, sex, medical appointment compliance, and pain management. These results demonstrate the utility of machine learning models in predicting attrition compared with regression analyses. VIM is a valuable tool that could be used in decision-making processes in the context of military personnel management and retention. Furthermore, ensemble approaches, compared with a single decision tree, improve overall predictive performance and result in a more robust model that is resistant to overfitting.<\/jats:p>\n               <\/jats:sec>","DOI":"10.1093\/milmed\/usaf101","type":"journal-article","created":{"date-parts":[[2025,3,14]],"date-time":"2025-03-14T00:19:57Z","timestamp":1741911597000},"page":"e2025-e2031","source":"Crossref","is-referenced-by-count":0,"title":["Identifying Risk and Protective Factors for Attrition Among Recently Enlisted Navy Personnel Using Variable Importance Measures"],"prefix":"10.1093","volume":"190","author":[{"given":"James M","family":"Zouris","sequence":"first","affiliation":[{"name":"Epidemiology and Data Management Support Department, Naval Health Research Center , San Diego, CA 92106,","place":["USA"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Andrew J","family":"MacGregor","sequence":"additional","affiliation":[{"name":"Epidemiology and Data Management Support Department, Naval Health Research Center , San Diego, CA 92106,","place":["USA"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Nathan C","family":"Carnes","sequence":"additional","affiliation":[{"name":"Military Population Health Department, Naval Health Research Center , San Diego, CA 92106,","place":["USA"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2025,4,1]]},"reference":[{"volume-title":"Predicting 36-month Attrition in the US Military","year":"2020","author":"Marrone","key":"2025092416174785200_R1"},{"issue":"9","key":"2025092416174785200_R2","doi-asserted-by":"crossref","first-page":"760","DOI":"10.1093\/milmed\/167.9.760","article-title":"Predictors of Navy Attrition. I. Analysis of 1-year attrition","volume":"167","author":"Booth-Kewley","year":"2002","journal-title":"Mil Med"},{"volume-title":"MILITARY PERSONNEL First-Term Recruiting and Attrition Continue to Require Focused Attention.General Accounting Office","year":"2000","author":"Rabkin","key":"2025092416174785200_R3"},{"issue":"9","key":"2025092416174785200_R4","doi-asserted-by":"crossref","first-page":"770","DOI":"10.1093\/milmed\/167.9.770","article-title":"Predictors of Navy attrition. II. A demonstration of potential usefulness for screening","volume":"167","author":"Larson","year":"2002","journal-title":"Mil Med"},{"issue":"1-2","key":"2025092416174785200_R5","doi-asserted-by":"crossref","first-page":"119","DOI":"10.1093\/milmed\/usaa278","article-title":"Toward a \u201cDashboard\u201d indicator of retention in U.S. Navy Personnel","volume":"186","author":"Taylor","year":"2021","journal-title":"Mil Med"},{"issue":"1-2","key":"2025092416174785200_R6","doi-asserted-by":"publisher","first-page":"e1603","DOI":"10.7205\/MILMED-D-15-00451","article-title":"Physical performance and attrition among U.S. Air Force trainees participating in the basic military training fueling initiative","volume":"182","author":"Bartlett","year":"2017","journal-title":"Mil Med"},{"issue":"4","key":"2025092416174785200_R7","doi-asserted-by":"publisher","first-page":"382","DOI":"10.7205\/MILMED-D-02-5308","article-title":"Predictors of one-year attrition in female marine corps recruits","volume":"174","author":"Pollack","year":"2009","journal-title":"Mil Med"},{"issue":"4","key":"2025092416174785200_R8","doi-asserted-by":"publisher","first-page":"436","DOI":"10.7205\/MILMED-D-14-00311","article-title":"Mental health diagnoses and attrition in Air Force recruits","volume":"180","author":"Garcia","year":"2015","journal-title":"Mil Med"},{"issue":"1","key":"2025092416174785200_R9","doi-asserted-by":"publisher","first-page":"63","DOI":"10.7205\/MILMED.172.1.63","article-title":"Attrition of U.S. Military enlistees with waivers for hearing deficiency, 1995\u20132004","volume":"172","author":"Niebuhr","year":"2007","journal-title":"Mil Med"},{"key":"2025092416174785200_R10","article-title":"Prevalence of past year substance use and mental illness by veteran status in a nationally representative sample","volume-title":"CBHSQ","author":"Pemberton","year":"2016"},{"issue":"2","key":"2025092416174785200_R11","doi-asserted-by":"publisher","first-page":"135","DOI":"10.1136\/military-2022-002151","article-title":"Medical correlates of first-term attrition in US Navy personnel","volume":"170","author":"Elkins","year":"2024","journal-title":"BMJ Mil Health"},{"issue":"6","key":"2025092416174785200_R12","doi-asserted-by":"publisher","first-page":"465","DOI":"10.7205\/MILMED.169.6.465","article-title":"Childhood exposure to family violence and attrition in the Navy","volume":"169","author":"Merrill","year":"2004","journal-title":"Mil Med"},{"issue":"12","key":"2025092416174785200_R13","doi-asserted-by":"publisher","first-page":"1037","DOI":"10.7205\/MILMED.170.12.1037","article-title":"Gender and Trauma as predictors of military attrition: a study of Marine corps recruits","volume":"170","author":"Wolfe","year":"2005","journal-title":"Mil Med"},{"issue":"11-12","key":"2025092416174785200_R14","doi-asserted-by":"publisher","first-page":"e3532","DOI":"10.1093\/milmed\/usad146","article-title":"Increasing maternity leave and decreasing attrition rates of U.S. Active Duty women in the prenatal and immediate postpartum period","volume":"188","author":"Herrick","year":"2023","journal-title":"Mil Med"},{"issue":"8","key":"2025092416174785200_R15","doi-asserted-by":"publisher","first-page":"897","DOI":"10.7205\/MILMED-D-13-00567","article-title":"Attrition due to orthopedic reasons during combat training: rates, types of injuries, and comparison between infantry and noninfantry units","volume":"179","author":"Schwartz","year":"2014","journal-title":"Mil Med"},{"issue":"2","key":"2025092416174785200_R16","first-page":"111","article-title":"Evidence based emergency medicine; Part 5 receiver operating curve and area under the curve","volume":"4","author":"Safari","year":"2016","journal-title":"Emerg"},{"issue":"1","key":"2025092416174785200_R17","doi-asserted-by":"publisher","DOI":"10.1186\/s12859-018-2264-5","article-title":"Random forest versus logistic regression: a large-scale benchmark experiment","volume":"19","author":"Couronn\u00e9","year":"2018","journal-title":"BMC Bioinf."},{"issue":"2","key":"2025092416174785200_R18","first-page":"172","article-title":"Career history archival medical and personnel system","volume":"170","author":"Gunderson","year":"2005","journal-title":"Mil Med"},{"author":"Health.mil","key":"2025092416174785200_R19","article-title":"MDR, M2, ICDs Functional References and Specifications"},{"issue":"25","key":"2025092416174785200_R20","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/1471-2105-8-25","article-title":"Bias in random forest variable importance measures: Illustrations, sources and a solution","volume":"8","author":"Strobl","year":"2007","journal-title":"BMC Bioinf."},{"key":"2025092416174785200_R21","doi-asserted-by":"crossref","first-page":"1049","DOI":"10.15585\/mmwr.mm6932a1","article-title":"Mental health, substance use, and suicidal ideation during the COVID-19 pandemic \u2014 United States, June 24\u201330, 2020","volume":"69","author":"Czeisler","year":"2020","journal-title":"MMWR Morb Mortal Wkly Rep"},{"issue":"7-8","key":"2025092416174785200_R22","doi-asserted-by":"publisher","first-page":"e2223","DOI":"10.1093\/milmed\/usac328","article-title":"Comparison of mental health visits at a military treatment facility emergency department pre- and post-COVID-19 pandemic","volume":"188","author":"Wojahn","year":"2023","journal-title":"Mil Med"}],"container-title":["Military Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/milmed\/article-pdf\/190\/9-10\/e2025\/62844324\/usaf101.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/milmed\/article-pdf\/190\/9-10\/e2025\/62844324\/usaf101.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,24]],"date-time":"2025-09-24T20:17:53Z","timestamp":1758745073000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/milmed\/article\/190\/9-10\/e2025\/8104087"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4,1]]},"references-count":22,"journal-issue":{"issue":"9-10","published-online":{"date-parts":[[2025,4,1]]},"published-print":{"date-parts":[[2025,9,1]]}},"URL":"https:\/\/doi.org\/10.1093\/milmed\/usaf101","relation":{},"ISSN":["0026-4075","1930-613X"],"issn-type":[{"type":"print","value":"0026-4075"},{"type":"electronic","value":"1930-613X"}],"subject":[],"published-other":{"date-parts":[[2025,9]]},"published":{"date-parts":[[2025,4,1]]}}}