Prediction Model for Attrition from a Combat Unit Training Program
Yael Fuks, Amir Hadid, Yael Arbel, Ran Yanovich, Daniel Sender Moran, Arie Laor · Medicine & Science in Sports & Exercise · 2011
PURPOSE: To develop a prediction model for the attrition rate of soldiers from a 12-month elite military training program based on physical, dietary, and psychological factors. METHODS: Two groups of 59 and 61 healthy, fit young males (18.7±0.7 yrs) entering a rigorous combat unit training program in the Israeli Defense Forces (IDF) volunteered to participate in this study. Data on anthropometrics, nutrition intake, psychological profile, blood measures (chemical and hematological), physical fitness, and bone quality (peripheral quantitative computed tomography- pQCT) were collected on induction day. Psychological questionnaires were completed again after four months of basic training (BT). RESULTS: Data from the two groups were pooled together for the analysis and to construct a new prediction model for attrition (Patt). Four explanatory variables were found significant with attrition by stepwise logistic regression analysis and were modeled as follows: Patt=26.75-0.98EstCom4-1.72Sc-0.34Fat-0.02Stibia38%; where EstCom4 is commander appreciation as perceived by the soldier after 4-months BT, Sc is a self-confidence grade, Fat is the body fat percentage, and Stibia38% is the total bone area of the tibia at 38% distal cross section. The new suggested model, based on three variables collected on induction and one variable collected after BT successfully predicted 79.2% of subject attrition in the combat unit during 5 to 12 months of the training program. CONCLUSIONS: Recruits with smaller tibias, relatively low body fat percentage, low self-esteem, and who feel unappreciated by their commander are at a higher dropout risk from a rigorous combat training program. This study was supported in part by a contract from the MRMC No. W911QY-08-P-0286.