Characteristics Associated With Participation in a Behavioral Weight Loss Randomized Control Trial in the U.S. Military.

Abstract

INTRODUCTION\nEffective recruitment and subsequent enrollment of diverse populations is often a challenge in randomized controlled trials, especially those focused on weight loss. In the civilian literature, individuals identified as racial and ethnic minorities, men, and younger and older adults are poorly represented in weight loss interventions. There are limited weight loss trials within military populations, and to our knowledge, none reported participant characteristics associated with enrollment. There may be unique motives and barriers for active duty personnel for enrollment in weight management trials. Given substantial costs and consequences of overweight and obesity in the U.S. military, identifying predictors and limitations to diverse enrollment can inform future interventions within this population. The study aims to describe the recruitment, screening, and enrollment process of a military weight loss intervention. Demographic and lifestyle characteristics of military personnel lost between screening and randomization are compared to characteristics of personnel randomized in the study and characteristics of the Air Force in general.\n\n\nMATERIALS AND METHODS\nThe Fit Blue study, a randomized controlled behavioral weight loss trial for active duty personnel, was approved by the Institutional Review Board of the Wilford Hall Ambulatory Surgical Center in San Antonio, TX, USA and acknowledged by the Institutional Review Board at the University of Tennessee Health Science Center. Logistic regressions compared participant demographics, anthropometric data, and health behaviors between personnel that attended a screening visit but were not randomized and those randomized. Multivariable models were constructed for the likelihood of being randomized using a liberal entry and stay criteria of 0.10 for the p-values in a stepwise variable selection algorithm. Descriptive statistics compared the randomized Fit Blue cohort demographics to those of the U.S. Air Force.\n\n\nRESULTS\nIn univariate analyses, older age (p < 0.02), having a college degree or higher (p < 0.007) and higher military rank (p < 0.02) were associated with completing the randomization process. The randomized cohort reported a lower percentage of total daily kilocalories for fat compared to the non-randomized cohort (p = 0.033). The non-randomized cohort reported more total minutes and intensity of physical activity (p = 0.073). In the multivariate model, only those with a college degree or higher were 3.2 times more likely to go onto randomization. (OR = 3.2, 95% CI = 2.0, 5.6, p < 0.0001). The Fit Blue study included a higher representation of personnel who identified as African American (19.4% versus 15.0%) and Hispanic/Latino (22.7% versus 14.3%) compared with the U.S. Air Force in general; however, men were underrepresented (49.4% versus 80.0%). TABLE I.Comparisons of Demographic Characteristics of Randomized Fit Blue Cohort to Screened Non-Randomized CohortFit Blue Randomized Participants (N = 248)Non-Randomized Cohort (N = 111)All Screened Participants (N = 359)p-ValueSex N (%)0.73\u2003Male122 (49.2)52 (46.8)174 (48.5)\u2003Female126 (50.8)59 (53.2)183 (51.5)Age Mean (±SD) years34 (±7.5)32 (±6.7)33 (±7.3)0.02Race N (%)0.89\u2003African American49 (19.8)22 (19.8)71 (19.8)\u2003Caucasian163 (65.7)75 (67.6)238 (66.3)\u2003Other36 (14.5)14 (12.2)50 (13.9)Ethnicity N (%)0.59\u2003Hispanic/Latino56 (22.6)28 (25.2)84 (23.4)\u2003Non-Hispanic/Latino192 (77.4)83 (74.8)275 (76.6)Education N (%)<0.0001\u2003Less than college degree123 (49.6)82 (73.9)205 (57.1)\u2003College degree or greater125 (50.4)29 (26.1)154 (42.9)Marital status N (%)0.83\u2003Single/never married40 (16.1)20 (18)60 (16.7)\u2003Married/living as married169 (68.1)72 (64.9)241 (67.1)\u2003Separated/divorced39 (15.7)19 (17.1)58 (16.2)Number of additional adults in household N (%)0.82\u2003046 (18.5)22 (19.8)68 (18.9)\u20031162 (65.3)73 (65.8)235 (65.5)\u2003231 (12.5)14 (12.6)45 (12.5)\u20033 or more9 (3.6)2 (1.8)11 (3.1)Number of children in household N (%)0.56\u2003091 (36.7)37 (33.3)128 (35.7)\u2003159 (23.8)23 (20.7)82 (22.8)\u2003257 (23)26 (23.4)83 (23.1)\u20033 or more41 (16.5)25 (22.5)66 (18.4)Years in service mean (± SD)12 (±6.6)11 (±6.1)12 (±6.4)0.20Military gradeaN (%)0.02\u2003E1-E434 (13.7)19 (17.1)53 (14.8)\u2003E5-E6105 (42.3)58 (52.3)163 (45.4)\u2003E7-E952 (21)21 (18.9)73 (20.3)\u2003O1-O317 (6.9)9 (8.1)26 (7.2)\u2003O4-O639 (15.7)4 (3.6)43 (12)Branch0.68\u2003Army4 (1.6)1 (0.9)5 (1.4)\u2003Air Force234 (94.4)105 (94.6)339 (94.4)\u2003Navy8 (3.2)5 (4.5)13 (3.6)\u2003Marine Corp2 (0.8)0 (0.0)2 (0.6)BMI (m2/kg) N (%)30.6 (±2.7)30.4 (±2.9)30.6 (±2.8)BMI category N (%)0.76\u2003Overweight115 (46.4)52 (48.1)167 (46.9)\u2003Obese133 (53.6)56 (51.9)189 (53.1)aMilitary ranking; Enlisted (E) categories: E1-E4 (enlisted), E5-E6 (non-commissioned officers), E7-E9 (senior non-commissioned officers) and two Officer categories (O): O1-O3 (Company Grade Officer) and O4-O6 (Field Grade Officer); standard deviation (SD).Table II.Comparisons of Anthropometric Characteristics of Randomized Fit Blue Cohort to Screened Non-Randomized CohortFit Blue Randomized Participants (N = 248)Non-Randomized Cohort (N = 111)All Screened Participants (N = 359)p-ValuePhysical activity\u2003Total physical activity2525 (±3218)2840 (±2541)2621 (±3028)0.027\u2003(mean (±SD) minutes per week)\u2003Total sedentary physical activity5046 (±239)472 (±221)494 (±234)0.35\u2003(mean (±SD) minutes per week)\u2003Vigorous physical activity34 (±145)54 (±152)40 (±147)0.036\u2003(mean (±SD) minutes per week)Dietary intake\u2003Total sweetened beverages (kcal per day)165 (±206)152.9 (±166)160.8 (±194)0.80\u2003Fruit and vegetable consumption (cups per day)3 (±1)3 (±1)3 (±1)0.52\u2003Dietary fat (% total kcal)35 (±4)34 (±4)35 (±4)0.033.\n\n\nCONCLUSIONS\nAccounting for all influencing characteristics, higher educational status was the only independent predictor of randomization. Perhaps, highly educated personnel are more invested in a military career, and thus, more concerned with consequences of failing required fitness tests. Thus, it may be important for future weight loss interventions to focus recruitment on less-educated personnel. Results suggest that weight loss interventions within a military population offer a unique opportunity to recruit a higher prevalence of males and individuals who identify as racial or ethnic minorities which are populations commonly underrepresented in weight loss research.