Trends in Overweight and Obesity in Soldiers Entering the US Army,

Transcription

1 Trends in Overweight and in Soldiers Entering the US Army, Adela Hruby 1,2, Owen T. Hill 3,4, Lakmini Bulathsinhala 3, Craig J. McKinnon 3, Scott J. Montain 1, Andrew J. Young 1, and Tracey J. Smith 1 Objective: The US Army recruits new soldiers from an increasingly obese civilian population. The change in weight status at entry into the Army between 1989 and 2012 and the demographic characteristics associated with overweight/obesity at entry were examined. Methods: 1,741,070 unique individuals with complete sex, age, and anthropometric information contributed data to linear and logistic regressions examining time trends and associations between demographic characteristics and overweight/obesity. Results: The prevalence of overweight (body mass index 25-<30 kg/m 2 ) generally increased, from 25.8% (1989) to 37.2% (2012), peaking at 37.9% (2011). The prevalence of obesity (body mass index 30 kg/m 2 ) also increased from 5.6% (1989) to 8.0% (2012), peaking at 12.3% (2009); annual prevalence exceeded 10%. The most consistent demographic characteristics predicting overweight/obesity were male sex, older age, Hispanic or Asian/Pacific Island race/ethnicity, and being married. There were no distinct geographic trends. Conclusions: The US Army is not immune to the US obesity epidemic. Demographic characteristics associated with being overweight or obese should be considered when developing military-sponsored weight management programs for new soldiers. (2015) 23, doi: /oby Introduction Between 1988 and 2010, the prevalence of obesity in US adults rose from an estimated 20.5% to 35.9% and 25.9% to 36.1% in men and women, respectively (1). Meanwhile, obesity in American adolescents nearly doubled: in , some 10.5% of 12- to 19-yearolds were obese; by , 20.5% were obese (2,3). Since members of the US armed forces are recruited from the general US population, increases in the prevalence of overweight/obesity in children and adults are a matter of concern to the Department of Defense (DoD), and overweight/obesity in military personnel has been identified as a national security threat (4-7). Overweight/obesity in military personnel increase risk of premature discharge (8), decreased length of service (9), higher incidence of heat illness (10), and musculoskeletal injury and related healthcare utilization (11). Obese active-duty personnel report higher rates of absenteeism and more present days with below-normal productivity (i.e., presenteeism), with associated annual costs of US$103 million and US$2.6 million, respectively (12). An estimated 20% of applicants to the Army are disqualified due to overweight/obesity or other medical conditions (13). Men and women entering (known as accessing into) the Army must meet age- and sex-specific weight-for-height screening criteria defined in Army Regulation (AR) (Standards of Medical Fitness) (14). When an individual exceeds the maximum allowable weight-for-height criteria, body fat standards, in use since 1991, are the final determinant in evaluating an applicant s acceptability (14). These screening criteria and body fat standards have changed over time. In addition, waivers to the criteria and standards, including those based on fitness tests, are used to permit accession. While the use of waivers has been noted to increase the number of overweight and over-fat entrants (15), they have not been found to impact attrition (15-17). 1 Military Nutrition Division, US Army Research Institute of Environmental Medicine, Natick, Massachusetts, USA. Correspondence: Tracey J. Smith (tracey.smith10.civ@mail.mil) 2 Department of Nutrition, Harvard School of Public Health, Boston, Massachusetts, USA 3 Military Performance Division, US Army Research Institute of Environmental Medicine, Natick, Massachusetts, USA 4 Center for the Intrepid, Fort Sam, Houston, Texas, USA. Funding agencies: Dr. Hruby s contribution to this research was supported by an appointment to the Postgraduate Research Participation Program at the US Army Medical Research Institute of Environmental Medicine administered by the Oak Ridge Institute for Science and Education an interagency agreement between the US Department of Energy and USAMRMC. Disclosure: The authors declared no conflicts of interest. The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or reflecting the views of the Army or the Department of Defense. Any citations of commercial organizations and trade names in this report do not constitute an official Department of the Army endorsement of approval of the products or services of these organizations. Additional Supporting Information may be found in the online version of this article. Received: 28 July 2014; Accepted: 4 November 2014; Published online 22 January doi: /oby VOLUME 23 NUMBER 3 MARCH

2 The age- and sex-specific weight-for-height criteria used by the Army are different from body mass index (BMI) thresholds used to define overweight (25-<30 kg/m 2 ) and obesity (>30 kg/m 2 ) in the civilian population (18). Weight-for-height screening criteria were developed to align with a range of healthy BMI and body fat (19,20), and allow BMIs in the overweight range to avoid misclassification due to excess muscle mass, primarily in males. For example, a 20-year-old man who entered the Army in 2012 at 72 inches tall could maximally weigh 200 pounds (14). At this weight, his BMI is considered overweight, at 27.1 kg/m 2. Notably, calculated maximum BMIs have historically been higher in males than females (up to 31.9 vs kg/m 2, respectively, at their highest), presumably because excess body weight is more likely to represent over-fatness in women. Several studies have examined short-term time trends in, and prevalence of, overweight and obesity in both newly accessing (21) and active-duty military personnel (22-24). In a study of 18-year-olds entering the Army between 1993 and 2006, the prevalence of overweight/obesity (i.e., BMI 25 kg/m 2 ) increased from 25.6% to 33.9% (21). Among active-duty military, approximately 50% and 54% of personnel reported being overweight/obese in 1995 and 1998, respectively (22). The prevalence of self-reported overweight/ obesity reached 57.2% in 2002 and 60.5% in 2005 (23), with obesity accounting for 12.9% of the latter. In 2011, self-reported obesity prevalence remained high, at 12.4% in the armed services overall, and was highest in the Army, at 15.8% (24). Several studies have also examined risk factors for overweight/obesity in the US military, implicating male gender, African-American race, Hispanic ethnicity, older age, and marital status (22,23,25). Because the DoD recruits its members from all regions of the US, the geographical distribution of obesity is also of interest. In 2012, the prevalence of self-reported obesity in adults exceeded 30% in 13 states, predominantly in the south and in the Midwest. Whether these geographic trends in obesity, which have been consistently observed since the 1990s (26), are reflected in military applicants is unknown. There have been no long-term (>15 years) studies of these trends in individuals newly accessing into the Army. In addition, geographic distribution of weight status, as well as other sociodemographic characteristics, and the long-term impact of weight-related standards, deserve examination. The present study therefore examined how weight status at accession into the Army has changed between 1989 and 2012; how weight status over the 24-year interval has changed relative to demographic characteristics; and whether these changes were impacted by revisions in weight-for-height criteria and body fat standards for entry. These observations will inform Army leaders about long-term trends of weight status at accession, and may allow recruitment offices to target appropriate populations with weightrelated education or intervention programs. Methods Study design and participants This study used existing historical data from the Total Army Injury and Health Outcomes Database (TAIHOD) for the years The TAIHOD is a data repository that maintains records of administrative and health-related data sets on active-duty Army soldiers to support epidemiologic research and analysis (27). Data from the TAIHOD on Army enlisted personnel was drawn from the Military Entrance Processing Command data set: date, height, weight, and geographic home of record at application (N 5 1,763,352 unique observations); and from the Defense Manpower Data Center Master Personnel and Transaction data set: date of birth, sex, race/ethnicity, education, and marital status (N 5 2,649,681 unique observations). The interval between application and successful accession into the Army was up to 18 months. In the combined data set, there were 1,758,445 unique observations appearing in both the accession and the personnel files, , indicating individuals successfully accessed into the Army and were considered personnel thereafter. We excluded 12,210 (0.69%) participants missing or having implausible recorded height (<4.5 or 7 ft), weight (<80 or 450 lbs) or calculated BMI (<10 or 50 kg/m 2 ), 4,668 (0.27%) participants missing date of birth, and an additional 497 (0.03%) participants missing information on sex, for 1,741,070 participants for the present analysis. The present secondary data analysis was approved by the Human Use Review Committee, US Army Research Institute of Environmental Medicine, Natick, MA. Measures and categories of body size BMI (kg/m 2 ) was calculated as weight (kg) divided by height (m), squared. Participants were categorized as underweight (BMI <18.5 kg/m 2 ), normal weight (18.5-<25 kg/m 2 ), overweight (25-<30 kg/m 2 ), or obese (30 kg/m 2 ) (18). Screening table weights (STW) from AR were used to categorize participants as meeting or exceeding the age- and sex-specific weight-for-height criteria for that year (14). Changes in these standards over the study period are described in Supporting Information Tables S1 and S2. In brief, there were five primary intervals in the 24-year time period to which we assigned indicators: (1) 1989 (study start) September 1991; (2) October 1991 (introduction of body fat standards) June 2006; (3) July 2006 [increase in body fat standards for 17- to 20-year-olds effective July 2006 and following widespread introduction of the Assessment of Recruit Motivation and Strength Study (ARMS) fitness-based waivers in February 2006 (17,28)] December 2007; (4) January 2008 (changes to minimum body weight in men and women and maximum body weight in women) April 2012; and (5) May 2012 (return to pre-2006 body fat standards for 17- to 20-year-olds) December 2012 (study end). Neither body fat nor physical fitness data were available. Demographic covariates Demographic covariates were categorized as follows: age (<20, 20-<30, 30-<40, 40 years old), sex (male, female), geographic region (state or country home of record) based on US Census Bureau-designated divisions (29), level of educational attainment (<high school, high school, some college or college, advanced degree), race/ethnicity (white, black, Hispanic, Asian/Pacific Islander, American Indian/ Native Alaskan), and marital status (never married, married, divorced/separated/widowed). In lieu of excluding participants with missing demographic information, individuals with missing data on variables other than age, sex, and anthropometry were classified as Other/Unknown. Statistical analyses Descriptive statistics (e.g., means, frequencies) were generated to examine the longitudinal course (over 24 years in time intervals and by year) of weight status as continuous (BMI) and dichotomous (overweight/obese vs. not, and exceeding vs. meeting STW) variables. Crude (unadjusted) prevalence is presented because the participants in the present analysis are considered to be the entire population VOLUME 23 NUMBER 3 MARCH

3 in Entering US Soldiers, Hruby et al. TABLE 1 Demographic characteristics of those accessing into the US Army, by time interval of changes in standard weight-for-height or body fat criteria, a December 2012) Overall N 250,440 1,026, , ,227 37,920 1,741,070 Sex, % female Age, % <20 yrs <30 yrs <40 yrs yrs BMI, kg/m (3.4) 24.4 (3.6) 24.9 (4.0) 24.9 (3.8) 24.6 (3.5) BMI categories, % Underweight Normal weight Overweight Obese Overweight or obese Exceeding STW, % Race/ethnicity, % Non-Hispanic white Non-Hispanic black Hispanic Asian/Pacific Islander American Indian/Native Alaskan Other/unknown Education, % <High school High school Some college or college Advanced degree Unknown Marital status, % Never married Married Separated/divorced/widowed Unknown Geographic region, % New England Mid-Atlantic South Atlantic East North Central East South Central West North Central West South Central Mountain Pacific US Territories Outside US Unknown a s are given as percentages in each interval and overall, except for BMI, which is mean (standard deviation). BMI, body mass index; STW, screening table weight, the age- and sex-specific weight-for-height screening criteria. 664 VOLUME 23 NUMBER 3 MARCH

4 Figure prevalence of (a) overweight (body mass index 25 <30 kg/m 2 ) and (b) obesity (body mass index 30 kg/m 2 ) in men (black bars), in women (gray bars), and in the total population (striped bars). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] accessing into the Army in this time period. Trends across the 24-year period for categories of BMI and STW were assessed by linear or logistic regression for continuous and binary outcomes, respectively, with time interval or year as primary predictors. Multivariate logistic regression was used to estimate demographic associations of overweight/obesity and STW within each time interval/year, adjusted for all demographic covariates simultaneously, to generate the odds ratio (OR) and 95% confidence intervals (95% CI) for each variable within each time period. Overall ORs were also estimated across the entire period, additionally adjusting for the effect of the time interval/year. All statistical procedures were performed using SAS (v9.3, SAS Institute, Cary, NC). While a twosided alpha level of <0.05 was considered statistically significant, owing to the very large sample size, P values for almost every statistical test were < We therefore limit the presentation of P values and rely on estimates and confidence intervals as indicators of the strength and/or consistency of association. Results Table 1 shows the unadjusted mean BMI, and annual and overall prevalence of overweight and obesity and demographic covariates for soldiers at accession into the US Army by time interval defined by revisions to weight-for-height criteria or body fat standards (Supporting Information Table S3 presents prevalence by year). Overweight/obesity Of the 1,741,070 individuals entering the Army in , 40.9% were overweight or obese, of which 32.9% were overweight (34.6% of men, 24.7% of women) and 8.0% were obese (9.4% of men, 1.0% of women). Across the study period, the prevalence of overweight trended upward from a low of 25.8% (1989), to a peak of 37.9% (2011) (Figure 1a). Overweight prevalence has remained >30% since The prevalence of obesity also generally increased, from a low of 5.0% (1993), to a high of 12.3% (2009), before decreasing to 8.0% (2012) (Figure 1b). In tests of time trend via multivariate regression with time interval as a predictor, BMI increased by kg/m 2 per interval, and odds of being overweight/obese increased by 13% per interval (OR 1.129, 95%CI ). Within each interval, the most consistent and strongest demographic characteristics predicting overweight/obesity were sex and age (Table 2). (Predictors by year are shown in Supporting Information Table S4.) Women were consistently less likely to be overweight/ obese at accession compared to men: odds of being overweight/ obese in women compared to men were initially 0.16 (95%CI ) in the first interval, before climbing to 0.45 (95%CI ) in the second interval, after the introduction of body fat standards. Since then, women were between 48% (interval 3) and 58% (intervals 4, 5) less likely to be overweight/obese at entry than men. Older age was consistently associated with higher odds of being overweight/obese at accession. Hispanic and Asian/Pacific Islanders had consistently higher odds, while American Indian/Native Alaskans had generally lower odds, of being overweight/obese compared to non-hispanic whites. Non-Hispanic blacks did not have strong odds of overweight/obesity relative to non-hispanic whites. Education had inconsistent trends, but in general, higher education attainment seemed to be associated with higher odds of overweight/obesity. Compared to those who were never married, being married, or VOLUME 23 NUMBER 3 MARCH

5 in Entering US Soldiers, Hruby et al. TABLE 2 Odds ratios (95% CI) of demographic characteristics associated with overweight/obesity across five time intervals and overall, a OR (95% CI) of overweight/obesity December 2012) Overall Sex Male 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) Female 0.16 (0.15, 0.16) 0.45 (0.45, 0.46) 0.52 (0.51, 0.54) 0.46 (0.45, 0.47) 0.42 (0.40, 0.45) 0.42 (0.42, 0.43) Age <20 yrs 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 20-<30 yrs 1.56 (1.53, 1.59) 1.58 (1.57, 1.60) 1.40 (1.36, 1.43) 1.56 (1.54, 1.59) 1.66 (1.58, 1.73) 1.56 (1.55, 1.57) 30-<40 yrs 2.10 (1.98, 2.22) 1.97 (1.93, 2.01) 1.46 (1.40, 1.53) 1.75 (1.70, 1.81) 2.33 (2.01, 2.69) 1.85 (1.82, 1.88) 40 yrs 2.70 (1.74, 4.19) 2.35 (2.05, 2.70) 2.40 (2.14, 2.68) 2.05 (1.89, 2.23) 2.46 (0.39, 15.44) 2.32 (2.19, 2.46) Race/ethnicity Non-Hispanic white 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) Non-Hispanic black 0.91 (0.89, 0.93) 1.01 (1.00, 1.02) 0.94 (0.91, 0.98) 0.99 (0.97, 1.01) 0.99 (0.94, 1.05) 0.99 (0.98, 0.99) Hispanic 1.32 (1.26, 1.37) 1.35 (1.33, 1.37) 1.48 (1.42, 1.54) 1.41 (1.37, 1.44) 1.48 (1.39, 1.58) 1.37 (1.36, 1.39) Asian/Pacific Islander 1.38 (1.25, 1.51) 1.34 (1.29, 1.40) 1.36 (1.21, 1.53) 1.20 (1.11, 1.30) 1.15 (0.94, 1.42) 1.32 (1.28, 1.36) American Indian/Native Alaskan 0.75 (0.69, 0.80) 0.91 (0.89, 0.94) 1.10 (1.05, 1.16) 0.96 (0.92, 0.99) 1.00 (0.90, 1.12) 0.94 (0.92, 0.96) Other/unknown 1.13 (1.06, 1.21) 1.07 (1.03, 1.12) 0.83 (0.59, 1.16) 0.54 (0.40, 0.73) (1.04, 1.12) Education <High school 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) High school 1.26 (1.20, 1.32) 1.00 (0.99, 1.02) 1.26 (1.17, 1.36) 1.09 (1.06, 1.13) 1.02 (0.91, 1.15) 1.04 (1.03, 1.06) Some college or college 1.60 (1.50, 1.71) 1.10 (1.08, 1.12) 1.16 (1.08, 1.25) 1.05 (1.01, 1.09) 1.07 (0.95, 1.21) 1.08 (1.07, 1.10) Advanced degree 1.49 (1.02, 2.17) 1.36 (1.24, 1.50) 1.69 (1.32, 2.16) 1.21 (1.09, 1.34) 0.90 (0.64, 1.27) 1.29 (1.21, 1.38) Unknown 1.15 (0.63, 2.12) 1.11 (1.09, 1.13) 1.26 (1.16, 1.36) 0.69 (0.63, 0.75) 0.80 (0.43, 1.50) 1.12 (1.11, 1.14) Marital status Never married 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) Married 1.20 (1.17, 1.23) 1.37 (1.36, 1.39) 1.50 (1.45, 1.55) 1.50 (1.47, 1.53) 1.30 (1.22, 1.39) 1.39 (1.37, 1.40) Separated/divorced/widowed 1.09 (1.02, 1.18) 1.34 (1.29, 1.38) 1.55 (1.43, 1.68) 1.48 (1.40, 1.57) 1.36 (1.07, 1.74) 1.35 (1.32, 1.39) Unknown 1.32 (1.17, 1.49) 1.14 (1.09, 1.19) 0.49 (0.30, 0.82) 0.49 (0.28, 0.84) (1.11, 1.20) Geographic region, % New England 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) Mid-Atlantic 0.92 (0.87, 0.97) 0.94 (0.91, 0.96) 0.97 (0.90, 1.05) 0.93 (0.89, 0.98) 1.02 (0.90, 1.16) 0.94 (0.92, 0.95) 666 VOLUME 23 NUMBER 3 MARCH

6 TABLE 2. (continued). OR (95% CI) of overweight/obesity December 2012) Overall South Atlantic 0.93 (0.89, 0.98) 0.93 (0.91, 0.95) 0.96 (0.90, 1.04) 0.88 (0.85, 0.92) 0.89 (0.79, 1.00) 0.92 (0.90, 0.94) East North Central 0.94 (0.89, 0.98) 0.95 (0.93, 0.97) 0.99 (0.92, 1.07) 0.94 (0.90, 0.98) 0.99 (0.88, 1.12) 0.95 (0.93, 0.97) West North Central 0.91 (0.86, 0.96) 0.95 (0.93, 0.98) 0.97 (0.89, 1.05) 0.94 (0.89, 0.99) 0.84 (0.73, 0.97) 0.94 (0.92, 0.96) East South Central 0.96 (0.91, 1.01) 0.95 (0.92, 0.97) 0.95 (0.88, 1.03) 1.00 (0.95, 1.05) 1.03 (0.89, 1.18) 0.96 (0.94, 0.98) West South Central 0.94 (0.89, 0.99) 0.96 (0.94, 0.99) 1.00 (0.93, 1.08) 0.94 (0.90, 0.99) 0.92 (0.82, 1.05) 0.95 (0.94, 0.97) Mountain 0.80 (0.75, 0.84) 0.85 (0.83, 0.87) 0.88 (0.81, 0.95) 0.85 (0.81, 0.89) 0.83 (0.72, 0.95) 0.84 (0.82, 0.86) Pacific 0.97 (0.93, 1.02) 0.98 (0.96, 1.01) 1.02 (0.95, 1.10) 0.96 (0.92, 1.01) 0.94 (0.83, 1.06) 0.97 (0.96, 0.99) US Territories 0.78 (0.70, 0.87) 0.83 (0.80, 0.87) 0.70 (0.61, 0.80) 0.80 (0.74, 0.88) 0.81 (0.62, 1.07) 0.81 (0.78, 0.84) Outside US 0.46 (0.32, 0.67) 0.63 (0.54, 0.72) 0.48 (0.26, 0.88) 0.78 (0.60, 1.03) 0.86 (0.43, 1.72) 0.63 (0.56, 0.70) Unknown 1.05 (0.90, 1.22) 1.07 (1.01, 1.14) 1.05 (0.78, 1.42) 1.73 (1.44, 2.08) 3.56 (1.93, 6.58) 1.11 (1.06, 1.17) a Associations are mutually adjusted for all demographic covariates within each interval. The overall association was also adjusted for the interval indicator, which had on OR (95% CI) of ( ), indicating 13% higher odds of overweight/obesity per interval period. separated/divorced/widowed was associated with higher odds of being overweight/obese. Finally, there were no distinct geographic trends with overweight/obesity, using New England as the arbitrarily selected reference category. Exceeding STW In the entire time period, 23.6% of individuals exceeded STW (21.1% of men, 35.5% of women) at entry. Of those who were not overweight/obese, 3.4% were nevertheless classified as exceeding STW, predominantly women (16.1%, compared to just 0.02% of men). Meanwhile, of those individuals who were overweight/obese, 52.6% were classified as exceeding STW (91.9% of women vs. 48.0% of men). The overall prevalence of exceeding STW increased with time, from a low of 5.7% (1989), jumping to 21.9% (1992) with the introduction of body fat standards, to a high of 31.0% (2006, 2007), shortly after the minimum body fat standard in the youngest age group was raised in mid Exceeding STW has since decreased to 25.2% in 2012 (Table 1 and Supporting Information Table S3). In tests of time trend via multivariate regressions with time interval as a predictor, odds of exceeding STW increased by 28% per interval (OR 1.276, 95%CI ), indicating the strong effect of the introduction of, and changes in, STW and body fat standards as a final arbiter of acceptable body composition. In contrast to odds of overweight/obesity, women had higher adjusted odds of exceeding STW only the first three time intervals, with the highest odds observed in the first interval, prior to the introduction of body fat standards (Table 3). (Predictors by year are shown in Supporting Information Table S5.) In the last two intervals (2008 forward), women were >35% less likely to exceed STW than men, likely owing to the approximately 10-pound increase in maximum weight-for-height beginning in January 2008, despite still being >50% less likely to be overweight/obese than men in the last two intervals. During the first three intervals, 19.2% of women who were not overweight/obese were classified as exceeding STW, versus 0.02% of men; in the last two intervals, just 0.61% of women who were not overweight/obese were classified as exceeding STW, versus 0.02% of men. Associations with age were less clear: while odds of exceeding STW were consistently higher in the 20-<30-year-olds relative to those <20 years old, the associations for those 30-<40 years old and 40 years old and older fluctuated from interval to interval. Across intervals, non-hispanic blacks and Hispanics were increasingly likely to exceed STW relative to non- Hispanic whites. There were no consistent trends of educational attainment, marital status, or geography with odds of exceeding STW. Discussion Our data show trends of overweight and obesity that generally reflect the trends of these conditions in the US adult civilian population. We observed increasing prevalence of overweight/obesity across the 24-year time period, nearing 50% of those entering into the Army in Overweight prevalence rose to approximately 35% in men and 25-30% in women, mimicking the prevalence of the US population, which was approximately 33% in men and 27% in women aged years old in (3). While overall rates of obesity in our population were just a third of the age-adjusted 34.9% of US adults in 2012 (3), obesity nevertheless more than doubled from approximately 5% in the late 1980s/early 1990s, to over 12% in Unlike US civilian populations in which women are slightly more likely to be obese than men, in the enlistee population we VOLUME 23 NUMBER 3 MARCH

7 in Entering US Soldiers, Hruby et al. TABLE 3 Odds ratios (95% CI) of demographic characteristics associated with exceeding STW across five time intervals and overall, a OR (95% CI) of exceeding STW December 2012) Overall Sex Male 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) Female (15.35, 16.53) 2.31 (2.28, 2.33) 2.05 (1.99, 2.12) 0.63 (0.62, 0.65) 0.53 (0.49, 0.57) 2.02 (2.00, 2.04) Age <20 yrs 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 20-<30 yrs 1.14 (1.09, 1.18) 1.34 (1.33, 1.35) 1.15 (1.12, 1.18) 1.30 (1.28, 1.33) 1.39 (1.32, 1.47) 1.31 (1.30, 1.32) 30-<40 yrs 0.85 (0.76, 0.96) 1.15 (1.12, 1.17) 0.87 (0.82, 0.91) 1.09 (1.05, 1.13) 1.43 (1.22, 1.67) 1.10 (1.08, 1.12) 40 yrs 1.85 (0.75, 4.58) 0.83 (0.70, 0.98) 0.97 (0.86, 1.09) 1.01 (0.92, 1.10) 0.96 (0.11, 8.70) 1.00 (0.94, 1.07) Race/Ethnicity Non-Hispanic White 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) Non-Hispanic Black 0.99 (0.95, 1.04) 1.10 (1.08, 1.11) 1.00 (0.97, 1.04) 1.12 (1.10, 1.14) 1.20 (1.13, 1.28) 1.09 (1.08, 1.10) Hispanic 0.92 (0.83, 1.01) 1.32 (1.30, 1.34) 1.51 (1.45, 1.57) 1.42 (1.38, 1.45) 1.42 (1.32, 1.52) 1.34 (1.32, 1.36) Asian/Pacific Islander 1.12 (0.94, 1.34) 1.31 (1.25, 1.37) 1.33 (1.18, 1.50) 1.32 (1.21, 1.43) 1.16 (0.92, 1.48) 1.30 (1.25, 1.35) American Indian/Native Alaskan 0.66 (0.56, 0.78) 0.93 (0.90, 0.96) 1.08 (1.02, 1.14) 0.95 (0.91, 0.99) 1.08 (0.95, 1.22) 0.95 (0.93, 0.97) Other/unknown 0.89 (0.77, 1.03) 0.99 (0.94, 1.04) 0.78 (0.53, 1.15) 0.49 (0.33, 0.75) (0.93, 1.02) Education <High school 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) High school 1.33 (1.17, 1.50) 0.99 (0.97, 1.01) 1.31 (1.21, 1.43) 1.12 (1.08, 1.16) 1.10 (0.97, 1.26) 1.04 (1.02, 1.06) Some college or college 1.27 (1.09, 1.48) 1.02 (1.00, 1.05) 1.19 (1.10, 1.29) 0.98 (0.94, 1.02) 0.94 (0.82, 1.08) 1.02 (1.00, 1.03) Advanced degree 1.15 (0.57, 2.33) 0.93 (0.83, 1.03) 1.29 (1.00, 1.67) 0.86 (0.77, 0.97) 0.66 (0.42, 1.02) 0.87 (0.81, 0.94) Unknown 0.63 (0.08, 4.85) 1.11 (1.09, 1.14) 1.27 (1.16, 1.38) 0.70 (0.63, 0.77) 1.03 (0.51, 2.08) 1.13 (1.11, 1.15) Marital Status Never married 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) Married 0.91 (0.86, 0.96) 1.23 (1.21, 1.24) 1.38 (1.34, 1.43) 1.39 (1.36, 1.42) 1.22 (1.14, 1.31) 1.25 (1.24, 1.26) Separated/divorced/widowed 0.80 (0.70, 0.91) 1.15 (1.11, 1.18) 1.36 (1.25, 1.47) 1.37 (1.29, 1.45) 0.98 (0.74, 1.30) 1.18 (1.15, 1.22) Unknown 0.91 (0.71, 1.18) 1.05 (1.00, 1.10) 0.38 (0.19, 0.77) 0.29 (0.11, 0.72) (0.99, 1.09) Geographic region, % New England 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) 1 (ref.) Mid-Atlantic 1.08 (0.96, 1.21) 0.97 (0.94, 1.00) 0.96 (0.88, 1.05) 0.89 (0.85, 0.94) 0.97 (0.83, 1.12) 0.96 (0.94, 0.98) 668 VOLUME 23 NUMBER 3 MARCH

8 TABLE 3. (continued). OR (95% CI) of exceeding STW December 2012) Overall South Atlantic 0.93 (0.83, 1.04) 0.99 (0.96, 1.01) 0.99 (0.92, 1.07) 0.87 (0.83, 0.91) 0.90 (0.79, 1.03) 0.99 (0.96, 1.01) East North Central 0.98 (0.88, 1.09) 1.00 (0.97, 1.03) 1.03 (0.95, 1.12) 0.92 (0.88, 0.96) 1.02 (0.89, 1.18) 0.99 (0.96, 1.01) West North Central 0.96 (0.85, 1.08) 0.98 (0.95, 1.01) 0.98 (0.90, 1.07) 0.90 (0.85, 0.95) 0.97 (0.83, 1.15) 0.97 (0.94, 0.99) East South Central 1.00 (0.89, 1.13) 0.97 (0.94, 1.00) 1.02 (0.93, 1.11) 1.01 (0.96, 1.07) 1.11 (0.95, 1.30) 0.95 (0.93, 0.97) West South Central 0.88 (0.79, 0.99) 1.04 (1.02, 1.07) 1.02 (0.95, 1.11) 0.94 (0.90, 0.99) 1.01 (0.87, 1.16) 1.01 (0.99, 1.03) Mountain 0.97 (0.86, 1.10) 0.90 (0.87, 0.93) 0.91 (0.83, 0.99) 0.85 (0.81, 0.89) 0.87 (0.75, 1.02) 0.89 (0.87, 0.91) Pacific 1.03 (0.92, 1.15) 0.99 (0.97, 1.02) 1.04 (0.96, 1.13) 0.93 (0.89, 0.98) 1.03 (0.89, 1.18) 0.98 (0.96, 1.00) US Territories 0.75 (0.58, 0.96) 0.73 (0.69, 0.77) 0.65 (0.56, 0.75) 0.61 (0.56, 0.67) 0.61 (0.44, 0.83) 0.70 (0.67, 0.73) Outside US 0.59 (0.34, 1.03) 0.69 (0.59, 0.81) 0.53 (0.28, 1.01) 0.58 (0.42, 0.81) 0.76 (0.33, 1.75) 0.65 (0.57, 0.75) Unknown 0.91 (0.65, 1.26) 1.14 (1.07, 1.22) 1.35 (0.99, 1.84) 1.77 (1.47, 2.12) 1.97 (1.08, 3.59) 1.16 (1.10, 1.23) a Associations are mutually adjusted for all demographic covariates within each interval. The overall association was also adjusted for the interval indicator, which had on OR (95% CI) of ( ), indicating 28% higher odds of exceeding STW per interval period. STW, screening table weight, the age- and sex-specific weight-for-height screening criteria. studied, it was men driving obesity prevalence: an average 9.4% of men versus 1.0% of women were classified as obese in the time period studied. In contrast to trends observed in the US population, we did not observe higher odds of overweight/obesity in non-hispanic blacks, as compared to non-hispanic whites, perhaps owing to disproportionate burden of overweight/obesity among female non-hispanic blacks in the civilian population (1), who represent just 5.5% of the present study population. However, Hispanics, as well as Asian/Pacific Island races/ethnicities tended to have higher odds of overweight/obesity, and of exceeding weight-for-height criteria, consistent with other studies (23). Contrary to the high rates of obesity observed in the southern and mid-western areas of the US over this period (26), we did not detect geographic trends associated with overweight/obesity. Our analyses complement published data on 18-year-olds first accessing into the military between 1993 and 2006, which showed that the crude prevalence of overweight/obesity increased from 25.6% to 33.9% in this age group over that time period (21). Our observations are also generally consistent with short-term studies involving active-duty military, such as those emanating from the DoD Surveys of Health-Related Behaviors Among Active Duty Military Personnel. Data from the 2002 and 2005 surveys indicated that the combined prevalence of self-reported overweight/obesity in military personnel increased to 2005 (60.5%) with 4.2% higher prevalence of obesity in 2005 (12.9%) compared to 2002 (8.7%). In 2011, the self-reported prevalence of obesity remained high, at 12.4% in the armed services overall (6.4% of females, and 13.5% of males), but it was 15.8% in the Army, the highest of the services (24). Similar to our findings, women were less likely than men to be overweight or obese in the survey years, but obesity was more prevalent with increasing age, black race or Hispanic ethnicity, and with being married (23,24). Some authors speculate that BMI does not accurately estimate body fat in male respondents, misclassifying as overweight what is actually excess lean mass in a very physically active population (23). However, in the present analysis, we observed considerably lower odds of overweight/obesity, but higher odds of exceeding STW in women compared to men, and more normal-weight women than men were nevertheless considered as exceeding STW. Changes to the STW since 2008 have brought the calculated maximum BMIs for men and women closer to each other (currently up to 28.6 vs kg/m 2, respectively), although they remain higher in males, perhaps to account for potential excess muscle mass in males, which is usually not a factor in females. This suggests that the Army screening criteria and body fat standard at entry play a key role in subsequent active-duty differences between sexes in terms of prevalence of overweight/obesity, and changes in screening weight-for-height criteria and body fat standards seem to have had an impact on women, in particular. The prevalence of overweight in women entering the Army increased sharply after the introduction of body fat standards in 1991, and rose again after the loosening of body fat standards and increases in maximum weightfor-height criteria in the mid-2000s, suggesting that screening criteria for women were initially too strict relative to those for men, as suggested by others (30). These observations about weight status in women at accession may underlie earlier findings that, unlike in the US civilian population, women enlisting and remaining on active duty are less likely to be overweight/obese than men (23,24). Overweight/obesity is a leading cause of medical disqualifications of applicants screened at recruiting stations, disqualifying some % VOLUME 23 NUMBER 3 MARCH

9 in Entering US Soldiers, Hruby et al. of applicants between 2007 and 2012 (13). Body weight has significant repercussions in the Army. In addition to being a barrier to entry, those nevertheless permitted to enter who exceed body fat standards are placed into the Army Weight Control Program, which consists of intensive exercise programs, nutritional counseling, and related interventions to monitor aggressive weight-loss goals consistent with weight and body fat retention standards described in AR 600-9, which are similar to accession standards (AR ) (16,19). If a soldier does not make satisfactory progress within six months, he/she may be processed for separation from the Army (16,19). In addition, between 1998 and 2010, personnel with incident overweight-related diagnoses left the service a median of 15 months earlier than their counterparts without such conditions. From , obese soldiers had shorter lengths of service following diagnosis than overweight soldiers (9). Beyond the importance of conforming to body size and composition criteria, overweight/obesity predisposes soldiers to cardiometabolic and other diseases typical of nonmilitary overweight/ obese populations, but with heightened risk of job-related military-specific risks, such as heat illness (10) and musculoskeletal injury (11). Our study has strengths and limitations. We benefitted from exceptionally large data sets that have generally consistently assessed the weight, height, and demographic characteristics of those accessing into the Army over this 24-year period. We were missing or had implausible values for <1% of participants for age, sex, and anthropometrics. While we did not have data on body fat or physical fitness, we may assume that a majority of those not meeting the weight-for-height criteria ultimately either met body fat standards or passed physical fitness tests. This study has practical implications: first, it provides military and public health leaders with long-term trends in overweight/obesity of individuals accessing into the Army, and indicates that the Army is not immune, nor has it been immune, to the increasing rates of obesity in the US civilian population. That the prevalence of overweight/obesity reached nearly half of those entering the Army just a few years ago is astounding. Second, it reveals the broad impact of the changing screening criteria and body fat standards on the prevalence of overweight/obesity in the Army, which can be expected to subsequently affect health and healthcare costs associated with these conditions in years to come. Future studies should assess the implications of BMI at accession into the Army on subsequent health status of soldiers, including musculoskeletal injuries and or/disease risk factors incident during a soldier s career, as well as analyses of cost-effectiveness of future revisions to accession standards. This will provide insight into the potential costs of the obesity epidemic to military readiness, to DoD healthcare expenditures, and ultimately, to the nation s health. O Acknowledgments The authors would like to thank the millions of US Army soldiers who, in entering into military service for this country, contributed their data to this analysis. VC 2015 The Society References 1. Fryar CD, Carroll MD, Ogden CL. NCHS Health E Stat Overweight, obesity, and extreme obesity among adults: United States, Trends Through Fryar CD, Carroll MD, Ogden CL. NCHS Health E Stats Overweight prevalence among children and adolescents Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of childhood and adult obesity in the United States, JAMA 2014;311: Yamane GK. in civilian adults: Potential impact on eligibility for U.S. military enlistment. Mil Med 2007;172: McLaughlin R, Wittert G. The obesity epidemic: implications for recruitment and retention of defence force personnel. Obes Rev 2009;10: Gattis VM. : A Threat to National Security?. Carlisle, PA: U.S. Army War College; Popkin BM. Is the obesity epidemic a national security issue around the globe? Curr Opin Endocrinol Diabetes Obes 2011;18: Packnett ER, Niebuhr DW, Bedno SA, Cowan DN. Body mass index, medical qualification status, and discharge during the first year of US Army service. Am J Clin Nutr 2011;93: Armed Forces Health Surveillance Center. Duration of service after overweightrelated diagnoses, active component, U.S. Armed Forces, Med Surveill Mon Rep 2011;18: Bedno SA, Li Y, Han W, et al. Exertional heat illness among overweight U.S. army recruits in basic training. Aviat Space Environ Med 2010;81: Cowan DN, Bedno SA, Urban N, Yi B, Niebuhr DW. Musculoskeletal injuries among overweight army trainees: incidence and health care utilization. Occup Med 2011;61: Dall TM, Zhang Y, Chen YJ, et al. Cost associated with being overweight and with obesity, high alcohol consumption, and tobacco use within the military health system s TRICARE prime-enrolled population. Am J Health Promot 2007;22: Anon. Accession Medical Standards Analysis and Research Activity Annual Report. Silver Spring, MD: Walter Reed Army Institute of Research; U.S. Department of the Army. Standards of Medical Fitness, Army Regulation Loughran DS, Orvis BR. The effect of the Assessment of Recruit Motivation and Strength (ARMS) program on army accessions and attrition. Santa Monica, CA: Rand Corporation; Bedno SA, Lang CE, Daniell WE, Wiesen AR, Datu B, Niebuhr DW. Association of weight at enlistment with enrollment in the Army Weight Control Program and subsequent attrition in the Assessment of Recruit Motivation and Strength Study. Mil Med 2010;175: Niebuhr DW, Scott CT, Li Y, Bedno SA, Han W, Powers TE. Preaccession fitness and body composition as predictors of attrition in U.S. Army recruits. Mil Med 2009;174: National Institutes of Health, National Heart, Lung, and Blood Institute. Education Initiative: Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults: executive summary. Expert Panel on the Identification, Evaluation, and Treatment of Overweight in Adults. Bethesda, MD: U.S. Department of Health and Human Services; Friedl KE. Can you be large and not obese?. The distinction between body weight, body fat, and abdominal fat in occupational standards. Diabetes Technol Ther 2004; 6: Friedl KE. Body composition and military performance Many things to many people. J Strength Cond Res Natl Strength Cond Assoc 2012;26(Suppl 2):S Hsu LL, Nevin RL, Tobler SK, Rubertone MV. Trends in overweight and obesity among 18-year-old applicants to the United States Military, J Adolesc Health 2007;41: Lindquist CH, Bray RM. Trends in overweight and physical activity among U.S. Military Personnel, Prev Med 2001;32: Smith TJ, Marriott BP, Dotson L, et al. Overweight and obesity in military personnel: Sociodemographic predictors. Obes Silver Spring Md 2012;20: Barlas FM, Higgins WB, Pflieger JC, Diecker K Health Related Behaviors Survey of Active Duty Military Personnel. Executive Summary. U.S. Department of Defense, TRICARE Management Activity, Defense Health Cost Assessment and Program Evaluation, and U.S. Coast Guard; Sanderson PW, Clemes SA, Biddle SJH. The correlates and treatment of obesity in military populations: A systematic review. Obes Facts 2011;4: Centers for Disease Control and Prevention (CDC). Adult Facts. Obes Overweight Prof Data Stat Adult Obes - DNPAO - CDC. 27. Amoroso PJ, Yore MM, Weyandt B, Jones BH. Chapter 8. Total Army injury and health outcomes database: A model comprehensive research database. Mil Med 1999;164: Niebuhr DW, Page WF, Cowan DN, Urban N, Gubata ME, Richard P. Costeffectiveness analysis of the U.S. Army Assessment of Recruit Motivation and Strength (ARMS) program. Mil Med 2013;178: U.S. Census Bureau. Regions and divisions. Econ Census Reg Div. 30. Bathalon GP, McGraw SM, Sharp MA, Williamson DA, Young AJ, Friedl KE. The effect of proposed improvements to the Army Weight Control Program on female soldiers. Mil Med 2006;171: VOLUME 23 NUMBER 3 MARCH