- Email: [email protected]
Overweight and Obesity Trends Among Active Duty Military Personnel
Overweight and Obesity Trends Among Active Duty Military Personnel A 13-Year Perspective Carolyn M. Reyes-Guzman, MPH, Robert M. Bray, PhD, Valerie L. Forman-Hoffman, PhD, MPH, Jason Williams, PhD Background: The U.S. population has shown increasing rates of overweight and obesity in recent years, but similar analyses do not exist for U.S. military personnel. It is important to understand these patterns in the military because of their impact on fitness and readiness. Purpose: To assess prevalence and trends in overweight/obesity among U.S. servicemembers and to examine the associations of sociodemographic characteristics, exercise, depression, and substance use with these patterns. Methods: Analyses performed in 2013 used five large population–based health-related behavior surveys conducted from 1995 to 2008. Main outcome measures were overweight and obesity among active duty military personnel based on BMI. Results: Combined overweight and obesity (BMIZ25) increased from 50.6% in 1995 to 60.8% in 2008, primarily driven by the rise in obesity (BMIZ30) from 5.0% to 12.7%. For overweight, military women showed the largest increase. For obesity, all sociodemographic groups showed significant increases, with the largest among warrant officers, senior enlisted personnel, and people aged 36–45 years. Adjusted multinomial logit analyses found that servicemembers aged 26 years and older, men, non-Hispanic blacks and Hispanics, enlisted personnel, married personnel, and heavy drinkers had the highest risk both for overweight and obesity.
Conclusions: Combined overweight and obesity in active duty personnel rose to more than 60% between 1995 and 2008, primarily because of increased obesity. The high prevalence of overweight and obesity needs attention and has implications for DoD efforts to improve the health, fitness, readiness, and quality of life of the Active Forces. (Am J Prev Med 2014;](]):]]]–]]]) & 2014 American Journal of Preventive Medicine
Introduction
O
verweight and obesity are paramount health concerns for the U.S. Using BMI, the most recent national estimates of overweight and obesity indicate that 33.8% of Americans aged Z20 years are obese (BMIZ30) and 68.0% are either overweight or obese (BMIZ25).1 These high rates are of great concern given that overweight and obesity increase the risk for several chronic health conditions, morbidity, and excess mortality.2,3 In addition to physical consequences, From RTI International, Research Triangle Park, North Carolina Ms. Reyes-Guzman and Dr. Bray contributed equally to this paper. Address correspondence to: Robert M. Bray, PhD, RTI International, 3040 Cornwallis Road, P.O. Box 12194, Research Triangle Park NC 27709. E-mail: [email protected]. 0749-3797/$36.00 http://dx.doi.org/10.1016/j.amepre.2014.08.033
& 2014 American Journal of Preventive Medicine
overweight and obesity are also related to selected mental health conditions (e.g., depression, anxiety) and substance use issues (e.g., alcohol use disorders, smoking).4–6 The resultant medical cost of overweight and obesity in the U.S., estimated at $147 billion per year,7 is staggering. Overweight and obesity are also of high concern to the U.S. military because they result in health risks, degrade military performance and readiness, and are costly in terms of productivity loss and morbidity. Each Service screens for overweight/obesity using a combination of BMI and waist circumference as an indicator of body fat percentage at least annually. Estimates indicate that overweight and obesity cost the active duty (AD) military nearly $106 million per year in lost productivity8 and $1.1 billion per year treating obesity-related conditions.8 Although overweight and obesity appear to be lower in the military than among civilians, studies during the past
Published by Elsevier Inc.
Am J Prev Med 2014;](]):]]]–]]] 1
2
Reyes-Guzman et al / Am J Prev Med 2014;](]):]]]–]]]
decade show increases in the military. Using BMI as an indicator, Lindquist and Bray9 found an increase in overweight (BMIZ25) among AD military personnel aged Z20 years from 50% in 1995 to 54% in 1998. A follow-on analysis on attainment of Healthy People 2000 objectives10 found a further increase in overweight to 58% in 2002 and an increase in the percentage of those not meeting the overweight objectives. The Millennium Cohort study11 reported a combined rate of 60% for overweight/obesity in 2005. Recent directives from Department of Defense (DoD) leadership provide policy guidance on overweight and obesity. In 2003, the DoD mandated that the military “place emphasis on Healthy People 2010 Leading Health Indicators,” which include overweight and obesity. In 2004, the DoD updated a directive on physical fitness and body fat standards.12 It has not been well documented whether these directives have had a positive impact on the weight of military personnel. Despite initial findings showing increases in overweight and obesity in the military, questions remain about the shape of this trend in more recent years and the sociodemographic and behavioral correlates associated with weight status. This paper addresses this gap and has a twofold purpose: (1) to assess the prevalence and trends in overweight and obesity over a 13-year period using population-based data; and (2) to identify the associations of sociodemographic and military characteristics, exercise, depression, and substance use with overweight and obesity among AD military personnel. This information will help military leaders understand changes in and correlates of weight status, and adjust programs and policies as needed to ensure the fitness and readiness of military servicemembers.
the 2008 survey) and stratifying by gender and pay grade; it also included personnel who were remote at the outset. Women and officers were oversampled. Data were collected by civilian teams using onsite group administrations of anonymous surveys or by mailing questionnaires to eligible participants who did not attend the onsite administrations. Sample sizes ranged from 12,756 (2002) to 28,546 (2008). Response rates showed declines from 1995 to 2005 (70%, 59%, 56%, and 52%, respectively) but improved in 2008 (70%) because of the replacement sampling. Data were weighted to represent the eligible AD population. IRB approval was obtained from RTI International and the DoD.
Measures The primary variables of interest included overweight and obesity status estimated using BMI. As recommended by CDC and the National Heart Lung and Blood Institute,10,19 healthy weight was defined as a BMI of 18.5–24.9, overweight as 25.0–29.9, and obese as Z30.0. Height and weight were self-reported. Variables examined for association with overweight and obesity included age; gender; branch of Service; race/ethnicity; pay grade (rank); education; and marital status. Marital status was dichotomized as married (or “living as married”) and non-married (separated, divorced, widowed, and single, never-married). The multivariate model also examined the associations of exercise, depression, heavy alcohol use, and heavy smoking with overweight and obesity. For this model, a dichotomous pay grade variable was created for enlisted personnel and officers. Exercise was defined as moderate (Z3 days per week for 20 minutes or longer) or vigorous (Z5 days per week for 30 minutes or longer). Heavy drinking was defined as five or more drinks (four or more for women) on the same occasion at least once a week in the past 30 days consistent with definitions used in civilian national surveys.20,21 Heavy smoking was defined as smoking one or more packs per day in the past 30 days. Finally, need for further depression evaluation was assessed using the Burnam screen.22,23 Personnel were defined as needing further evaluation if they felt sad, blue, or depressed for 2 weeks or more in the past 12 months or reported Z2 years in their lifetime of feeling depressed and felt depressed “much of the time” in the past 12 months and felt depressed on Z1 days in the past week.
Methods Data Source, Sampling, and Data Collection Data were drawn from the 1995,13 1998,14 2002,15 2005,16 and 200817 DoD Surveys of Health-Related Behaviors Among AD Military Personnel (HRB). Another survey, conducted in 2011,18 was not included because of incompatible variables and unknown effects of changes in survey methodology. The sampling design and data collection methods were similar for the 1995–2008 population-based surveys and included personnel from the Army, Navy, Marine Corps, and Air Force. The eligible population for each survey included all AD personnel at the time of data collection except recruits, academy cadets, and personnel who were absent without leave, incarcerated, or undergoing a permanent change of station. First-stage sampling involved random selection of installations (and Navy ships) within each of the four services, both within and outside of the continental U.S. Second-stage sampling consisted of randomly selecting personnel at these installations without replacement (except for
Statistical Analysis Analyses were conducted in 2013 using SUDAAN, version 10, to account for the complex survey design of the HRB data. Analyses also used HRB sampling weights to generate results to represent the entire AD military population and account for survey nonresponse. Descriptive statistics were calculated to summarize demographic and weight-related characteristics of AD personnel for each of the survey years from 1995 to 2008. A generalized multinomial logit model examined the associations of sociodemographic characteristics, exercise, depression, and other psychosocial characteristics (heavy cigarette smoking and heavy alcohol use) with overweight and obesity outcomes.
Results The demographic distributions of AD personnel across survey years are presented in Table 1. The military www.ajpmonline.org
3
Reyes-Guzman et al / Am J Prev Med 2014;](]):]]]–]]]
Table 1. Estimated sociodemographics of active duty military personnel, 1995–2008, % (SE) unless otherwise noted Characteristic
1995
1998
2002
2005
2008
N¼16,193
N¼17,264
N¼12,756
N¼16,146
N¼24,690
N¼1,325,396
N¼1,113,710
N¼1,125,083
N¼1,011,852
N¼1,311,581
69.6
59.0
55.6
51.8
70.1
r25
43.8 (1.4)
38.6 (1.3)
46.7 (1.9)
46.7 (1.9)
46.9 (2.0)
26–35
36.2 (0.7)
38.1 (0.7)
31.8 (0.7)
32.8 (1.0)
31.9 (0.8)
36–45
18.1 (0.8)
20.8 (0.8)
18.9 (1.3)
17.8 (1.0)
18.1 (1.1)
Z46
2.0 (0.2)
2.5 (0.2)
2.6 (0.4)
2.7 (0.3)
3.1 (0.4)
Army
31.9 (1.7)
34.0 (1.6)
33.8 (2.2)
31.8 (5.0)
38.7 (4.2)
Navy
28.8 (1.8)
25.8 (1.7)
25.4 (2.1)
26.8 (3.5)
23.9 (4.1)
Marine Corps
11.0 (0.7)
12.2 (1.1)
13.5 (1.9)
12.7 (2.2)
13.4 (2.9)
Air Force
28.4 (1.3)
28.0 (1.3)
27.4 (2.7)
28.7 (3.0)
24.0 (3.6)
Male
87.6 (0.9)
86.3 (0.7)
83.1 (0.8)
85.2 (0.7)
85.7 (0.9)
Female
12.4 (0.9)
13.7 (0.7)
16.9 (0.8)
14.8 (0.7)
14.3 (0.9)
NH White
67.7 (1.1)
66.0 (0.9)
67.3 (1.3)
64.4 (1.3)
64.0 (1.2)
NH Black/AA
17.2 (0.8)
18.0 (0.8)
20.7 (1.4)
17.6 (1.0)
16.7 (0.8)
Hispanic
8.5 (0.4)
8.8 (0.5)
7.1 (0.4)
8.8 (0.5)
10.4 (0.5)
Other
6.6 (0.4)
7.3 (0.4)
5.0 (0.5)
9.2 (0.6)
8.9 (0.5)
E1–E3
21.8 (1.0)
18.9 (0.9)
22.0 (1.6)
24.0 (1.8)
21.0 (1.5)
E4–E6
52.2 (1.4)
52.5 (1.2)
51.9 (1.0)
49.6 (1.8)
51.7 (2.4)
E7–E9
10.4 (0.5)
10.8 (0.5)
10.8 (0.8)
9.7 (0.8)
10.2 (0.6)
W1–W5
1.0 (0.2)
1.2 (0.2)
1.2 (0.2)
1.0 (0.1)
1.4 (0.7)
O1–O3
8.7 (0.8)
9.5 (0.8)
8.3 (0.5)
9.4 (1.1)
9.3 (0.7)
O4–O10
5.9 (0.8)
7.2 (0.7)
5.8 (1.1)
6.3 (0.8)
6.4 (0.8)
Married
63.9 (0.9)
62.5 (0.8)
58.6 (1.1)
59.2 (1.3)
59.4 (1.1)
Unmarried
36.1 (0.9)
37.5 (0.8)
41.4 (1.1)
40.8 (1.3)
40.6 (1.1)
Unweighted total Weighted total a
Response rate (%) Age (years)
Branch of service
Gender
Race/ethnicityb
Pay grade
c
Marital status
a
Response rate is defined as the rate that usable questionnaires from eligible personnel (see exclusion criteria in Methods section) were obtained for both phases of data collection. Other category includes Asian, Native Hawaiian, Pacific Islander, American Indian/Alaska Native, and “other” responses; these were collapsed because of small sample sizes. For 1998, 2.18% of responses for race/ethnicity were missing (not displayed). c E1–E3¼junior enlisted personnel; E4–E6¼mid-level enlisted personnel; E7–E9¼senior enlisted personnel; W1–W5¼warrant officers; O1–O3¼junior officers; O4–O10¼senior officers. AA, African American; NH, Non-Hispanic. b
] 2014
4
Reyes-Guzman et al / Am J Prev Med 2014;](]):]]]–]]]
population was predominantly male, white, relatively young, married, and concentrated in pay grades E4–E6. During the 13-year span, most characteristics remained relatively stable. There were increases in the percentages of the youngest personnel, personnel in the Army, Hispanic or “other” racial minorities, and unmarried personnel. Figure 1 presents the overall changes in overweight and obesity from 1995 to 2008. The proportion of overweight individuals showed a slight increase of 2.5 percentage points during the 13-year period, but the proportion of obese personnel increased nearly 8 percentage points from 1995 to 2008. The combined overweight and obese categories increased 10 percentage points from 50.6 % to 60.8%. To better understand these changes, Table 2 examined overweight and obesity within demographic and military groupings. Consistent with the findings from Figure 1, trends in overweight for each of the demographic groups were relatively stable, although there were a few exceptions. The most striking increase in overweight occurred among military women, who had a gain of nearly 14 percentage points (20.8% to 34.6%). There were also some differences among pay grade groups. Junior officers (O1–O3) showed an increase of 8 percentage points, followed by junior enlisted personnel and warrant officers. By contrast, the rates of obesity showed striking increases. The groups with the largest gains were warrant officers (13.4 percentage points) and senior enlisted personnel (13.2 percentage points) followed by people aged 36–45 years (12.3 percentage points); non-Hispanic blacks/African Americans (11.3 percentage points); and Hispanics (10.7 percentage points). Married personnel
showed larger increases in the prevalence of obesity (9.2 percentage points) compared to unmarried personnel. All demographic and service groups showed significant increases in the rates from 1995 to 2008, and the majority showed increases of Z5 percentage points. Tests of the trends were also conducted in Table 2. These included linear contrasts across the five survey periods for each demographic variable and pairwise comparisons for each combination of survey pairs (e.g., 1995 versus 1998, 2002, 2005, and 2008) within each demographic variable. Results for all tests for trends were statistically significant at po0.05. Table 3 presents adjusted associations of overweight or obesity and demographic, military, and behavioral factors from the 2008 survey. Multinomial models were created to examine the relationship between overweight and obesity and behavioral predictors while controlling for the demographic and military variables examined in Table 2. The first multinomial comparison examined the associations among demographic, military, and behavioral factors and overweight, compared with servicemembers in the healthy weight group. Results showed that the highest overweight risks were for servicemembers aged Z36 years and for men. In addition, non-Hispanic African Americans and Hispanics had 36%–62% higher overweight prevalence, respectively, than those of other race/ethnicities. There was also a moderate association between overweight and heavy drinking (21% greater risk than lighter drinking or no drinking) and between overweight and heavy smoking (19% lower risk than among lighter smokers and nonsmokers). Enlisted and married personnel both had higher prevalence of overweight compared to their reference groups (40% and 26%, respectively).
70.0 BMI 25.0-29.9
BMI 30+
60.0
Percentage
50.0
5.0
6.2
8.8
13.1
12.7
49.0
48.2
48.1
2002
2005
2008
40.0 30.0 20.0
45.6
48.4
1995
1998
10.0 0.0 Survey Year *BMI is based on self-reports of height and weight; does not differentiate between muscle and fat.
Figure 1. Trends in BMI, 1995–2008. www.ajpmonline.org
] 2014
Table 2. Prevalence of overweight and obese active duty personnel by sociodemographic characteristics, 1995–2008, % (SE) 1995 Characteristic
1998
2002
2005
2008
Overweight
Obese
Overweight
Obese
Overweight
Obese
Overweight
Obese
Overweight
Obese
r25
36.8 (0.7)
3.2 (0.1)
38.7 (0.6)
4.1 (0.2)
41.3 (0.7)
5.4 (0.3)
42.0 (0.8)
9.0 (0.3)
40.7 (0.9)
7.9 (0.2)
26–35
49.8 (0.4)
5.8 (0.2)
51.9 (0.4)
7.3 (0.2)
52.1 (0.5)
10.9 (0.3)
51.0 (0.7)
15.0 (0.2)
53.0 (0.5)
15.3 (0.2)
36–45
57.0 (0.5)
7.5 (0.1)
58.6 (0.5)
8.4 (0.2)
60.2 (0.7)
13.0 (0.2)
57.4 (0.6)
19.7 (0.3)
56.1 (0.6)
19.8 (0.3)
Z46
57.5 (0.1)
5.1 (0.1)
58.8 (0.1)
4.8 (0.1)
60.9 (0.3)
10.9 (0.1)
55.8 (0.2)
14.1 (0.1)
59.4 (0.2)
14.5 (0.1)
Army
44.9 (0.7)
4.0 (0.2)
47.7 (0.8)
5.4 (0.2)
48.3 (1.0)
9.3 (0.3)
49.7 (2.2)
10.2 (0.6)
48.7 (1.8)
13.1 (0.6)
Navy
46.0 (0.9)
7.7 (0.2)
50.5 (0.8)
10.3 (0.3)
49.8 (0.9)
11.6 (0.4)
46.8 (1.7)
18.5 (0.7)
48.9 (1.5)
14.4 (0.4)
Marine Corps
47.0 (0.3)
1.9 (0.1)
47.9 (0.6)
2.3 (0.1)
46.9 (0.8)
3.7 (0.1)
48.2 (1.1)
7.1 (0.1)
49.7 (1.4)
6.2 (0.2)
Air Force
45.6 (0.7)
4.4 (0.1)
47.7 (0.7)
5.2 (0.1)
50.1 (1.3)
8.3 (0.2)
47.8 (1.5)
13.7 (0.5)
45.6 (1.2)
13.9 (0.4)
49.1 (0.6)
5.4 (0.3)
52.0 (0.5)
7.0 (0.3)
52.9 (0.6)
9.8 (0.4)
50.6 (0.6)
13.9 (0.5)
50.4 (0.6)
13.8 (0.4)
20.8 (0.2)
1.4 (0.1)
25.5 (0.2)
1.4 (0.1)
29.5 (0.3)
3.9 (0.1)
33.9 (0.3)
8.4 (0.1)
34.6 (0.4)
6.7 (0.1)
NH White
45.5 (0.7)
4.3 (0.2)
48.6 (0.6)
5.4 (0.2)
49.8 (0.8)
7.9 (0.3)
48.6 (1.0)
11.4 (0.4)
47.9 (0.7)
11.1 (0.3)
NH Black/AA
46.4 (0.4)
8.0 (0.2)
49.9 (0.4)
9.7 (0.2)
47.6 (0.6)
11.7 (0.2)
46.9 (0.5)
20.8 (0.4)
47.7 (0.5)
19.3 (0.2)
Hispanic
48.9 (0.3)
3.3 (0.1)
49.7 (0.3)
5.5 (0.1)
48.7 (0.3)
11.4 (0.2)
50.8 (0.3)
13.0 (0.1)
52.8 (0.3)
14.0 (0.1)
Other
40.3 (0.2)
6.3 (0.1)
42.6 (0.2)
5.3 (0.1)
43.7 (0.3)
6.4 (0.1)
45.3 (0.3)
10.5 (0.1)
45.6 (0.3)
12.1 (0.1)
E1–E3
34.4 (0.4)
3.0 (0.1)
37.2 (0.5)
3.8 (0.1)
39.8 (0.7)
5.5 (0.2)
40.1 (0.7)
9.9 (0.3)
39.0 (0.5)
6.6 (0.1)
E4–E6
47.5 (0.7)
5.9 (0.3)
48.7 (0.7)
7.4 (0.3)
48.4 (0.6)
9.8 (0.3)
47.9 (1.0)
14.8 (0.3)
48.0 (1.1)
14.4 (0.4)
E7–E9
57.9 (0.3)
7.8 (0.1)
61.4 (0.3)
8.0 (0.1)
60.1 (0.5)
13.9 (0.2)
58.1 (0.4)
19.2 (0.2)
60.5 (0.4)
21.0 (0.1)
W1–W5
55.4 (0.1)
4.1 (0.1)
60.2 (0.1)
5.8 (0.1)
58.5 (0.1)
8.8 (0.1)
61.8 (0.1)
9.8 (0.1)
59.6 (0.4)
17.5 (0.1)
O1–O3
41.6 (0.3)
1.8 (0.1)
47.5 (0.4)
3.7 (0.1)
52.1 (0.3)
5.6 (0.1)
51.9 (0.6)
7.9 (0.2)
50.0 (0.3)
8.0 (0.1)
O4–O10
52.1 (0.4)
3.8 (0.1)
54.9 (0.3)
4.2 (0.1)
59.5 (0.6)
7.1 (0.1)
56.1 (0.5)
9.8 (0.1)
51.8 (0.4)
11.5 (0.1)
Age (years)
Gender Male Female Race/ethnicity
Pay grade
a
Reyes-Guzman et al / Am J Prev Med 2014;](]):]]]–]]]
Branch of service
b
5
(continued on the next page)
6
Note: Only estimates for overweight and obese personnel are reported. Percentages for overweight, obese, and healthy weight groups add to 100%. Linear contrasts over time and pairwise comparisons between survey years were statistically significant at po0.05. a Other category includes Asian, Native Hawaiian, Pacific Islander, American Indian/Alaska Native, and "other" responses; these were collapsed because of small sample sizes. For 1998, 2.28% of responses were missing (not displayed). b E1–E3 ¼ junior enlisted personnel; E4–E6¼midlevel enlisted personnel; E7–E9¼senior enlisted personnel; W1–W5¼warrant officers; O1–O3¼junior officers; O4–O10¼senior officers. AA, African American; NH, Non-Hispanic.
12.7 (0.4) 13.1 (0.6) 49.0 (0.6) 6.2 (0.3) 48.4 (0.5) 5.0 (0.3) 45.6 (0.5) Total
38.6 (0.5) Unmarried
8.8 (0.5)
48.2 (0.7)
48.1 (0.4)
9.0 (0.1) 43.5 (0.5) 9.6 (0.3) 44.2 (0.7) 43.0 (0.6) 4.7 (0.2) 42.0 (0.4)
52.3 (0.5) 6.1 (0.3) 49.5 (0.6) Married
Marital status
2.8 (0.1)
53.1 (0.9) 7.1 (0.2)
Obese Overweight Obese Overweight Characteristic
5.5 (0.2)
15.4 (0.4) 11.1 (0.3)
50.8 (0.9)
51.3 (0.6)
Overweight Obese Overweight Overweight
Obese
2005 2002 1998 1995
Table 2. Prevalence of overweight and obese active duty personnel by sociodemographic characteristics, 1995–2008, % (SE) (continued)
2008
Obese
15.3 (0.3)
Reyes-Guzman et al / Am J Prev Med 2014;](]):]]]–]]]
The second multinomial comparison examined the associations among the same demographic, military, and behavioral factors and obesity compared to servicemembers of healthy weight. Similar to the overweight analysis, the highest risk of obesity was also found in the oldest two age groups and for men (ORsZ4). Other strong predictors of obesity were being aged 26–35 years, being in the Army, Navy, or Air Force, being non-Hispanic African American, married, and enlisted (ORsZ2). Individuals with a need for further depression evaluation were more likely to be obese compared to those not depressed (OR¼1.36, 95% CI¼1.18, 1.55), and heavy drinkers were more likely to be obese than non-heavy drinkers (OR¼1.24, 95% CI¼1.04, 1.46). Exercise was not a statistically significant predictor of either overweight or obesity.
Discussion This study examined changes in overweight and obesity among AD military personnel from 1995 to 2008. Findings showed an overall increasing trend for combined overweight and obesity of 10 percentage points from 50.6% to 60.8%, primarily driven by the nearly 8 percentage point rise in obesity. Of note, the increases of combined overweight and obesity leveled off between 2005 and 2008, suggesting a possible change in the trend line. These findings also mirror civilian trends showing a deceleration in the late 2000s of previously increasing rates.1 In the current study, the pattern in overall trends was found across most examined subgroups. Although it is possible that awareness of the detrimental effects of obesity and military-specific initiatives to curb the growing epidemic may have slowed the increase in rates, the last period of no change may reflect historic trends of steep increases followed by periods of relative stability.24 This study found several key correlates of overweight and obesity, with age and gender being the strongest predictors, consistent with findings from other studies.25 Of interest, the 36- to 45-year age group had equal or higher overweight and obesity rates than people aged 46 years or older. This could be due to a reduction in lean muscle mass associated with aging and not by a reduction in adipose tissue.26 These findings, however, support the notion that aging servicemembers may experience a decrease in daily exercise as they transition to more senior phases of their careers that typically involve more sedentary time spent behind a desk rather than engaged in physical combat–readiness activities. Both heavy drinkers and servicemembers who screened positive for depression were more likely to be obese than their counterparts.27–32 These findings suggest that servicemembers identified as heavy drinkers or depressed www.ajpmonline.org
7
Reyes-Guzman et al / Am J Prev Med 2014;](]):]]]–]]]
Table 3. Associations between sociodemographic and behavioral risk factors for overweight or obese status, 2008, AOR (95% CI)a Overweight vs healthy
Group
Obese vs healthy
Age (years) 1.00
Yes
26–35
2.13 (1.94, 2.34)
2.88 (2.58, 3.20)
36–45
3.03 (2.66, 3.45)
4.82 (4.15, 5.61)
Z46
3.40 (2.86, 4.03)
4.16 (3.37, 5.14)
Army
1.03 (0.93, 1.14)
2.00 (1.62, 2.47)
Navy
1.11 (1.00, 1.25)
2.35 (1.95, 2.82)
Air Force
0.96 (0.88, 1.04)
2.30 (1.95, 2.72)
Branch of service
Marine Corps
1.00
Gender Male
2.36 (2.14, 2.61)
Female
1.00
NH White
1.11 (0.98, 1.25)
1.01 (0.82, 1.24)
NH Black/AA
1.36 (1.17, 1.59)
2.09 (1.67, 2.61)
Hispanic
1.62 (1.39, 1.88)
1.76 (1.41, 2.21)
Other
1.00
Marital status Married
1.26 (1.16, 1.36)
Unmarried
1.67 (1.52, 1.84) 1.00
Pay grade Enlisted
1.40 (1.23, 1.60)
Officers
2.37 (2.02, 2.77) 1.00
b
Yes
1.08 (0.97, 1.20)
No
0.98 (0.84, 1.14) 1.00
c
Heavy drinker Yes
1.21 (1.10, 1.33)
No
1.23 (1.04, 1.46) 1.00
Heavy smokerd
No
0.81 (0.69, 0.95)
0.88 (0.70, 1.10) 1.00 (continued)
] 2014
Overweight vs healthy
Obese vs healthy
1.04 (0.91, 1.18)
1.36 (1.18, 1.55)
No
1.00
Note: Boldface indicates statistical significance. a This multilog model uses a three-level weight category variable as the outcome. The weight categories include obese individuals (BMIZ30); overweight individuals (BMI¼25–29.9); and individuals with a healthy weight (BMI¼18.5–24.9). BMI is defined as weight (kg)/[height (m)]2. The survey uses self-reports of height and weight; however, BMI does not differentiate between muscle and fat. b This variable includes vigorous exercise for Z20 minutes on Z3 days per week, and moderate exercise for Z30 minutes on Z5 days per week. c Heavy drinking was defined as Z5 drinks on the same drinking occasion at least once a week in the past 30 days. d Heavy smoking was defined as smoking Z1 packs per day (415 cigarettes) in the past 30 days. AA, African American; NH, Non-Hispanic.
3.95 (3.38, 4.61)
Race/ethnicity
Yes
Group Depression
r25
Exercise
Table 3. (continued )
should be carefully screened and monitored for obesity. Conversely, the drinking behaviors and levels of depression of overweight or obese servicemembers should be assessed and treated by healthcare providers as well. Providers should consider these associations when recommending a particular weight loss, depression, or substance use intervention and the potential impact of this treatment on other comorbid conditions. Although some literature suggests an intricate relationship between overweight/obesity and alcohol use, where light to moderate drinkers are less likely to be overweight or obese33 than nondrinkers, the present study did not find similar results. It is possible that these discrepancies are applicable to women of a higher socioeconomic status (e.g., wine drinkers) but are not applicable to the U.S. military. Of note, vigorous/moderate exercise was not associated with overweight or obesity consistent with findings by Lindquist and Bray.9 This suggests that factors besides a sedentary lifestyle play an important role in explaining the observed increases. These factors might include genetics, body composition, dietary intake, or fluctuating exercise during the year. The use of BMI as a measure of overweight and obesity has several caveats worth noting. BMI is correlated with body fat percentage34 and thus is a useful surveillance tool at the population level.35 There are times, however, when an individual has an elevated BMI due to increased muscle mass instead of body fat. We hypothesized that muscular servicemembers with high standards of fitness
8
Reyes-Guzman et al / Am J Prev Med 2014;](]):]]]–]]]
and physical readiness may be classified as overweight or obese according to BMI despite having healthy or even low levels of adiposity. A study of servicemembers by Heinrich and colleagues,36 however, showed that categorizing obesity based on measured BMI actually underestimated the prevalence of obesity compared to using body fat testing to classify obesity. Although it is true that the current study used self-reported rather than measured height and weight to calculate BMI, the underestimation may be particularly pronounced because self-reported height and weight tend to underestimate the prevalence of overweight and obesity in nationally representative samples.37 Nonetheless, prior studies have not examined whether categorizing overweight (but not obese) status based on BMI leads to similar patterns of misclassification in military samples. In evaluating these findings, some limitations should be acknowledged. Modest response rates during some of the survey years raise some questions about possible nonresponse bias. A bias would be expected to result in under-reporting of weight such that BMI would be within normal limits. The finding of substantial overweight/ obesity coupled with increases over time argues against a serious problem. A second limitation is that the use of self-reported height and weight data to calculate BMI and classify overweight and obesity may have introduced reporting bias. Prior studies, however, have found reasonable correlations of self-reported and measured height and weight data38 and suggest that self-reports are reasonable to use in epidemiologic studies.39 Also, potential bias in self-reported estimates was likely nondifferential across time and thus less likely to affect trend data. Third, the use of cross-sectional data to study trends also has some limitations in that trends can be influenced by the characteristics of each sample and period effects of public health campaigns, advances in healthcare access and delivery, and various other cultural influences on weight-related behaviors. Fortunately, differences in sample characteristics are likely to be minimal because of the highly homogeneous nature of the military population, and the statistical weighting that was done to ensure that samples represented the broader military population. Despite these limitations, the current study has several notable strengths. It is based on five large probability samples representative of the AD population. The study used sampling weights that were post-stratified to reflect population totals and used anonymous questionnaires collected by civilian rather than military survey teams. The large sample size permits observation of robust trend estimates overall and within subgroups of interest (e.g., women, depressed personnel, heavy drinkers). The identification of specific demographic and behavioral risk characteristics correlated with overweight and
obesity elucidates potential targets for weight-related prevention and treatment efforts. Overall, the 2008 survey met the Healthy People 2010 objective of o15% obesity prevalence (12.8%); however, several subgroups failed to meet this goal. Servicemembers aged 26–45 years, of non-Hispanic black/African American race/ ethnicity, who reported being married, and who were senior enlisted personnel or warrant officers had obesity prevalence estimates 415%. In multivariate models, heavy drinkers and those who screened positive for depression also had elevated obesity risk. These findings suggest potential targets for weight-related prevention and intervention efforts and highlight characteristics associated with overweight and obesity for healthcare providers to use in patient assessment. In addition to clinical significance, these findings suggest the need for future research and policy considerations. Although the DoD has traditionally focused on health promotion and disease prevention among its troops and Military Health System (MHS) beneficiaries, the prevalence of substantial numbers of both overweight and obese servicemembers requires attention. As a complement to the DoD’s efforts, the MHS has obesity intervention programs, counseling for nutrition improvement, and other health education and wellness classes. Future research should obtain a more nuanced understanding of the predisposing, enabling, and maintenance factors of weight problems in AD servicemembers. In addition, evaluation of current DoD and MHS initiatives may help identify helpful characteristics of successful programs. Overweight and obesity in the military are critical issues currently affecting servicemembers’ health, fitness, and quality of life. Development and implementation of programs that foster a healthy, active force may help control rising medical and non-medical costs while ultimately improving the health and well-being of servicemembers. The authors wish to thank Justin Faerber for editorial assistance in preparing the manuscript. The paper was prepared using internal funds provided by RTI International. The views, opinions, and findings contained in this report are those of the authors and should not be construed as an official Department of Defense position, policy, or decision, unless so designated by other official documentation. No financial disclosures were reported by the authors of this paper.
References 1. Flegal KM, Carroll MD, Ogden CL, Curtin LR. Prevalence and trends in obesity among US adults, 1999-2008. JAMA 2010;303(3):235–41.
www.ajpmonline.org
Reyes-Guzman et al / Am J Prev Med 2014;](]):]]]–]]] 2. Allison DB, Fontaine KR, Manson JE, Stevens J, VanItallie TB. Annual deaths attributable to obesity in the United States. JAMA 1999;282(16): 1530–8. 3. Flegal KM, Graubard BI, Williamson DF, Gail MH. Cause-specific excess deaths associated with underweight, overweight, and obesity. JAMA 2007;298(17):2028–37. 4. Cilli M, De Rosa R, Pandolfi C, et al. Quantification of subclinical anxiety and depression in essentially obese patients and normal-weight healthy subjects. Eat Weight Disord 2003;8(4): 319–20. 5. Rivenes AC, Harvey SB, Mykletun A. The relationship between abdominal fat, obesity, and common mental disorders: results from the HUNT study. J Psychosom Res 2009;66(4):269–75. 6. Petry NM, Barry D, Pietrzak RH, Wagner JA. Overweight and obesity are associated with psychiatric disorders: results from the National Epidemiologic Survey on Alcohol and Related Conditions. Psychosom Med 2008;70(3):288–97. 7. Finkelstein EA, Trogdon JG, Cohen JW, Dietz W. Annual medical spending attributable to obesity: payer-and service-specific estimates. Health Aff (Millwood) 2009;28(5):w822–w831. 8. 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(2):120–39. 9. Lindquist CH, Bray RM. Trends in overweight and physical activity among U.S. military personnel, 1995–1998. Prev Med 2001;32(1): 57–65. 10. Bray RM, Rae Olmsted KL, Williams J, Sanchez RP, Hartzell M. Progress toward healthy people 2000 objectives among U.S. military personnel. Prev Med 2006;42(5):390–6. 11. Smith TJ, Marriott BP, Dotson L, et al. Overweight and obesity in military personnel: sociodemographic predictors. Obesity (Silver Spring) 2012;20(7):1534–8. 12. Department of Defense (DoD). DoD Physical Fitness and Body Fat Program. Washington DC: Department of Defense, 2004. 13. Bray RM, Kroutil LA, Wheeless SC, et al. 1995 Department of Defense Survey of Health Related Behaviors Among Military Personnel. Raleigh NC: Research Triangle Institute, 1995. 14. Bray RM, Sanchez RP, Ornstein ML, et al. 1998 Department of Defense Survey of Health Related Behaviors Among Military Personnel. Raleigh NC: Research Triangle Institute, 1999. 15. Bray RM, Hourani L, Rae KL, et al. 2002 Department of Defense Survey of Health Related Behaviors Among Military Personnel. Raleigh NC: Research Triangle Institute, 2003. 16. Bray RM, Hourani LL, Rae Olmsted KL, et al. 2005 Department of Defense Survey of Health Related Behaviors Among Active Duty Military Personnel. Raleigh NC: Research Triangle Insitute, 2006. 17. Bray RM, Pemberton MR, Hourani L, et al. 2008 Department of Defense Survey of Health Related Behaviors Among Active Duty Military Personnel. Raleigh NC: Research Triangle Institute, 2009. 18. Barlas FM, Higgins WB, Pflieger JC, Diecker K. 2011 Health Related Behaviors Survey of Active Duty Military Personnel. Executive summary 2013, June 27, 2013. tricare.mil/tma/dhcape/surveys/coresur veys/surveyhealthrelatedbehaviors/ADS.aspx. 19. CDC. Healthy weight—it’s not a diet, it’s a lifestyle! www.cdc.gov/ healthyweight/. 20. Substance Abuse and Mental Health Services Administration (SAMHSA). Results from the 2010 National Survey on Drug Use and Health: national findings. Rockville MD: Substance Abuse and Mental Health Services Administration, Office of Applied Studies, 2011.
] 2014
9
21. Johnston LD, O’Malley PM, Bachman JG, Schulenberg JE. Monitoring the Future national survey results on drug use, Volume II: 1975-2010. College students and adults ages 19-50. Bethesda MD: National Institute on Drug Abuse, 2011. 22. Burnam MA, Wells KB, Leake B, Landsverk J. Development of a brief screening instrument for detecting depressive disorders. Med Care 1988;26(8):775–89. 23. Rost K, Burnam MA, Smith GR. Development of screeners for depressive disorders and substance disorder history. Med Care 1993;31(3):189–200. 24. Costa D, Steckel R. Long-term trends in health, welfare, and economic growth in the United States. In: Steckel R, Floud R, eds. Health and welfare during industrialization. Chicago IL: University of Chicago Press, 1997:47–90. 25. Kumanyika SK. Special issues regarding obesity in minority populations. Ann Intern Med 1993;119(7 Pt 2):650–4. 26. Gallagher D, Visser M, Sepulveda D, Pierson RN, Harris T, Heymsfield SB. How useful is body mass index for comparison of body fatness across age, sex, and ethnic groups? Am J Epidemiol 1996;143(3):228–39. 27. Must A, Spadano J, Coakley EH, Field AE, Colditz G, Dietz WH. The disease burden associated with overweight and obesity. JAMA 1999;282(16):1523–9. 28. Gregg EW, Cheng YJ, Cadwell BL, et al. Secular trends in cardiovascular disease risk factors according to body mass index in US adults. JAMA 2005;293(15):1868–74. 29. Wilson PW, D'Agostino RB, Sullivan L, Parise H, Kannel WB. Overweight and obesity as determinants of cardiovascular risk: the Framingham experience. Arch Intern Med 2002;162(16):1867–72. 30. Lahti-Koski M, Pietinen P, Heliovaara M, Vartiainen E. Associations of body mass index and obesity with physical activity, food choices, alcohol intake, and smoking in the 1982–1997 FINRISK Studies. Am J Clin Nutr 2002;75(5):809–17. 31. Onyike CU, Crum RM, Lee HB, Lyketsos CG, Eaton WW. Is obesity associated with major depression? Results from the Third National Health and Nutrition Examination Survey. Am J Epidemiol 2003; 158(12):1139–47. 32. Shelton N, Knott C. Association between alcohol calorie intake and overweight and obesity in english adults. Am J Public Health 2014;104(4): 639–1. 33. Wang L, Lee IM, Manson JE, Buring JE, Sesso HD. Alcohol consumption, weight gain, and risk of becoming overweight in middle-aged and older women. Arch Intern Med 2010;170(5):453–61. 34. Flegal KM, Shepherd JA, Looker AC, et al. Comparisons of percentage body fat, body mass index, waist circumference, and waist-stature ratio in adults. Am J Clin Nutr 2009;89(2):500–8. 35. Bouchard C. How much progress have we made over the last few decades? Int J Obes 2008;32(S7):S2–S7. 36. Heinrich KM, Jitnarin N, Suminski RR, et al. Obesity classification in military personnel: a comparison of body fat, waist circumference, and body mass index measurements. Mil Med 2008;173(1):67–73. 37. Yun S, Zhu B, Black W, Brownson R. A comparison of national estimates of obesity prevalence from the behavioral risk factor surveillance system and the National Health and Nutrition Examination Survey. Int J Obes 2005;30(1):164–70. 38. Kuczmarski MF, Kuczmarski RJ, Najjar M. Effects of age on validity of self-reported height, weight, and body mass index: findings from the Third National Health and Nutrition Examination Survey, 1988–1994. J Am Diet Assoc 2001;101(1):28–34; (quiz 5–6). 39. Spencer EA, Appleby PN, Davey GK, Key TJ. Validity of self-reported height and weight in 4808 EPIC-Oxford participants. Public Health Nutr 2002;5(4):561–5.
Conclusions: Combined overweight and obesity in active duty personnel rose to more than 60% between 1995 and 2008, primarily because of increased obesity. The high prevalence of overweight and obesity needs attention and has implications for DoD efforts to improve the health, fitness, readiness, and quality of life of the Active Forces. (Am J Prev Med 2014;](]):]]]–]]]) & 2014 American Journal of Preventive Medicine
Introduction
O
verweight and obesity are paramount health concerns for the U.S. Using BMI, the most recent national estimates of overweight and obesity indicate that 33.8% of Americans aged Z20 years are obese (BMIZ30) and 68.0% are either overweight or obese (BMIZ25).1 These high rates are of great concern given that overweight and obesity increase the risk for several chronic health conditions, morbidity, and excess mortality.2,3 In addition to physical consequences, From RTI International, Research Triangle Park, North Carolina Ms. Reyes-Guzman and Dr. Bray contributed equally to this paper. Address correspondence to: Robert M. Bray, PhD, RTI International, 3040 Cornwallis Road, P.O. Box 12194, Research Triangle Park NC 27709. E-mail: [email protected]. 0749-3797/$36.00 http://dx.doi.org/10.1016/j.amepre.2014.08.033
& 2014 American Journal of Preventive Medicine
overweight and obesity are also related to selected mental health conditions (e.g., depression, anxiety) and substance use issues (e.g., alcohol use disorders, smoking).4–6 The resultant medical cost of overweight and obesity in the U.S., estimated at $147 billion per year,7 is staggering. Overweight and obesity are also of high concern to the U.S. military because they result in health risks, degrade military performance and readiness, and are costly in terms of productivity loss and morbidity. Each Service screens for overweight/obesity using a combination of BMI and waist circumference as an indicator of body fat percentage at least annually. Estimates indicate that overweight and obesity cost the active duty (AD) military nearly $106 million per year in lost productivity8 and $1.1 billion per year treating obesity-related conditions.8 Although overweight and obesity appear to be lower in the military than among civilians, studies during the past
Published by Elsevier Inc.
Am J Prev Med 2014;](]):]]]–]]] 1
2
Reyes-Guzman et al / Am J Prev Med 2014;](]):]]]–]]]
decade show increases in the military. Using BMI as an indicator, Lindquist and Bray9 found an increase in overweight (BMIZ25) among AD military personnel aged Z20 years from 50% in 1995 to 54% in 1998. A follow-on analysis on attainment of Healthy People 2000 objectives10 found a further increase in overweight to 58% in 2002 and an increase in the percentage of those not meeting the overweight objectives. The Millennium Cohort study11 reported a combined rate of 60% for overweight/obesity in 2005. Recent directives from Department of Defense (DoD) leadership provide policy guidance on overweight and obesity. In 2003, the DoD mandated that the military “place emphasis on Healthy People 2010 Leading Health Indicators,” which include overweight and obesity. In 2004, the DoD updated a directive on physical fitness and body fat standards.12 It has not been well documented whether these directives have had a positive impact on the weight of military personnel. Despite initial findings showing increases in overweight and obesity in the military, questions remain about the shape of this trend in more recent years and the sociodemographic and behavioral correlates associated with weight status. This paper addresses this gap and has a twofold purpose: (1) to assess the prevalence and trends in overweight and obesity over a 13-year period using population-based data; and (2) to identify the associations of sociodemographic and military characteristics, exercise, depression, and substance use with overweight and obesity among AD military personnel. This information will help military leaders understand changes in and correlates of weight status, and adjust programs and policies as needed to ensure the fitness and readiness of military servicemembers.
the 2008 survey) and stratifying by gender and pay grade; it also included personnel who were remote at the outset. Women and officers were oversampled. Data were collected by civilian teams using onsite group administrations of anonymous surveys or by mailing questionnaires to eligible participants who did not attend the onsite administrations. Sample sizes ranged from 12,756 (2002) to 28,546 (2008). Response rates showed declines from 1995 to 2005 (70%, 59%, 56%, and 52%, respectively) but improved in 2008 (70%) because of the replacement sampling. Data were weighted to represent the eligible AD population. IRB approval was obtained from RTI International and the DoD.
Measures The primary variables of interest included overweight and obesity status estimated using BMI. As recommended by CDC and the National Heart Lung and Blood Institute,10,19 healthy weight was defined as a BMI of 18.5–24.9, overweight as 25.0–29.9, and obese as Z30.0. Height and weight were self-reported. Variables examined for association with overweight and obesity included age; gender; branch of Service; race/ethnicity; pay grade (rank); education; and marital status. Marital status was dichotomized as married (or “living as married”) and non-married (separated, divorced, widowed, and single, never-married). The multivariate model also examined the associations of exercise, depression, heavy alcohol use, and heavy smoking with overweight and obesity. For this model, a dichotomous pay grade variable was created for enlisted personnel and officers. Exercise was defined as moderate (Z3 days per week for 20 minutes or longer) or vigorous (Z5 days per week for 30 minutes or longer). Heavy drinking was defined as five or more drinks (four or more for women) on the same occasion at least once a week in the past 30 days consistent with definitions used in civilian national surveys.20,21 Heavy smoking was defined as smoking one or more packs per day in the past 30 days. Finally, need for further depression evaluation was assessed using the Burnam screen.22,23 Personnel were defined as needing further evaluation if they felt sad, blue, or depressed for 2 weeks or more in the past 12 months or reported Z2 years in their lifetime of feeling depressed and felt depressed “much of the time” in the past 12 months and felt depressed on Z1 days in the past week.
Methods Data Source, Sampling, and Data Collection Data were drawn from the 1995,13 1998,14 2002,15 2005,16 and 200817 DoD Surveys of Health-Related Behaviors Among AD Military Personnel (HRB). Another survey, conducted in 2011,18 was not included because of incompatible variables and unknown effects of changes in survey methodology. The sampling design and data collection methods were similar for the 1995–2008 population-based surveys and included personnel from the Army, Navy, Marine Corps, and Air Force. The eligible population for each survey included all AD personnel at the time of data collection except recruits, academy cadets, and personnel who were absent without leave, incarcerated, or undergoing a permanent change of station. First-stage sampling involved random selection of installations (and Navy ships) within each of the four services, both within and outside of the continental U.S. Second-stage sampling consisted of randomly selecting personnel at these installations without replacement (except for
Statistical Analysis Analyses were conducted in 2013 using SUDAAN, version 10, to account for the complex survey design of the HRB data. Analyses also used HRB sampling weights to generate results to represent the entire AD military population and account for survey nonresponse. Descriptive statistics were calculated to summarize demographic and weight-related characteristics of AD personnel for each of the survey years from 1995 to 2008. A generalized multinomial logit model examined the associations of sociodemographic characteristics, exercise, depression, and other psychosocial characteristics (heavy cigarette smoking and heavy alcohol use) with overweight and obesity outcomes.
Results The demographic distributions of AD personnel across survey years are presented in Table 1. The military www.ajpmonline.org
3
Reyes-Guzman et al / Am J Prev Med 2014;](]):]]]–]]]
Table 1. Estimated sociodemographics of active duty military personnel, 1995–2008, % (SE) unless otherwise noted Characteristic
1995
1998
2002
2005
2008
N¼16,193
N¼17,264
N¼12,756
N¼16,146
N¼24,690
N¼1,325,396
N¼1,113,710
N¼1,125,083
N¼1,011,852
N¼1,311,581
69.6
59.0
55.6
51.8
70.1
r25
43.8 (1.4)
38.6 (1.3)
46.7 (1.9)
46.7 (1.9)
46.9 (2.0)
26–35
36.2 (0.7)
38.1 (0.7)
31.8 (0.7)
32.8 (1.0)
31.9 (0.8)
36–45
18.1 (0.8)
20.8 (0.8)
18.9 (1.3)
17.8 (1.0)
18.1 (1.1)
Z46
2.0 (0.2)
2.5 (0.2)
2.6 (0.4)
2.7 (0.3)
3.1 (0.4)
Army
31.9 (1.7)
34.0 (1.6)
33.8 (2.2)
31.8 (5.0)
38.7 (4.2)
Navy
28.8 (1.8)
25.8 (1.7)
25.4 (2.1)
26.8 (3.5)
23.9 (4.1)
Marine Corps
11.0 (0.7)
12.2 (1.1)
13.5 (1.9)
12.7 (2.2)
13.4 (2.9)
Air Force
28.4 (1.3)
28.0 (1.3)
27.4 (2.7)
28.7 (3.0)
24.0 (3.6)
Male
87.6 (0.9)
86.3 (0.7)
83.1 (0.8)
85.2 (0.7)
85.7 (0.9)
Female
12.4 (0.9)
13.7 (0.7)
16.9 (0.8)
14.8 (0.7)
14.3 (0.9)
NH White
67.7 (1.1)
66.0 (0.9)
67.3 (1.3)
64.4 (1.3)
64.0 (1.2)
NH Black/AA
17.2 (0.8)
18.0 (0.8)
20.7 (1.4)
17.6 (1.0)
16.7 (0.8)
Hispanic
8.5 (0.4)
8.8 (0.5)
7.1 (0.4)
8.8 (0.5)
10.4 (0.5)
Other
6.6 (0.4)
7.3 (0.4)
5.0 (0.5)
9.2 (0.6)
8.9 (0.5)
E1–E3
21.8 (1.0)
18.9 (0.9)
22.0 (1.6)
24.0 (1.8)
21.0 (1.5)
E4–E6
52.2 (1.4)
52.5 (1.2)
51.9 (1.0)
49.6 (1.8)
51.7 (2.4)
E7–E9
10.4 (0.5)
10.8 (0.5)
10.8 (0.8)
9.7 (0.8)
10.2 (0.6)
W1–W5
1.0 (0.2)
1.2 (0.2)
1.2 (0.2)
1.0 (0.1)
1.4 (0.7)
O1–O3
8.7 (0.8)
9.5 (0.8)
8.3 (0.5)
9.4 (1.1)
9.3 (0.7)
O4–O10
5.9 (0.8)
7.2 (0.7)
5.8 (1.1)
6.3 (0.8)
6.4 (0.8)
Married
63.9 (0.9)
62.5 (0.8)
58.6 (1.1)
59.2 (1.3)
59.4 (1.1)
Unmarried
36.1 (0.9)
37.5 (0.8)
41.4 (1.1)
40.8 (1.3)
40.6 (1.1)
Unweighted total Weighted total a
Response rate (%) Age (years)
Branch of service
Gender
Race/ethnicityb
Pay grade
c
Marital status
a
Response rate is defined as the rate that usable questionnaires from eligible personnel (see exclusion criteria in Methods section) were obtained for both phases of data collection. Other category includes Asian, Native Hawaiian, Pacific Islander, American Indian/Alaska Native, and “other” responses; these were collapsed because of small sample sizes. For 1998, 2.18% of responses for race/ethnicity were missing (not displayed). c E1–E3¼junior enlisted personnel; E4–E6¼mid-level enlisted personnel; E7–E9¼senior enlisted personnel; W1–W5¼warrant officers; O1–O3¼junior officers; O4–O10¼senior officers. AA, African American; NH, Non-Hispanic. b
] 2014
4
Reyes-Guzman et al / Am J Prev Med 2014;](]):]]]–]]]
population was predominantly male, white, relatively young, married, and concentrated in pay grades E4–E6. During the 13-year span, most characteristics remained relatively stable. There were increases in the percentages of the youngest personnel, personnel in the Army, Hispanic or “other” racial minorities, and unmarried personnel. Figure 1 presents the overall changes in overweight and obesity from 1995 to 2008. The proportion of overweight individuals showed a slight increase of 2.5 percentage points during the 13-year period, but the proportion of obese personnel increased nearly 8 percentage points from 1995 to 2008. The combined overweight and obese categories increased 10 percentage points from 50.6 % to 60.8%. To better understand these changes, Table 2 examined overweight and obesity within demographic and military groupings. Consistent with the findings from Figure 1, trends in overweight for each of the demographic groups were relatively stable, although there were a few exceptions. The most striking increase in overweight occurred among military women, who had a gain of nearly 14 percentage points (20.8% to 34.6%). There were also some differences among pay grade groups. Junior officers (O1–O3) showed an increase of 8 percentage points, followed by junior enlisted personnel and warrant officers. By contrast, the rates of obesity showed striking increases. The groups with the largest gains were warrant officers (13.4 percentage points) and senior enlisted personnel (13.2 percentage points) followed by people aged 36–45 years (12.3 percentage points); non-Hispanic blacks/African Americans (11.3 percentage points); and Hispanics (10.7 percentage points). Married personnel
showed larger increases in the prevalence of obesity (9.2 percentage points) compared to unmarried personnel. All demographic and service groups showed significant increases in the rates from 1995 to 2008, and the majority showed increases of Z5 percentage points. Tests of the trends were also conducted in Table 2. These included linear contrasts across the five survey periods for each demographic variable and pairwise comparisons for each combination of survey pairs (e.g., 1995 versus 1998, 2002, 2005, and 2008) within each demographic variable. Results for all tests for trends were statistically significant at po0.05. Table 3 presents adjusted associations of overweight or obesity and demographic, military, and behavioral factors from the 2008 survey. Multinomial models were created to examine the relationship between overweight and obesity and behavioral predictors while controlling for the demographic and military variables examined in Table 2. The first multinomial comparison examined the associations among demographic, military, and behavioral factors and overweight, compared with servicemembers in the healthy weight group. Results showed that the highest overweight risks were for servicemembers aged Z36 years and for men. In addition, non-Hispanic African Americans and Hispanics had 36%–62% higher overweight prevalence, respectively, than those of other race/ethnicities. There was also a moderate association between overweight and heavy drinking (21% greater risk than lighter drinking or no drinking) and between overweight and heavy smoking (19% lower risk than among lighter smokers and nonsmokers). Enlisted and married personnel both had higher prevalence of overweight compared to their reference groups (40% and 26%, respectively).
70.0 BMI 25.0-29.9
BMI 30+
60.0
Percentage
50.0
5.0
6.2
8.8
13.1
12.7
49.0
48.2
48.1
2002
2005
2008
40.0 30.0 20.0
45.6
48.4
1995
1998
10.0 0.0 Survey Year *BMI is based on self-reports of height and weight; does not differentiate between muscle and fat.
Figure 1. Trends in BMI, 1995–2008. www.ajpmonline.org
] 2014
Table 2. Prevalence of overweight and obese active duty personnel by sociodemographic characteristics, 1995–2008, % (SE) 1995 Characteristic
1998
2002
2005
2008
Overweight
Obese
Overweight
Obese
Overweight
Obese
Overweight
Obese
Overweight
Obese
r25
36.8 (0.7)
3.2 (0.1)
38.7 (0.6)
4.1 (0.2)
41.3 (0.7)
5.4 (0.3)
42.0 (0.8)
9.0 (0.3)
40.7 (0.9)
7.9 (0.2)
26–35
49.8 (0.4)
5.8 (0.2)
51.9 (0.4)
7.3 (0.2)
52.1 (0.5)
10.9 (0.3)
51.0 (0.7)
15.0 (0.2)
53.0 (0.5)
15.3 (0.2)
36–45
57.0 (0.5)
7.5 (0.1)
58.6 (0.5)
8.4 (0.2)
60.2 (0.7)
13.0 (0.2)
57.4 (0.6)
19.7 (0.3)
56.1 (0.6)
19.8 (0.3)
Z46
57.5 (0.1)
5.1 (0.1)
58.8 (0.1)
4.8 (0.1)
60.9 (0.3)
10.9 (0.1)
55.8 (0.2)
14.1 (0.1)
59.4 (0.2)
14.5 (0.1)
Army
44.9 (0.7)
4.0 (0.2)
47.7 (0.8)
5.4 (0.2)
48.3 (1.0)
9.3 (0.3)
49.7 (2.2)
10.2 (0.6)
48.7 (1.8)
13.1 (0.6)
Navy
46.0 (0.9)
7.7 (0.2)
50.5 (0.8)
10.3 (0.3)
49.8 (0.9)
11.6 (0.4)
46.8 (1.7)
18.5 (0.7)
48.9 (1.5)
14.4 (0.4)
Marine Corps
47.0 (0.3)
1.9 (0.1)
47.9 (0.6)
2.3 (0.1)
46.9 (0.8)
3.7 (0.1)
48.2 (1.1)
7.1 (0.1)
49.7 (1.4)
6.2 (0.2)
Air Force
45.6 (0.7)
4.4 (0.1)
47.7 (0.7)
5.2 (0.1)
50.1 (1.3)
8.3 (0.2)
47.8 (1.5)
13.7 (0.5)
45.6 (1.2)
13.9 (0.4)
49.1 (0.6)
5.4 (0.3)
52.0 (0.5)
7.0 (0.3)
52.9 (0.6)
9.8 (0.4)
50.6 (0.6)
13.9 (0.5)
50.4 (0.6)
13.8 (0.4)
20.8 (0.2)
1.4 (0.1)
25.5 (0.2)
1.4 (0.1)
29.5 (0.3)
3.9 (0.1)
33.9 (0.3)
8.4 (0.1)
34.6 (0.4)
6.7 (0.1)
NH White
45.5 (0.7)
4.3 (0.2)
48.6 (0.6)
5.4 (0.2)
49.8 (0.8)
7.9 (0.3)
48.6 (1.0)
11.4 (0.4)
47.9 (0.7)
11.1 (0.3)
NH Black/AA
46.4 (0.4)
8.0 (0.2)
49.9 (0.4)
9.7 (0.2)
47.6 (0.6)
11.7 (0.2)
46.9 (0.5)
20.8 (0.4)
47.7 (0.5)
19.3 (0.2)
Hispanic
48.9 (0.3)
3.3 (0.1)
49.7 (0.3)
5.5 (0.1)
48.7 (0.3)
11.4 (0.2)
50.8 (0.3)
13.0 (0.1)
52.8 (0.3)
14.0 (0.1)
Other
40.3 (0.2)
6.3 (0.1)
42.6 (0.2)
5.3 (0.1)
43.7 (0.3)
6.4 (0.1)
45.3 (0.3)
10.5 (0.1)
45.6 (0.3)
12.1 (0.1)
E1–E3
34.4 (0.4)
3.0 (0.1)
37.2 (0.5)
3.8 (0.1)
39.8 (0.7)
5.5 (0.2)
40.1 (0.7)
9.9 (0.3)
39.0 (0.5)
6.6 (0.1)
E4–E6
47.5 (0.7)
5.9 (0.3)
48.7 (0.7)
7.4 (0.3)
48.4 (0.6)
9.8 (0.3)
47.9 (1.0)
14.8 (0.3)
48.0 (1.1)
14.4 (0.4)
E7–E9
57.9 (0.3)
7.8 (0.1)
61.4 (0.3)
8.0 (0.1)
60.1 (0.5)
13.9 (0.2)
58.1 (0.4)
19.2 (0.2)
60.5 (0.4)
21.0 (0.1)
W1–W5
55.4 (0.1)
4.1 (0.1)
60.2 (0.1)
5.8 (0.1)
58.5 (0.1)
8.8 (0.1)
61.8 (0.1)
9.8 (0.1)
59.6 (0.4)
17.5 (0.1)
O1–O3
41.6 (0.3)
1.8 (0.1)
47.5 (0.4)
3.7 (0.1)
52.1 (0.3)
5.6 (0.1)
51.9 (0.6)
7.9 (0.2)
50.0 (0.3)
8.0 (0.1)
O4–O10
52.1 (0.4)
3.8 (0.1)
54.9 (0.3)
4.2 (0.1)
59.5 (0.6)
7.1 (0.1)
56.1 (0.5)
9.8 (0.1)
51.8 (0.4)
11.5 (0.1)
Age (years)
Gender Male Female Race/ethnicity
Pay grade
a
Reyes-Guzman et al / Am J Prev Med 2014;](]):]]]–]]]
Branch of service
b
5
(continued on the next page)
6
Note: Only estimates for overweight and obese personnel are reported. Percentages for overweight, obese, and healthy weight groups add to 100%. Linear contrasts over time and pairwise comparisons between survey years were statistically significant at po0.05. a Other category includes Asian, Native Hawaiian, Pacific Islander, American Indian/Alaska Native, and "other" responses; these were collapsed because of small sample sizes. For 1998, 2.28% of responses were missing (not displayed). b E1–E3 ¼ junior enlisted personnel; E4–E6¼midlevel enlisted personnel; E7–E9¼senior enlisted personnel; W1–W5¼warrant officers; O1–O3¼junior officers; O4–O10¼senior officers. AA, African American; NH, Non-Hispanic.
12.7 (0.4) 13.1 (0.6) 49.0 (0.6) 6.2 (0.3) 48.4 (0.5) 5.0 (0.3) 45.6 (0.5) Total
38.6 (0.5) Unmarried
8.8 (0.5)
48.2 (0.7)
48.1 (0.4)
9.0 (0.1) 43.5 (0.5) 9.6 (0.3) 44.2 (0.7) 43.0 (0.6) 4.7 (0.2) 42.0 (0.4)
52.3 (0.5) 6.1 (0.3) 49.5 (0.6) Married
Marital status
2.8 (0.1)
53.1 (0.9) 7.1 (0.2)
Obese Overweight Obese Overweight Characteristic
5.5 (0.2)
15.4 (0.4) 11.1 (0.3)
50.8 (0.9)
51.3 (0.6)
Overweight Obese Overweight Overweight
Obese
2005 2002 1998 1995
Table 2. Prevalence of overweight and obese active duty personnel by sociodemographic characteristics, 1995–2008, % (SE) (continued)
2008
Obese
15.3 (0.3)
Reyes-Guzman et al / Am J Prev Med 2014;](]):]]]–]]]
The second multinomial comparison examined the associations among the same demographic, military, and behavioral factors and obesity compared to servicemembers of healthy weight. Similar to the overweight analysis, the highest risk of obesity was also found in the oldest two age groups and for men (ORsZ4). Other strong predictors of obesity were being aged 26–35 years, being in the Army, Navy, or Air Force, being non-Hispanic African American, married, and enlisted (ORsZ2). Individuals with a need for further depression evaluation were more likely to be obese compared to those not depressed (OR¼1.36, 95% CI¼1.18, 1.55), and heavy drinkers were more likely to be obese than non-heavy drinkers (OR¼1.24, 95% CI¼1.04, 1.46). Exercise was not a statistically significant predictor of either overweight or obesity.
Discussion This study examined changes in overweight and obesity among AD military personnel from 1995 to 2008. Findings showed an overall increasing trend for combined overweight and obesity of 10 percentage points from 50.6% to 60.8%, primarily driven by the nearly 8 percentage point rise in obesity. Of note, the increases of combined overweight and obesity leveled off between 2005 and 2008, suggesting a possible change in the trend line. These findings also mirror civilian trends showing a deceleration in the late 2000s of previously increasing rates.1 In the current study, the pattern in overall trends was found across most examined subgroups. Although it is possible that awareness of the detrimental effects of obesity and military-specific initiatives to curb the growing epidemic may have slowed the increase in rates, the last period of no change may reflect historic trends of steep increases followed by periods of relative stability.24 This study found several key correlates of overweight and obesity, with age and gender being the strongest predictors, consistent with findings from other studies.25 Of interest, the 36- to 45-year age group had equal or higher overweight and obesity rates than people aged 46 years or older. This could be due to a reduction in lean muscle mass associated with aging and not by a reduction in adipose tissue.26 These findings, however, support the notion that aging servicemembers may experience a decrease in daily exercise as they transition to more senior phases of their careers that typically involve more sedentary time spent behind a desk rather than engaged in physical combat–readiness activities. Both heavy drinkers and servicemembers who screened positive for depression were more likely to be obese than their counterparts.27–32 These findings suggest that servicemembers identified as heavy drinkers or depressed www.ajpmonline.org
7
Reyes-Guzman et al / Am J Prev Med 2014;](]):]]]–]]]
Table 3. Associations between sociodemographic and behavioral risk factors for overweight or obese status, 2008, AOR (95% CI)a Overweight vs healthy
Group
Obese vs healthy
Age (years) 1.00
Yes
26–35
2.13 (1.94, 2.34)
2.88 (2.58, 3.20)
36–45
3.03 (2.66, 3.45)
4.82 (4.15, 5.61)
Z46
3.40 (2.86, 4.03)
4.16 (3.37, 5.14)
Army
1.03 (0.93, 1.14)
2.00 (1.62, 2.47)
Navy
1.11 (1.00, 1.25)
2.35 (1.95, 2.82)
Air Force
0.96 (0.88, 1.04)
2.30 (1.95, 2.72)
Branch of service
Marine Corps
1.00
Gender Male
2.36 (2.14, 2.61)
Female
1.00
NH White
1.11 (0.98, 1.25)
1.01 (0.82, 1.24)
NH Black/AA
1.36 (1.17, 1.59)
2.09 (1.67, 2.61)
Hispanic
1.62 (1.39, 1.88)
1.76 (1.41, 2.21)
Other
1.00
Marital status Married
1.26 (1.16, 1.36)
Unmarried
1.67 (1.52, 1.84) 1.00
Pay grade Enlisted
1.40 (1.23, 1.60)
Officers
2.37 (2.02, 2.77) 1.00
b
Yes
1.08 (0.97, 1.20)
No
0.98 (0.84, 1.14) 1.00
c
Heavy drinker Yes
1.21 (1.10, 1.33)
No
1.23 (1.04, 1.46) 1.00
Heavy smokerd
No
0.81 (0.69, 0.95)
0.88 (0.70, 1.10) 1.00 (continued)
] 2014
Overweight vs healthy
Obese vs healthy
1.04 (0.91, 1.18)
1.36 (1.18, 1.55)
No
1.00
Note: Boldface indicates statistical significance. a This multilog model uses a three-level weight category variable as the outcome. The weight categories include obese individuals (BMIZ30); overweight individuals (BMI¼25–29.9); and individuals with a healthy weight (BMI¼18.5–24.9). BMI is defined as weight (kg)/[height (m)]2. The survey uses self-reports of height and weight; however, BMI does not differentiate between muscle and fat. b This variable includes vigorous exercise for Z20 minutes on Z3 days per week, and moderate exercise for Z30 minutes on Z5 days per week. c Heavy drinking was defined as Z5 drinks on the same drinking occasion at least once a week in the past 30 days. d Heavy smoking was defined as smoking Z1 packs per day (415 cigarettes) in the past 30 days. AA, African American; NH, Non-Hispanic.
3.95 (3.38, 4.61)
Race/ethnicity
Yes
Group Depression
r25
Exercise
Table 3. (continued )
should be carefully screened and monitored for obesity. Conversely, the drinking behaviors and levels of depression of overweight or obese servicemembers should be assessed and treated by healthcare providers as well. Providers should consider these associations when recommending a particular weight loss, depression, or substance use intervention and the potential impact of this treatment on other comorbid conditions. Although some literature suggests an intricate relationship between overweight/obesity and alcohol use, where light to moderate drinkers are less likely to be overweight or obese33 than nondrinkers, the present study did not find similar results. It is possible that these discrepancies are applicable to women of a higher socioeconomic status (e.g., wine drinkers) but are not applicable to the U.S. military. Of note, vigorous/moderate exercise was not associated with overweight or obesity consistent with findings by Lindquist and Bray.9 This suggests that factors besides a sedentary lifestyle play an important role in explaining the observed increases. These factors might include genetics, body composition, dietary intake, or fluctuating exercise during the year. The use of BMI as a measure of overweight and obesity has several caveats worth noting. BMI is correlated with body fat percentage34 and thus is a useful surveillance tool at the population level.35 There are times, however, when an individual has an elevated BMI due to increased muscle mass instead of body fat. We hypothesized that muscular servicemembers with high standards of fitness
8
Reyes-Guzman et al / Am J Prev Med 2014;](]):]]]–]]]
and physical readiness may be classified as overweight or obese according to BMI despite having healthy or even low levels of adiposity. A study of servicemembers by Heinrich and colleagues,36 however, showed that categorizing obesity based on measured BMI actually underestimated the prevalence of obesity compared to using body fat testing to classify obesity. Although it is true that the current study used self-reported rather than measured height and weight to calculate BMI, the underestimation may be particularly pronounced because self-reported height and weight tend to underestimate the prevalence of overweight and obesity in nationally representative samples.37 Nonetheless, prior studies have not examined whether categorizing overweight (but not obese) status based on BMI leads to similar patterns of misclassification in military samples. In evaluating these findings, some limitations should be acknowledged. Modest response rates during some of the survey years raise some questions about possible nonresponse bias. A bias would be expected to result in under-reporting of weight such that BMI would be within normal limits. The finding of substantial overweight/ obesity coupled with increases over time argues against a serious problem. A second limitation is that the use of self-reported height and weight data to calculate BMI and classify overweight and obesity may have introduced reporting bias. Prior studies, however, have found reasonable correlations of self-reported and measured height and weight data38 and suggest that self-reports are reasonable to use in epidemiologic studies.39 Also, potential bias in self-reported estimates was likely nondifferential across time and thus less likely to affect trend data. Third, the use of cross-sectional data to study trends also has some limitations in that trends can be influenced by the characteristics of each sample and period effects of public health campaigns, advances in healthcare access and delivery, and various other cultural influences on weight-related behaviors. Fortunately, differences in sample characteristics are likely to be minimal because of the highly homogeneous nature of the military population, and the statistical weighting that was done to ensure that samples represented the broader military population. Despite these limitations, the current study has several notable strengths. It is based on five large probability samples representative of the AD population. The study used sampling weights that were post-stratified to reflect population totals and used anonymous questionnaires collected by civilian rather than military survey teams. The large sample size permits observation of robust trend estimates overall and within subgroups of interest (e.g., women, depressed personnel, heavy drinkers). The identification of specific demographic and behavioral risk characteristics correlated with overweight and
obesity elucidates potential targets for weight-related prevention and treatment efforts. Overall, the 2008 survey met the Healthy People 2010 objective of o15% obesity prevalence (12.8%); however, several subgroups failed to meet this goal. Servicemembers aged 26–45 years, of non-Hispanic black/African American race/ ethnicity, who reported being married, and who were senior enlisted personnel or warrant officers had obesity prevalence estimates 415%. In multivariate models, heavy drinkers and those who screened positive for depression also had elevated obesity risk. These findings suggest potential targets for weight-related prevention and intervention efforts and highlight characteristics associated with overweight and obesity for healthcare providers to use in patient assessment. In addition to clinical significance, these findings suggest the need for future research and policy considerations. Although the DoD has traditionally focused on health promotion and disease prevention among its troops and Military Health System (MHS) beneficiaries, the prevalence of substantial numbers of both overweight and obese servicemembers requires attention. As a complement to the DoD’s efforts, the MHS has obesity intervention programs, counseling for nutrition improvement, and other health education and wellness classes. Future research should obtain a more nuanced understanding of the predisposing, enabling, and maintenance factors of weight problems in AD servicemembers. In addition, evaluation of current DoD and MHS initiatives may help identify helpful characteristics of successful programs. Overweight and obesity in the military are critical issues currently affecting servicemembers’ health, fitness, and quality of life. Development and implementation of programs that foster a healthy, active force may help control rising medical and non-medical costs while ultimately improving the health and well-being of servicemembers. The authors wish to thank Justin Faerber for editorial assistance in preparing the manuscript. The paper was prepared using internal funds provided by RTI International. The views, opinions, and findings contained in this report are those of the authors and should not be construed as an official Department of Defense position, policy, or decision, unless so designated by other official documentation. No financial disclosures were reported by the authors of this paper.
References 1. Flegal KM, Carroll MD, Ogden CL, Curtin LR. Prevalence and trends in obesity among US adults, 1999-2008. JAMA 2010;303(3):235–41.
www.ajpmonline.org
Reyes-Guzman et al / Am J Prev Med 2014;](]):]]]–]]] 2. Allison DB, Fontaine KR, Manson JE, Stevens J, VanItallie TB. Annual deaths attributable to obesity in the United States. JAMA 1999;282(16): 1530–8. 3. Flegal KM, Graubard BI, Williamson DF, Gail MH. Cause-specific excess deaths associated with underweight, overweight, and obesity. JAMA 2007;298(17):2028–37. 4. Cilli M, De Rosa R, Pandolfi C, et al. Quantification of subclinical anxiety and depression in essentially obese patients and normal-weight healthy subjects. Eat Weight Disord 2003;8(4): 319–20. 5. Rivenes AC, Harvey SB, Mykletun A. The relationship between abdominal fat, obesity, and common mental disorders: results from the HUNT study. J Psychosom Res 2009;66(4):269–75. 6. Petry NM, Barry D, Pietrzak RH, Wagner JA. Overweight and obesity are associated with psychiatric disorders: results from the National Epidemiologic Survey on Alcohol and Related Conditions. Psychosom Med 2008;70(3):288–97. 7. Finkelstein EA, Trogdon JG, Cohen JW, Dietz W. Annual medical spending attributable to obesity: payer-and service-specific estimates. Health Aff (Millwood) 2009;28(5):w822–w831. 8. 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(2):120–39. 9. Lindquist CH, Bray RM. Trends in overweight and physical activity among U.S. military personnel, 1995–1998. Prev Med 2001;32(1): 57–65. 10. Bray RM, Rae Olmsted KL, Williams J, Sanchez RP, Hartzell M. Progress toward healthy people 2000 objectives among U.S. military personnel. Prev Med 2006;42(5):390–6. 11. Smith TJ, Marriott BP, Dotson L, et al. Overweight and obesity in military personnel: sociodemographic predictors. Obesity (Silver Spring) 2012;20(7):1534–8. 12. Department of Defense (DoD). DoD Physical Fitness and Body Fat Program. Washington DC: Department of Defense, 2004. 13. Bray RM, Kroutil LA, Wheeless SC, et al. 1995 Department of Defense Survey of Health Related Behaviors Among Military Personnel. Raleigh NC: Research Triangle Institute, 1995. 14. Bray RM, Sanchez RP, Ornstein ML, et al. 1998 Department of Defense Survey of Health Related Behaviors Among Military Personnel. Raleigh NC: Research Triangle Institute, 1999. 15. Bray RM, Hourani L, Rae KL, et al. 2002 Department of Defense Survey of Health Related Behaviors Among Military Personnel. Raleigh NC: Research Triangle Institute, 2003. 16. Bray RM, Hourani LL, Rae Olmsted KL, et al. 2005 Department of Defense Survey of Health Related Behaviors Among Active Duty Military Personnel. Raleigh NC: Research Triangle Insitute, 2006. 17. Bray RM, Pemberton MR, Hourani L, et al. 2008 Department of Defense Survey of Health Related Behaviors Among Active Duty Military Personnel. Raleigh NC: Research Triangle Institute, 2009. 18. Barlas FM, Higgins WB, Pflieger JC, Diecker K. 2011 Health Related Behaviors Survey of Active Duty Military Personnel. Executive summary 2013, June 27, 2013. tricare.mil/tma/dhcape/surveys/coresur veys/surveyhealthrelatedbehaviors/ADS.aspx. 19. CDC. Healthy weight—it’s not a diet, it’s a lifestyle! www.cdc.gov/ healthyweight/. 20. Substance Abuse and Mental Health Services Administration (SAMHSA). Results from the 2010 National Survey on Drug Use and Health: national findings. Rockville MD: Substance Abuse and Mental Health Services Administration, Office of Applied Studies, 2011.
] 2014
9
21. Johnston LD, O’Malley PM, Bachman JG, Schulenberg JE. Monitoring the Future national survey results on drug use, Volume II: 1975-2010. College students and adults ages 19-50. Bethesda MD: National Institute on Drug Abuse, 2011. 22. Burnam MA, Wells KB, Leake B, Landsverk J. Development of a brief screening instrument for detecting depressive disorders. Med Care 1988;26(8):775–89. 23. Rost K, Burnam MA, Smith GR. Development of screeners for depressive disorders and substance disorder history. Med Care 1993;31(3):189–200. 24. Costa D, Steckel R. Long-term trends in health, welfare, and economic growth in the United States. In: Steckel R, Floud R, eds. Health and welfare during industrialization. Chicago IL: University of Chicago Press, 1997:47–90. 25. Kumanyika SK. Special issues regarding obesity in minority populations. Ann Intern Med 1993;119(7 Pt 2):650–4. 26. Gallagher D, Visser M, Sepulveda D, Pierson RN, Harris T, Heymsfield SB. How useful is body mass index for comparison of body fatness across age, sex, and ethnic groups? Am J Epidemiol 1996;143(3):228–39. 27. Must A, Spadano J, Coakley EH, Field AE, Colditz G, Dietz WH. The disease burden associated with overweight and obesity. JAMA 1999;282(16):1523–9. 28. Gregg EW, Cheng YJ, Cadwell BL, et al. Secular trends in cardiovascular disease risk factors according to body mass index in US adults. JAMA 2005;293(15):1868–74. 29. Wilson PW, D'Agostino RB, Sullivan L, Parise H, Kannel WB. Overweight and obesity as determinants of cardiovascular risk: the Framingham experience. Arch Intern Med 2002;162(16):1867–72. 30. Lahti-Koski M, Pietinen P, Heliovaara M, Vartiainen E. Associations of body mass index and obesity with physical activity, food choices, alcohol intake, and smoking in the 1982–1997 FINRISK Studies. Am J Clin Nutr 2002;75(5):809–17. 31. Onyike CU, Crum RM, Lee HB, Lyketsos CG, Eaton WW. Is obesity associated with major depression? Results from the Third National Health and Nutrition Examination Survey. Am J Epidemiol 2003; 158(12):1139–47. 32. Shelton N, Knott C. Association between alcohol calorie intake and overweight and obesity in english adults. Am J Public Health 2014;104(4): 639–1. 33. Wang L, Lee IM, Manson JE, Buring JE, Sesso HD. Alcohol consumption, weight gain, and risk of becoming overweight in middle-aged and older women. Arch Intern Med 2010;170(5):453–61. 34. Flegal KM, Shepherd JA, Looker AC, et al. Comparisons of percentage body fat, body mass index, waist circumference, and waist-stature ratio in adults. Am J Clin Nutr 2009;89(2):500–8. 35. Bouchard C. How much progress have we made over the last few decades? Int J Obes 2008;32(S7):S2–S7. 36. Heinrich KM, Jitnarin N, Suminski RR, et al. Obesity classification in military personnel: a comparison of body fat, waist circumference, and body mass index measurements. Mil Med 2008;173(1):67–73. 37. Yun S, Zhu B, Black W, Brownson R. A comparison of national estimates of obesity prevalence from the behavioral risk factor surveillance system and the National Health and Nutrition Examination Survey. Int J Obes 2005;30(1):164–70. 38. Kuczmarski MF, Kuczmarski RJ, Najjar M. Effects of age on validity of self-reported height, weight, and body mass index: findings from the Third National Health and Nutrition Examination Survey, 1988–1994. J Am Diet Assoc 2001;101(1):28–34; (quiz 5–6). 39. Spencer EA, Appleby PN, Davey GK, Key TJ. Validity of self-reported height and weight in 4808 EPIC-Oxford participants. Public Health Nutr 2002;5(4):561–5.