Effectiveness of Lifestyle Interventions in an Active Duty Air Force Population

Abstract

Purpose:

To examine the effectiveness of 3 lifestyle intervention programs in an active duty military population.

Design:

Experimental design with stratified random assignment to 1 of 3 intervention groups. Measures were taken at baseline, 3 months and 6 months.

Setting:

A Military Treatment Facility in the western U.S.

Subjects/Intervention:

122 active duty service members were enrolled and randomly assigned to 1 of 3 lifestyle intervention programs: the Diabetes Prevention Program-Group Lifestyle Balance (DPP-GLB), the Better Body Better Life (BBBL) program or the Fitness Improvement Program (FIP).

Measures:

weight, abdominal circumference, lipid and HbA1c levels, physical activity, and well-being as measured by the RAND SF-36 questionnaire.

Analysis:

Statistical analyses were performed to assess changes over time.

Results:

83 participants completed the study (BBBL N = 23, FIP N = 30, DPP-GLB N = 30). The DPP-GLB participants had statistically significant decreases in weight (-3.1 pounds, p = .01) and abdominal circumference (-0.9 inches; p = .01) over time. HbA1c was also significantly lower in this group at 6 months compared to baseline (p = .036). There were no statistically significant changes in weight, abdominal circumference, or HbA1c in the FIP or BBBL groups. No significant changes were observed in lipids in any of the groups.

Conclusion:

Results from this study indicate that the DPP-GLB program may be effective in reducing weight, abdominal circumference, and HbA1c in an active duty U.S. military population.

Keywords

lifestyle intervention military fitness weight loss

Purpose

Military personnel are required to maintain certain levels of physical fitness, functional status and health to accomplish their missions in various environments. Therefore, personnel are assessed periodically for physical fitness and medical conditions that could impede their ability to perform their mission. In the U.S. Air Force (USAF), active duty (AD) members who develop medical conditions leading to long-term inability to deploy may face discharge from service. Unfortunately, the rate of overweight among the AD military mirrors the adult civilian population¹: 71.6% of adults in the U.S. are overweight with a body mass index (BMI) of greater than or equal to 25 kg/m².² Interestingly, military women may be less likely than men to be overweight which differs from civilian populations.³

To date, most studies investigating interventions to address behaviors related to obesity have been conducted on the general populace or among veterans but not in the AD population. Meta-analyses in non-military populations suggest initial improvement in weight, waist circumference, and blood pressure,^4

-10 although it is ambiguous whether weight loss is sustained over time.^6,7,10 Results also demonstrate improvement in physical activity engagement,^5
-7,10 and several studies indicate improvements in hemoglobin A1c (HbA1c), glucose, and/or lipids.^6,8
-10 Finally, one study showed an improvement in Quality of Life (QoL).¹¹ It is unknown whether the above studied lifestyle interventions would produce the same results in a military population.

Studies conducted in military populations have shown mixed results. One study demonstrated that a 1-day in-person class during which participants were instructed on healthy lifestyle behaviors was effective in boosting fitness scores for Airmen who had failed a fitness assessment.¹² However, this study only involved cohorts of Airmen who failed their fitness assessment without a comparison to other Airmen who passed their fitness assessment. Conversely, a review of 8 lifestyle programs in military populations showed that most did not result in significant weight loss at 6 months post-intervention.¹³ However, a recent study comparing counselor initiated and self-paced adaptations of the LookAHEAD Intensive Lifestyle Intervention program in AD personnel showed that both interventions produced significant weight loss and decrease in abdominal circumference.¹⁴

Currently, there are 2 lifestyle intervention programs available at most USAF bases for personnel desiring to lose weight and improve health: the Fitness Improvement Program (FIP) and the Better Body Better Life (BBBL) program. Despite widespread utilization in the military setting, no published evidence was found to support either program. A third intervention, DPP Group Lifestyle Balance (DPP-GLB), adapted from the Diabetes Prevention Program lifestyle intervention¹⁵ has been offered intermittently to personnel at some USAF bases. This program has been shown to be effective for improving weight and physical activity in multiple community settings.^{16

-25} In one study, the DPP-GLB program also demonstrated improvement in Health-related QoL.²⁶ While the DPP-GLB program has been studied extensively in the non-military community, it has not been examined in an exclusively AD population. The purpose of this investigation is to examine the effectiveness of the 3 above interventions in an AD population.

Methods

Ethics Statement

The voluntary, fully informed written consent of the subjects used in this research was obtained as required by 32 CFR 219 and DODI3216.02_AFI40-402, Protection of Human Subjects and Adherence to Ethical Standards in Air Force Supported Research. The work reported herein was performed under United States Air Force Surgeon General approved Clinical Investigation Number FDG 20150017H and was reviewed and conducted under the oversight of the David Grant United States Air Force Medical Center Institutional Review Board.

Study Design

This was a randomized control trial with 3 study arms. It used an experimental design to evaluate the effectiveness of 3 interventions that have been previously implemented in the USAF community (FIP, BBBL, and DPP-GLB). The specific aims were: 1) to determine the effectiveness of the FIP, BBBL, and DPP-GLB lifestyle intervention programs on weight, abdominal circumference, disease risk reduction as measured by HbA1c and lipids, physical activity, and well-being in an at-risk active duty population; and 2) to determine the acceptability of the 3 interventions to AD personnel.

Setting

This setting was an Air Force base in the western U.S. Participants were recruited through flyers posted in medical clinics where AD personnel receive care as well as at the base fitness center. Flyers were also distributed by Unit Fitness Monitors (ie, personnel embedded in units/squadrons who are responsible for tracking fitness testing for AD unit members). Interested participants were asked to contact research coordinators who provided additional study information and assessed eligibility. We estimated that there was a population of approximately 1800 AD individuals who met eligibility criteria to recruit from.

Subjects

Individuals were eligible to participate if they met the following inclusion criteria:

An AD member of any U.S. armed service

Had at least one of the following characteristics:

Abdominal circumference over 35 inches for men or 31.5 inches for women

BMI over 25 kg/m²

Willing to commit to 1 weekly 1-hour class for 12 weeks and two 1-hour classes per month for an additional 3 months

Exclusion criteria were:

Women who were pregnant or breastfeeding

Participants who were within 8 months of a Permanent Change of Station or deployment

Anyone who was restricted from participating in moderate activity equivalent to a brisk walk

Taking glucose-lowering medication

Recently (in the past 6 months) started on a cholesterol lowering medication or had a dosing change in a cholesterol lowering medication (participants who had been on a stable dose of cholesterol lowering medication for greater than 6 months were eligible)

Anyone who for medical reasons could not have a calorie-restricted diet

Randomization

Eligible participants were assigned to 1) FIP, 2) BBBL, or 3) DPP-GLB through a computer-generated stratified randomization plan. Since many studies performed in military populations lack female participants, stratification was performed to ensure that female participants were appropriately represented and equally allocated to each group. Participants and research staff were not blinded to treatment assignment.

Sample Size

A statistical power analysis indicated a total of 23 participants per group would be needed to detect a relatively small change in weight (effect size = .3) over the course of the study using a repeated measures ANOVA with a power of 0.80 and an alpha level of 0.05. In 3 similar studies, attrition ranged from 17%-19% at 6-month follow-up.^27,28 An attrition rate of 20% was thus accounted for in our original sample size calculations. However, the actual attrition in our study was much higher than anticipated (ie, average: 32%; range: 25%-43%). The sample size was recalculated part way through the study to allow for the higher attrition rate, indicating that at least 40 participants per group would be needed to secure 23 completers per group.

Primary and Secondary Outcome Measures

Primary outcomes included weight, abdominal circumference, hemoglobin A1c (HbA1c), lipid profiles, physical activity, and well-being. Weight was measured in pounds on a calibrated scale and abdominal circumference was measured in inches according to USAF fitness assessment standards (the arithmetic mean of 3 tape measurements taken immediately above the iliac crests).²⁹ Fasting blood samples for HbA1c and lipids were drawn and processed through the clinical laboratory at the local Military Treatment Facility (MTF). Physical activity was measured by the Modifiable Activity Questionnaire (MAQ). The MAQ has been used to assess activity in a variety of populations and has been shown to be reliable and valid through comparisons with activity monitors and fitness testing. It assesses hours per week for all activities. (Note that although the MAQ includes both leisure and occupational activity, we used the leisure activity component only, since occupational activity is less likely to be modifiable by the interventions we studied.) Values are weighted by their estimated metabolic cost and expressed as MET-hours per week by multiplying hours per week for each specific activity by the estimated metabolic value of that activity.³⁰ Well-being was measured by the RAND Short Form 36 (SF-36). The SF-36 was developed for the Medical Outcomes Study to measure quality of life and has been found to be a valid and reliable tool for assessing health-related quality of life outcomes in adult patients.³¹ Study coordinators took all measurements and administered all questionnaires. However, blood work was collected at the MTF clinical laboratory, with study coordinators obtaining results through the electronic health record.

All measurements were taken at baseline and again at the 6-month time frame. This allowed for completion of all interventions and a reasonable amount of time to effect change. Some non-blood work measures (i.e., weight and abdominal circumference) were also taken at the 3-month time frame. The secondary outcome of acceptability was assessed through a qualitative content analysis of open-ended responses on a questionnaire administered after the interventions were completed (see Figure 1).

Figure 1.

Intervention Feedback Questionnaire .

Procedure

Once participant eligibility was determined, individuals with continued interest scheduled an appointment to obtain informed consent. As part of the informed consent, research coordinators explained the study and procedures including the randomization process, potential time commitment required, and how measurements would be taken. Following this, research coordinators enrolled the participant into the study and collected baseline data including: demographics, weight, height, and abdominal circumference, as well as activity and well-being questionnaires. Blood work for HbA1c and lipids were also ordered at this time and participants were instructed to have fasting lab work done at their earliest convenience. Research coordinators then assigned the participant to the FIP, BBBL, or DPP-GLB group using a standardized randomization plan generated from www.randomizer.com. Randomization was initially stratified to ensure women represented at least 20% of each group. Once this minimum representation was reached, randomization continued but was no longer stratified by gender. Lastly, the participant was scheduled into either a DPP-GLB class, a BBBL class, or a time frame to take the FIP.

For the DPP-GBL and BBBL groups, participants were scheduled into the next available class. Participants in the FIP group were asked to complete the training within 4 weeks after their baseline visit and to give the research coordinator a printed copy of their FIP certificate of completion. Attendance at both the DPP-GLB and BBBL classes were tracked. Research coordinators followed-up with any participants who missed a class and arranged for them to make up the class through watching the DVD (DPP-GLB) or by scheduling a make-up class (BBBL).

Interventions

Fitness Improvement Program (FIP)

The FIP is a self-directed on-line computer-based training that can be accessed by any AD personnel through the Advanced Distributed Learning System. The content was developed by the USAF with a focus on the AD population and covers physical fitness, nutrition, and overall well-being. It consists of an introduction, 3 core components (nutrition, physical training, and spiritual well-being), and a summary. Knowledge is assessed through short quizzes, and participants are asked to set goals. It takes approximately 90 minutes in total to view all of the content, which can be viewed all at once or in increments. Once the training is complete, there is no follow-up or requirement to repeat the training. Participants are responsible for using the information for their own self-directed program.

Better Body Better Life (BBBL)

The BBBL, also developed by the USAF, consists of 5 independent modules that are taught in-person by instructors who receive on the job curriculum training. Each 2-hour module is offered once a week and conducted in a classroom with 15-20 attendees. Modules provide information on nutrition, physical activity and overall well-being; instructors discuss the content with the group and answer specific questions. Individuals can attend the modules in any order, but are required to complete a preliminary survey and 3-day food record prior to attending their first class. Since participants can take the classes in any order, class attendees are not necessarily the same throughout the program.

Diabetes Prevention Program-Group Lifestyle Balance (DPP-GLB)

The DPP-GLB is a direct adaptation of the Diabetes Prevention Program lifestyle intervention and was designed for individuals at risk for diabetes.¹⁵ The face-to-face group sessions are led by a trained facilitator (i.e. “GLB lifestyle coach”) and are held as follows: 1 one-hour class weekly for 12 weeks followed by 2 one-hour classes per month for 3 months followed by 1 one-hour class per month for 6 months (www.diabetesprevention.pitt.edu). The facilitator weighs and checks-in with each participant separately at the beginning of each class. As with the other 2 interventions, content includes material on nutrition, physical fitness, and overall well-being; facilitators guide discussion and are available to answer questions. Participants are also encouraged to set goals. For study purposes, data were only collected for 6 months (i.e. after 17 one-hour sessions were completed). However, participants were encouraged to attend the remainder of the program even after data collection was complete.

Analyses

STATA/SE 14.2 (College Station, TX) was used for all analyses; alpha level was set at 0.05. Appropriate descriptive statistics (eg, mean, median, standard deviation, range) were chosen to summarize central tendency and dispersion. Pairwise deletion was used when missing data were present. Parametric techniques were employed if the relevant assumptions (i.e. normality, sphericity) were met; otherwise, nonparametric methods were used. Specific analyses for each measure are described in the Results section.

Lastly, a conventional content analysis was performed on the text from the acceptability questionnaire responses. Responses were entered into a spreadsheet, coded, and analyzed for trends in each group by 3 research team members with experience in qualitative methods. Each team member coded content separately initially and then through dialogue as a group, came to consensus on coded items, and named themes for each grouping.

Results

Participants were recruited over the course of 2 ½ years. Although 180 individuals expressed interest in participating in the study, only 122 met all eligibility criteria and were enrolled. The most common reason for ineligibility of pre-screened participants was not being overweight; 43% did not meet the BMI or waist circumference inclusion criteria. In addition, 15% couldn’t participate for the entire length of the study due to upcoming permanent change of duty station (PCS) or deployment. Other, less common, causes of ineligibility included participating in other weight loss programs, conflicting work schedules, and pending separation from AD service.

Of the 122 participants enrolled in the study, 40 were randomized to the BBBL group, 42 to the FIP group, and 40 to the DPP-GLB group. The stratification by gender was successful in creating more equal numbers of males and females than the previously anticipated 20% female and 80% male participation. Percentage of females in each group ranged from 39% in the BBBL group to 60% in the DPP-GLB group.

Of the 122 enrolled participants, 83 completed the study (see Table 1). There was no statistically significant association between gender (chi-square = 1.01; p = .31) nor race/ethnicity (chi-square = 3.46; p = .46) on completion status (i.e., 83 completers vs. 39 non-completers). We used the Wilcoxon-Mann-Whitney test to examine the effect of age on completion status since the age data were skewed (i.e., not normally distributed). The results showed a statistically significant effect of age on completion status; on average, participants who completed the study were older (median age = 34 years) compared to participants who did not complete the study (median age = 29 years; z = -2.16; p = .03). The variables military rank and age are highly related (i.e., high collinearity); thus, we chose not to separately explore the relationship between rank and completion status.

Table 1.

Baseline Demographic Characteristics of All Enrolled Subjects.

	BBBL		FIP		DPP-GLB		Total
	N	Mean (SD)	N	Mean (SD)	N	Mean (SD)	N	Mean (SD)
Age (years)	40	30.58 (5.68)	42	32.52 (7.81)	40	33.7 (8.4)	122	32.27 (7.45)

Gender	N	%	N	%	N	%	N	%
Female	14	35%	17	40.48%	24	60%	55	45.08%
Male	26	65%	25	59.52%	16	40%	67	54.92%

Race/Ethnicity	N	%	N	%	N	%	N	%
White	18	45%	23	54.76%	19	47.5%	60	49.18%
African American	8	20%	8	19.05%	10	25%	26	21.31%
Hispanic/Latino	6	15%	4	9.52%	8	20%	18	14.75%
Asian	5	12.5%	4	9.52%	2	5%	11	9.02%
Multiple/Other	3	7.5%	3	7.14%	1	2.5%	7	5.74%

Military rank	N	%	N	%	N	%	N	%
E1-E4	13	32.5%	10	23.81%	10	25%	33	27.05%
E5-E6	20	50%	18	42.86%	14	35%	52	42.62%
E7-E9	2	5%	8	19.05%	6	15%	16	13.11%
O1-O3	3	7.5%	4	9.52%	6	15%	13	10.66%
O4+	2	5%	2	4.76%	4	10%	8	6.56%

Note: there were no statistically significant differences (p > .05) among the 3 groups (BBBL, FIP, DPP-GLB) in demographic characteristics.

It was confirmed that there were no statistically significant differences (p > .05) among groups (BBBL, FIP, DPP-GLB) in baseline characteristics (i.e., age, gender, race/ethnicity, military weight, abdominal circumference, HbA1c, fasting total cholesterol, LDL, HDL, and triglycerides).

Weight and Abdominal Circumference

After checking that appropriate assumptions were met, repeated measures ANOVAs were performed assessing weight (lbs) and abdominal circumference (inches) as a function of time (baseline, 3 months, 6 months) in each intervention group (BBBL, FIP, DPP-GLB). In the DPP-GLB group, there was a statistically significant difference in weight (F_{2, 57} = 4.68; p = .01; see Figure 2) and abdominal circumference (F_{2, 57} = 4.35; p = .02; see Figure 3) over time.

In the BBBL group, there was no statistically significant difference in weight (F_{2, 44} = 0.99; p = .38) nor abdominal circumference (F_{2, 44} = 2.47; p = .10) over time. In the FIP group, there was no statistically significant difference in weight (F_{2, 58} = 0.88; p = .42) nor abdominal circumference (F_{2, 58} = 3.11; p = .052) over time, although the latter comparison fell just short of statistical significance.

Figure 2.

Average weight (pounds), shown as arithmetic mean ± standard deviation

Figure 3.

Average abdominal circumference (inches), shown as arithmetic mean ± standard deviation

Further analyses of the DPP-GLB group data after the statistically significant omnibus result showed that at 6 months, DPP-GLB participants lost an average of 3.1 pounds and 0.9 inches of abdominal circumference. Pairwise comparisons for weight in the DPP-GLB group showed statistically significant differences between baseline and 3 months (p = .02) and baseline and 6 months (p = .03). Pairwise comparisons for abdominal circumference in the DPP-GLB group showed a statistically significant difference between baseline and 6 months (p = .01); the comparison between baseline and 3 months fell just short of significance (p = .054). However, the results of these additional pairwise comparisons should be interpreted cautiously, since when the Bonferroni correction procedure (a conservative method used to adjust for multiple comparisons) was applied, only the difference in abdominal circumference between baseline and 6 months remained statistically significant.

HbA1c

The Skillings-Mack test (a nonparametric version of the repeated measures ANOVA which is useful when there are many ties and/or missing data)^32,33 was used to assess HbA1c as a function of time (baseline vs. 6 months) in each intervention group. HbA1c was significantly lower (i.e., an average of .08 units) in the DPP-GLB group at 6 months compared to baseline (p = .036). There was no statistically significant difference in HbA1c over time in the BBBL group (p = .30) nor the FIP group (p = 1.00; see Table 2).

Table 2.

Summary of Lipid and HbA1c levels; Data Are Reported as Mean (Standard Deviation).

	BBBL (n = 23)			FIP (n = 30)			DPP-GLB (n = 30)
	Baseline Mean ( SD )	6 months Mean ( SD )	p value	Baseline Mean ( SD )	6 months Mean ( SD )	p value	Baseline Mean ( SD )	6 months Mean ( SD )	p value
Total Cholesterol	186.13 (36.5)	180.52 (36.15)	.30	171.72 (33.29)	170.67 (31.73)	.33	186.93 (35.37)	188.17 (34.59)	.86
LDL	105.36 (32.30)	103.01 (34.84)	1.00	97 (32.57)	94.95 (29.55)	.85	109.63 (33.06)	108.75 (28.89)	.47
HDL	57.35 (19.35)	53.83 (19.3)	.30	55.86 (13.62)	57.00 (14.44)	.58	54.27 (16.93)	52.17 (16.41)	.20
Triglycerides	116.96 (68.52)	118.83 (70.52)	.53	94.41 (45.16)	93.2 (65.24)	.58	115.43 (65.63)	146.47 (125.62)	.14
HbA1c	5.28 (0.24)	5.22 (0.45)	.30	5.24 (0.47)	5.26 (0.40)	1.0	5.33 (.36)	5.26 (.43)	.03*

* = p value < .05.

Note: one subject in the FIP group did not have lipid nor HbA1c data at baseline.

Lipids

The Skillings-Mack test showed no statistically significant effect of time (baseline vs. 6 months) on total cholesterol, LDL, HDL, and triglycerides in any of the groups (see Table 2).

Physical Activity

The Skillings-Mack test showed a statistically significant effect of time (baseline vs. 6 months) on the number of metabolic hours per week of leisure activity as measured by the MAQ in the BBBL group (p = .03) and the DPP-GLB group (p = .01). On average, participants in both the BBBL group and the DPP-GLB groups showed a statistically significant increase in metabolic hours of leisure activity at 6 months relative to baseline. There was no statistically significant difference in metabolic hours of leisure activity over time in the FIP group (p = .72) (see Table 3).

Table 3.

Leisure Activity/week in Each Intervention Group; Data are Calculated in Metabolic Hours From the Modifiable Activity Questionnaire and Are Reported as Median (Minimum-Maximum).

Group	Baseline	6 months	p value
BBBL (n = 23)	19.8 (2.3-82.5)	30.2 (8.8-59.4)	.03*
FIP (n = 30)	30.7 (1.7-133.0)	26.9 (4.1-209.3)	.72
DPP-GLB (n = 30)	24.0 (0.5-370.4)	36.2 (8.1-91.0)	.01*

* = p value < .05.

Note: one subject in the BBBL group did not have data at 6 months.

RAND SF-36

The Skillings-Mack test was used to individually analyze the 8 health concepts (ie, categories) of the SF-36: physical functioning, bodily pain, role limitations due to physical health problems, role limitations due to personal or emotional problems, emotional well-being, social functioning, energy/fatigue, and general health perceptions over time (baseline, 3 months, 6 months).

There were statistically significant improvements over time on “bodily pain” in the BBBL group (p = .039) and the FIP group (p = .029) as well as “general health perceptions” in the BBBL group (p < .0001) and the FIP group (p = .037). In the DPP-GLB group, there was no statistically significant effect of time on bodily pain (p = .12) or general health perceptions (p = .14), nor were there statistically significant effects of time on any of the other categories in any of the intervention groups.

Qualitative Analyses

Results from the content analysis performed on feedback questionnaire responses showed some trends in comments for each intervention. The most common response from participants regarding what they liked best for the BBBL and DPP-GLB interventions was the information provided; the most common response for what they liked least was the schedule. Conversely for the FIP intervention, the most common response regarding what they liked best was the schedule, and for what they liked least was that there was no facilitator. This is not surprising given the FIP was an on-line only content that could be viewed at any time, whereas the DPP-GLB and BBBL content was delivered via an in-person class setting by a facilitator.

Four questions contained binary responses (yes/no); results are reported as percentages for each response by intervention. The DPP-GLB group had the highest “Yes” response percentages compared to the other 2 groups for questions asking if the intervention was useful, beneficial and recommended. The BBBL group had the highest percentage of “Yes” responses to the question asking if the intervention was difficult. For the last question regarding what could be improved about the intervention, the most common response from the BBBL group was more flexibility to the class schedule (30%), the most common for the FIP group was to not have the content delivered as a Computer Based Training (CBT) (23%), and for the DPP-GLB it was to increase physical activity content (27%). See Table 4 for a summary of the results.

Table 4.

Summary of Intervention Questionnaire Results.

	Liked Best	Liked Least	Useful	Difficult	Beneficial	Recommend	Improve
BBBL N = 23	Good Info- 10 (43%)	Schedule - 10 (43%)	Yes – 15 (65%)No – 5 (22%)	Yes-13 (57%)No-9 (39%)	Yes-18 (78%)No-4 (17%)	Yes-19 (83%)No-4 (17%)	More flexible schedule- 7 (30%)
FIP N = 30	Schedule- 9 (30%)	No facilitator- 14 (47%)	Yes – 20 (67%)No – 7 (23%)	Yes-3 (10%)No-27 (90%)	Yes-21 (70%)No-9 (30%)	Yes-23 (77%)No-7 (23%)	No response-9 (30%)No computer- based training – 7 (23%)
DPP-GLB N = 30	Good Info- 16 (53%)	Nothing – 8 (27%)Schedule- 6 (20%)	Yes – 28 (93%)Blank – 2 (7%)	Yes – 11 (37%)No – 19 (63%)	Yes – 29 (97%)Blank – 1 (3%)	Yes – 29 (97%)Blank – 1 (3%)	Physical activity – 8 (27%)

Discussion

There is a clear need for AD members to have access to lifestyle interventions that can optimize their health and well-being. However, interventions that have been successful in civilian at risk populations may not have the same success in military populations due to unique circumstances that could inhibit participation. Therefore, it is important to understand which interventions are effective for this specific population prior to implementing new programs. According to results from this study, interventions currently available to Airmen may not be effective in helping overweight AD personnel in losing weight or lowering disease risk. However, the DPP-GLB was effective in reducing weight, abdominal circumference, and HbA1c. This is consistent with similar DPP-GLB studies involving non-AD populations.^{22

-25,34,35} Of note, weight reductions observed with the DPP-GLB group were modest compared to similar studies in civilian populations and may not be clinically significant. Nevertheless, even modest weight reductions are meaningful in an active duty population whose performance evaluations include weight measurements.

None of the interventions demonstrated an effect on lipids. This differs from studies performed in civilian populations where lifestyle interventions have led to favorable changes in lipids.³⁶ This could be because the AD population is generally young and healthy compared with many civilian populations and therefore may not demonstrate a change in lipids related to lifestyle interventions. Lastly, participants indicated that the information from the 3 lifestyle interventions was helpful, but that they preferred to have a facilitator deliver the content as well as more flexible scheduling for classes.

Limitations to this study include that it was conducted at only one Air Force base and may not be representative of populations in other geographic locations. In addition, the study only evaluated changes in measures over the course of 6 months; therefore it is unknown whether weight, abdominal circumference and HbA1c changes are sustained over a longer period of time. Furthermore, we did not study specific characteristics of each program that may have contributed to their effectiveness, such as number of classes attended and total time/attention given to participants (ie, class attendance was tracked but these data were not separately analyzed). These components may be interesting to investigate in future research. Finally, while it is difficult to do long-term studies on AD personnel due to frequent deployments and moves, a longer term study that involves several locations would help determine whether lifestyle interventions such as the DPP-GLB are effective in sustainment of weight loss and lowering disease risk over a longer period of time. Many platforms have recently been offered virtually due to the COVID-19 pandemic, and it would be interesting to assess how this may affect outcomes.

This study provides important information that can help military leadership make informed decisions about which lifestyle interventions to make available to the AD population. Although this study did not find evidence to support effectiveness of the currently available FIP program or BBBL programs on weight, abdominal circumference or HbA1c, it did find that the DPP-GLB program had a positive effective on those measures which could help improve performance on the fitness assessment and decrease risk for disease. Leaders may want to consider making the DPP-GLB program more widely available for AD personnel through access to trained facilitators and classes with flexible scheduling. Live virtual options for the program could also be considered.

So What?

What Is Already Known About This Topic?

Some lifestyle interventions that focus on behaviors related to physical activity and nutrition may positively affect physical health indicators and well-being in some high-risk populations.

What Does This Article Add?

This study fills a knowledge gap regarding the effectiveness of different lifestyle interventions in an active duty military population. The Diabetes Prevention Program-Group Lifestyle Balance intervention was found to be effective for decreasing weight, abdominal circumference, and HbA1c in an at risk active duty population. No effects on weight, abdominal circumference, or HbA1c were found for interventions currently implemented in the U.S. Air Force in the active duty population studied.

What Are the Implications For Health And Research?

Results from this study indicate that the DPP-GLB program may be effective in reducing weight, abdominal circumference, and HbA1c in an active duty Air Force population. Active duty service members who achieve normal measurements in these areas could potentially have improved performance and decreased chronic disease risk. Additional research encompassing a broader active duty population on lifestyle intervention programs targeted toward active duty service members is needed to understand long-term effects of such programs.

Footnotes

Acknowledgments

This study was funded by a grant from the TriService Nursing Research Program (TSNRP) grant number N15-006.

We would like to acknowledge Mr. Arthur Stout and Ms. Kathryn Buthker for their dedicated efforts that were key to the success of this study.

Clinical Trials Registration

This article has been registered with Clinical Trials.gov under identifier NCT02556112.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Disclaimer

The views expressed in this material are those of the authors and do not reflect the official policy or position of the U.S. Government, the Department of Defense, or the Department of the Air Force.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Nicole H. Armitage, PhD

References

Tanofsky-Kraff

Sbrocco

Theim

, et al. Obesity and the US military family. Obesity. 2013;21(11):2205–2220. doi:10.1002/oby.20566

Centers for Disease Control and Prevention. National Center for Health Statistics. Obesity and overweight. Accessed December 28, 2020. https://www.cdc.gov/nchs/fastats/obesity-overweight.htm

Smith

Marriott

Dotson

, et al. Overweight and obesity in military personnel: sociodemographic predictors. Obesity. 2012;20(7):1534–1538. doi:10.1038/oby.2012.25

Borel

Nazare

Baillot

, et al. Cardiometabolic risk improvement in response to a 3-year lifestyle modification program in men: contribution of improved cardiorespiratory fitness vs. weight loss. Am J Physio Endocrinol Metabol. 2017;321(4):E273–E281. doi:10.1152/ajpendo.00278.2016

Burgess

Hassmen

Welvaert

Pumpa

Behavioural treatment strategies improve adherence to lifestyle intervention programmes in adults with obesity: a systematic review and meta-analysis. Clinical Obesity. 2017;7(2):105–114. doi:10.111/cob.12180

Cradock

OLaighin

Finucane

Gainforth

Quinlan

Ginis

Behaviour change techniques targeting both diet and physical activity in type 2 diabetes: a systematic review and meta-analysis. Int J Beh Nutrition Phy Act. 2017;14(1):1–18.

Gilis-Januszewska

Lindström

Tuomilehto

, et al. Sustained diabetes risk reduction after real life and primary health care setting implementation of the diabetes in Europe prevention using lifestyle, physical activity and nutritional intervention (DE-PLAN) project. BMC Public Health. 2017;17(1):1–7. doi:10.1186/s12889-017-4104-3

Johansen

MacDonald

Hansen

, et al. Effect of an intensive lifestyle intervention on glycemic control in patients with type 2 diabetes: a randomized clinical trial. JAMA. 2017;18(7):637–646. doi:10.1001/jama.2017.10169

Sbroma

Pippi

Reginato

, et al. Intensive lifestyle intervention is particularly advantageous in poorly controlled type 2 diabetes. Nutr Metab Cardiovasc Dis. 2017;27(8):688–694. doi:10.1016/j.numecd.2017.06.009

10.

Zhang

Imperatore

Thomas

, et al. Effect of lifestyle interventions on glucose regulation among adults without impaired glucose tolerance or diabetes: a systematic review and meta-analysis. Diabetes Res Clin Pract. 2017;123:149–164. doi:10.1016/j.diabres.2016.11.020

11.

Saboya

Bodanese

Zimmermann

, et al. Lifestyle intervention on metabolic syndrome and its impact on quality of life: a randomized controlled trial. Arq Bras Cardiol. 2017;108(1):60–69. doi:10.5935/abc.20160186

12.

Webber

Nelson

Gildengorin

. Indicators of sequential fitness assessment failures for Travis Air Force base airmen who attend the be well course. Mil Med. 2012;177(3):302–307.

13.

Murray

Aboul-Enein

Bernstein

Kruk

. Selected weight management interventions for military populations in the United States: a narrative report. Nutr Health. 2017;23(2):67–74. doi:10.1177/0260106017704797

14.

Krukowski

Hare

Talcott

, et al. Dissemination of the Look AHEAD intensive lifestyle intervention in the United States military: a randomized controlled trial. Obesity. 2018;26(10):1558–1565. doi:10.1002/oby.22293

15.

Diabetes prevention program research group, Knowler

Barrett-Connor

, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med. 2002;346(6):393–403. doi:10.1056/NEJMoa012512

16.

Seidel

Powell

Zgibor

Siminerio

Piatt

. Translating the diabetes prevention program into an urban medically underserved community: a nonrandomized prospective intervention study. Diabetes Care. 2008;31(4):684–689.

17.

McTigue

Conroy

Bigi

Murphy

McNeil

Weight loss through living well: translating an effective lifestyle intervention into clinical practice. Diabetes Educ. 2009;35(2):199–208.

18.

Kramer

Kriska

Venditti

, et al. Translating the diabetes prevention program: a comprehensive model for prevention training and program delivery. Am J Prev Med. 2009;37(6):505–511.

19.

Kramer

Kriska

Venditti

, et al. A novel approach to diabetes prevention: evaluation of the group lifestyle balance program delivered via DVD. Diabetes Res Clin Pract. 2010;90(3):e60–e63.

20.

Kramer

McWilliams

Chen

Siminerio

. A community-based diabetes prevention program: Evaluation of the group lifestyle balance program delivered by diabetes educators. The Diabetes Educ. 2011;37(5):659–668.

21.

Piatt

Seidel

Chen

Powell

Zgibor

. Two-year results of translating the diabetes prevention program into an urban, underserved community. The Diabetes Educ. 2012;38(6):798–804.

22.

Yank

Xiao

, et al. Translating the diabetes prevention program lifestyle intervention for weight loss into primary care: a randomized trial. Arch Intern Med. 2012:173(2):1–9.

23.

Kramer

Venditti

Semler

Kriska

Miller

Orchard

. Long-term strategies for diabetes prevention: evaluation of the group lifestyle balance post-core sessions focusing on carbohydrate and hunger management. J Diabetes Metab. 2012;2(6):1–6.

24.

Kramer

Molenaar

Arena

, et al. Improving employee health: evaluation of a worksite lifestyle change program to decrease risk factors for diabetes and cardiovascular disease. J Occup Environ Med. 2015;57(3):284–291. PMID:25742535.

25.

Kramer

Vanderwood

Arena

, et al. Evaluation of a diabetes prevention program lifestyle intervention in older adults: a randomized controlled study in three senior/community centers of varying socioeconomic status. The Diabetes Educ. 2018;44(2):118–129.

26.

Eaglehouse

Schafer

Arena

Kramer

Miller

Kriska

. Impact of a community-based lifestyle intervention program on health-related quality of life. Qual Life Res. 2016;25(8):1903–1912. doi:10.1007/s11136-016-1240-7. Epub 2016 Feb 20. PubMed PMID: 26896960.

27.

Yank

Xiao

, et al. Translating the diabetes prevention program lifestyle intervention for weight loss into primary care: a randomized trial. JAMA Int Med. 2013;173(2):1–17. doi:10.1001/2013.jamainternmed.987

28.

Shrestha

Combest

Fonda

Alfonso

Guerrero

. Effect of an accelerometer on body weight and fitness in overweight and obese active duty soldiers. Mil Med. 2013;178(1):82–87. doi:10.7205/MILMED-D-12-00275

29.

Air Force Manual 36-2905. Air force physical fitness program. December 11, 2020. Retrieved December 28, 2020. www.e-Publishing.af.mil

30.

Kriska

. Modifiable activity questionnaire. ACSM Journals. 1997;29(6):S73–S78.

31.

Rand Health. Scoring instructions for the MOS 36-Item short form survey instrument. 2009. Retrieved January 29, 2014. http://www.rand.org/health/surveys_tools/mos/mos_core_36item.html

32.

Chatfield

Mander

. The Skillings-Mack test (Friedman test when there are missing data). Stata J. 2009;9(2):299–305.

33.

Skillings

Mack

. On the use of a Friedman-type statistic in balanced and unbalanced block designs. Technometrics. 1981;3(2):171–177.

34.

Eaglehouse

Venditti

Kramer

, et al. Factors related to lifestyle goal achievement in a diabetes prevention program dissemination study. Transl Behav Med. 2017;7(4):873–880. doi:10.1007/s13142-017-0494-0

35.

Piatt

Powell

Seidel

Zgibor

. Comparative effectiveness of lifestyle intervention efforts in the community: results of the rethingking eating and ACTivity (REACT) study. Diabetes Care. 2013;36(2):202–209. doi:10.2337/dc12-0824

36.

Zhang

Devlin

Smith

, et al. Effect of lifestyle interventions on cardiovascular risk factors among adults without impaired glucose tolerance or diabetes: a systematic review and meta-analysis. PLOS One. 2017;12(5):e0176436. doi:10.1371/journal.ps.0176436