Low dietary intake among women in the US Army and in other militaries is a source of worldwide concern. Even so, very few nutritional surveys have been conducted among females in the military. Specific nutritional deficiencies have been shown in energy, protein, vitamin B6, folic acid, magnesium, zinc, iron, and particularly in calcium where dietary intake of female soldiers did not meet the Military Recommended Dietary Allowances (MRDA) (2,4). Although an increasing number of women are participating in Israel Defense Forces (IDF) combat units, their dietary intake has never been assessed using Food Frequency Questionnaires (FFQ). In addition, no assessment of the male and the female fighters' nutrition status at induction to reflect dietary intake background has ever been made. This study was conducted within the framework of a multidisciplinary prospective research project on the Karakal combat unit.
In Army Basic Training (ABT), a main concern is for adequate energy, protein, calcium, and iron intake due to their direct involvement in energy metabolism and physical performance (3). Inadequate dietary intake can impair immune system function, cause prolonged recovery from injuries and illnesses, and have a negative impact on physical performance (13). Lack of calcium in adolescence can also increase the risk for loss of bone density and development of fractures at a later stage (4). The Military Dietary Reference Intake (MDRI) has been adapted from the MRDA of the Food and Nutrition Board committee and is a quantitative estimate of the recommended dietary intake for healthy army populations (14). These recommendations are used for establishing nutritional programs for the various military units tailored to their specific demands, with particular emphasis on the health maintenance of the young soldiers. As most army recruits in Israel are aged 18-19 yr and therefore in their late adolescent period, there is a need for special nutritional considerations, particularly among those fighters who undergo different and difficult physical demands.
King (2) summarized six studies evaluating a total of 233 female soldiers. It was found that they consumed less than 70% of the MRDA for different nutrients. The reasons mentioned for the low intake of these women were the field and environmental conditions, the stress of training, the type of military rations served, and the soldiers' nutritional habits. Data are contradictory regarding total energy intake, with Klicka and King (4) noting sufficient energy in West Point cadets of both genders, whereas Rock et al. (8) noting only 69-74% of the nutrition standard for operational rations in artillery soldiers. Besides energy, there is a concern regarding protein intake due to its nutritional role in regulating muscle mass and strength, its response to injury and infections, and its possible impact on cognitive performance (7). The protein requirements by the MDRI are established at 0.8 to 1.5 g·kg−1 body weight·d−1 with a range of 50-93 g·d−1 for women and 63-119 g·d−1 for men (14). In 1996, Marriott found that mean protein intake in military women decreased from 96 to 82 g·d−1 during ABT, but as operational rations are high in protein content, this is still within the limits of the MRDA (5). During adolescence, calcium intake is of great importance in developing peak bone mass (4,5), and an adequate consumption has been implicated in preventing stress fractures although data are controversial (1). Almost 90% of female adolescents do not comply with the calcium daily dietary reference intakes. One report from the United States found an intake of less than 70% of the MRDA among female soldiers (2). In another report, the investigator showed a mean calcium intake of 787 mg·d−1 among women aged 18-30 yr (5). The nutritional importance of iron is well known. In addition to the relationship between hemoglobin levels and physical fitness (6), iron also affects a variety of functions including cognitive performance through oxidative metabolism (7). Although the MDRI takes the average activity of soldiers into consideration, variations in actual activity levels, nutritional supplies, and actual utilization of the supplies may result in nutritional deficiencies, particularly during combat activities. As several studies have shown that energy intake in some civilian women can be severely insufficient, the excessive physical activity demands imposed on female soldiers and the need to meet nutritional requirements can exacerbate the deficiency in individuals (2).
The Karakal unit is a light infantry regiment comprised of both female and male fighters who train together. As a light infantry regiment, its activity level is categorized as light to moderate according to the IDF's physical activity scale. As the IDF has not yet established an MDRI for its own purposes, this study uses the established US MDRI as a first assessment for dietary intake references. Accordingly, the energy requirement in the Karakal unit is 2200-2300 kcal·d−1 for females and 3000-3250 kcal·d−1 for males (14). The purpose of this study was to assess and evaluate dietary intake among male and female recruit fighters and to compare their intake to female nonfighters. In addition, we assessed the dietary intake of female fighter recruits during integrated ABT, with the expectation that females were able to complete the same physical activity demands as males.
MATERIALS AND METHODS
The research group included all Karakal females (KF) enlisted to serve in one platoon of the Karakal regiment. The Karakal males (KM) served as a comparison group, and a company of female recruits training as medics served as a control group (control females [CF]). The CF group was chosen because it has much lighter physical training demands during the 4-month ABT. This is because training is combined with the medic course. Some of the recruits enlisted in the Karakal had volunteered for a year's national service (NS) before their regular army service. This year was spent doing community work (with no excessive physical activity) and involved living away from home. All recruits enlisted in the Karakal group were screened using standard IDF medical profiling. The study was voluntary and all participants signed informed consents according to the international review board approval of both the IDF and the US Army.
The weight of each subject was measured after bladder voiding on an electronic scale with an accuracy of ±100 g. Height was assessed to the nearest 1 cm. Skinfold thickness was measured with a Skyndex® digital caliper (Caldwell; Justiss & Co., Inc., Fayetteville, AR) at the biceps, the triceps, the subscapular, and the suprailiac loci. Each measurement was repeated three times on the right side of the body, and the mean of the three measurements was used. All skinfold measurements were carried out by the same experienced laboratory technician. Body weight and skinfold thickness were reassessed after 8 and 16 wk of ABT by the same technician using the same equipment. Body fat percentage was calculated using Siri's (11) equation. Body fat weight and lean body weight were derived by simple multiplication.
Food frequency questionnaire
Food intake was assessed using the FFQ, developed and validated for the Israeli population by The S. Daniel Abraham International Center for Health and Nutrition at the Ben-Gurion University of the Negev (9,10). The FFQ is a long-term dietary assessment tool. Its time frame may vary, most often referring to the previous year, but it can also refer to the previous month or previous 6 months. It is designed to measure the participant's typical food intake. We used a time frame from the previous 2 months up to a year before date of recruitment to incorporate all variations in the backgrounds of the subjects.
The FFQ includes 126 food items in nine food groups and can be analyzed for nutrients and food group intake. The food groups include 1) eggs, milk, and milk products; 2) fats (including sauces); 3) chicken, meat, and fish; 4) bread and baked products; 5) starches and legumes; 6) fruit; 7) vegetables; 8) snacks and cookies; and 9) beverages. Each participant was instructed to relate to a defined period. The questionnaire was divided into columns of time frequency (day/week/month or nonintake), and the subject filled in the appropriate consumption frequency. For the assessment of portion sizes, pictures of typical foods or dishes were displayed.
The recruits completed the questionnaires with verbal clarification of a dietitian. For new immigrants and for subjects not fluent in Hebrew, the FFQ forms were interpreted orally by the study dietitian. The recruits filled in the FFQ at three time points: at induction, referring uniquely to their preinduction status; after 8 wk of ABT; and after 16 wk, at the end of ABT (post-ABT) whereby data at 8 and 16 wk were compared and found to be consistent. The FFQ at post-ABT refers to the period during ABT. As one of the purposes in this study was to assess the nutrition status of the soldiers preinduction, not only those completing the full ABT were analyzed. The different number (n) of subjects that is presented during ABT is due to dropouts and nutrition analyses outliers.
Army kitchens provided food according to one of several "food tables"-detailed food serving lists for each meal over a 2-wk cycle. The tables varied according to the type of company being served. The Karakal recruits were served food according to standard army food table 5, with precalculated 3620 kcal·d−1 daily nutrient content, 884 mg·d−1 calcium, and 25 mg·d−1 iron. The medics were served food according to army food table 2, with precalculated 2220 kcal·d−1 daily nutrient content, 525 mg·d−1 calcium, and 16.3 mg·d−1 iron. However, it appears that the above tables may be insufficient in calcium requirements and probably other nutrients as well.
Analysis of the nutritional data was performed using the Statistical Package for the Social Sciences version 12 (SPSS Inc., Chicago, IL). Descriptive statistics included mean, standard deviation, standard error, and range. Energy intake was checked for outliers of the 5th and 95th percentiles. At each data collection point, outliers (n = 12) were assessed for consistency between energy intake and BMI measurements. If the results were inconsistent, the subject was excluded from that time point onward. MDRI was used to assess the adequacy of the soldiers' dietary intakes by calculating the mean nutrient intakes as a percentage of the MDRI. For comparing groups and preinduction versus post-ABT, we used repeated-measures ANOVA, general linear model, and SAS, testing the effect of group and time and the interaction between group and time. Significance was set at P < 0.05.
During ABT, all subjects gained weight. KF and KM reduced body fat by 2% and 1.5%, respectively, and gained 2 and 2.5 kg of lean body mass (LBM), respectively, whereas the CF gained 2.5% body fat and no specific change was found in LBM during ABT (Table 1).
Selected intakes of nutrients at preinduction and during ABT are presented in Table 2. During the year before preinduction, the nutrition intake in KF was lower than the CF for energy (Fig. 1) and for all nutrients that were analyzed (P < 0.001). All groups were deficient in calcium, vitamin D, and zinc; however, these deficiencies were more prominent in the Karakal (Table 2).
Macronutrient intakes preinduction and during ABT are presented in Table 3. The protein, carbohydrate, and fat contents in comparison to the total energy of the soldiers were shown to be adequate.
During ABT, there was a significant decrease in nutrient intake in all groups. The decrease was more pronounced in the CF group (P < 0.001). The differences between the groups (P < 0.01) were for energy, iron, zinc, Mg, vitamin B12, and protein (Fig. 2). Preinduction, the KF consumed less than the CF for each of the nutrients, in most cases significantly. The KF also consumed less than the KM, but in this case not significantly. During training, the order changed. The percentage of change from preinduction to ABT was more pronounced in the CF group, with the medics dropping to third place in every factor (Fig. 3). KM nutrient intake was significantly higher for most nutrients than in both groups of females.
During ABT, KF and KM consumed only 47% and 58%, respectively, of the food table's energy intake that was provided (3620 kcal·d−1) and 66% and 75%, respectively, of the food table's provided calcium (884 mg·d−1). During the medic course, CF consumed 73% of the food table's energy intake (2220 kcal·d−1 was provided) and 97% of the calcium (525 mg·d−1 was provided).
When comparing the national service (NS) recruits to the other female recruits, preinduction energy intake was considerably higher among those who were inducted directly from a home environment (P < 0.0001). The KF who were inducted from home (KFhome; n = 37) had a mean energy intake of 2177 ± 853 kcal·d−1 compared with the KF arriving from NS (KFNS; n = 55) who had mean energy intake of 1869 ± 625 kcal·d−1. The mean intake values for KFhome and KFNS were 773 ± 332 and 654 ± 373 mg·d−1, respectively, for calcium and 14.6 ± 6.3 and 12.8 ± 4.7 mg·d−1, respectively, for iron. This was also true for other nutrients. Mean intake values for the KF vegetarian subgroup (KFveg; n = 55) were lower than the meat consumers (KFmeat; n = 37): energy = 1869 ± 625 and 2177 ± 853 kcal·d−1, respectively; iron = 12.8 ± 4.7 and 14.6 ± 6.3 mg·d−1, respectively.
Table 4 represents the main food group source among all groups preinduction and post-ABT. The main source of energy in the Karakal recruits (both genders), preinduction and during ABT, was grains and cereals. This contrasted with the CF who derived most of their energy from sweets and beverages. Vitamin D, lacking in all groups, came mainly from milk products. Preinduction, the secondary source for vitamin D in CF was only soy and fish (only likely to be served at home) with a mean intake of 11 ± 2 IU. The main source of fat in all females preinduction was milk. During training, milk became secondary to meat. The source of fat for the KM group was meat throughout. Preinduction, the main source of vitamin B12 for all groups was meat. During ABT, the meat group remained the main source for CF and KM. Although the intake declined to 2.4 ± 0.7 μg·d−1 among the CF, it still met 100% of the MDRI. Among the KF, milk was the main source that provided vitamin B12. The main source of iron for all the groups, both preinduction and during ABT, was grains and cereals.
This is the first detailed nutritional assessment conducted in the IDF on combat soldiers. As such, it adds to the very limited information on what soldiers eat. As more is learned on the importance of nutrition for health and performance, particularly in women, this study helps identify specific nutritional deficiencies (2). Although preinduction data are important in itself, particularly where it demonstrates unhealthy nutrition, our purpose was to study what occurs during ABT. In this circumstance, the army is responsible not only for the health of the recruits but is also in the unique position of having some effect on dietary intake, a situation envied by almost all dietitians. In this study, we compared our findings with the MDRI (14), which sets a general standard of recommended nutrient intake for each gender according to levels of physical activity. After careful consideration and consultation, the activity in the Karakal unit was defined as light to moderate, and therefore the energy requirements for the women were 2250 and 3125 kcal·d−1 for the men.
During ABT, all groups were deficient in most nutrients (Fig. 2). Energy consumption was only about 70% of MDRI, calcium 60%, and vitamin D 35%. Regarding the females, these data were significantly less than those for the same subjects during the year preinduction. This decrease was also significant for females coming from a year's NS where their diet was considerably more deficient than the controls who were inducted directly from a home environment. Similar deficiencies have been shown in the US Army: approximately 70% for energy and less than 70% for Ca (2-4). One potential reason for the nutrient deficiencies in the IDF is that the army food tables do not appear to meet the MDRI for most nutrients that were analyzed. The recruits were not consuming similar portions of what they were served. If the IDF menus do not meet the MDRI for some nutrients, then recruits who eat the army menus daily will not meet the MDRI requirements for those nutrients that are deficient in the tables. Because the IDF food tables might not be sufficient, the tables need to be analyzed for nutrient content and reconstructed to meet the MDRI requirements. They could then be used as a comparative standard for nutrient intake. On the other hand, the IDF could establish nutrient standards for its own purposes.
Another possible explanation is that the recruits were not provided with enough food or that they did not eat enough from what they were provided. In other studies, this has been explained by the field conditions, by the stress of training, or by the willingness of the soldier to eat during ABT (2). In addition, one other possibility could be the shortcomings of the FFQ. This is usually true with similar questionnaires that relate to a relatively long period and rely on memory and personal perception. We overcame several of these difficulties by using well-validated FFQ for this population. The FFQ included pictures of typical foods or dishes and were completed under surveillance with adequate assistance. We also repeated the FFQ during the ABT at 8 and 16 wk to validate our findings and found good correlation. However, in this study, we reported the results from preinduction and post-ABT because there were no significant differences between the 8- and 16-wk results. Therefore, we considered the end of ABT (post-ABT) as reflecting consumption during ABT.
Calcium intake was reduced to half of the recommended allowance during basic training among the female groups (KF = 580 mg·d−1; CF = 512 mg·d−1). These levels are lower than the mean calcium intake of 787 mg·d−1 among civilian women aged 18-30 yr (5). As most females in compulsory service in the IDF are between 18 and 22 yr, they are still at the critical age of building their lifetime bone mass. This serious calcium deficiency may compromise their peak bone mass and increase the risk of osteoporosis later in life (5). We calculated the amount of calcium in the army food table served to the CF group to be only 525 mg·d−1, or about half of the recommended requirement. The calcium intake was not much higher in the study group (580 mg·d−1) who were provided with 884 mg·d−1, suggesting that the main reason for low dietary intake stemmed from lack of eating what was actually served. Therefore, we cannot expect the groups to consume the MDRI for calcium if it is not provided or eaten.
Females inducted from NS had lower preinduction dietary intake than the controls. This might be partly related to the unusually high proportion of vegetarians. Thirty-four percent of the females and 11% of the males inducted from NS were vegetarian, a value more than 10-fold their prevalence in the Israeli society. Surprisingly, 14% of the females not inducted from NS were also vegetarians. We have no explanation for this high proportion of vegetarians. Another observation is that the females from NS were the only subgroup to improve their nutrition status during ABT. This may be related to nutritional deficiency during the year of NS due to a lack of orderly meals and a lack of experience in providing their own food. In contrast, during ABT they were served three meals a day with additional rations between meals. The main food groups providing most of the calorie intake among KF (54% after NS) were grains and cereals and milk products (Table 4). This finding contrasts the sweets and the beverages consumed by the CF group. It also helps explain the differences in the eating habits and nutrition consumption between females who were recruited from home to those inducted after NS. With regard to both genders, the main source of energy in ABT was grains and cereals (Table 4). The absolute values in kilocalorie decreased in the females from preinduction to ABT for both grains and cereals and milk products (416-348 and 274-193 kcal·d−1, respectively). This decreased consumption of grains and cereals and milk products increases the risk of underconsumption of related nutrients, a finding similar to that described in the Australian army where carbohydrates, vitamin B2, and calcium were deficient (12). As the recruits in the IDF receive food and snacks from home during training and are also permitted to buy at the army canteen, there is a special need to increase their awareness to eat healthy foods that will include all the essential nutrients.
During most of ABT, the Karakal recruits dined in an organized mess hall and were served according to IDF food table 5 (parallel to US Army ration A). This excluded 2 wk of field activities where rations were served for groups of four soldiers, prepared in the regimental kitchen the same day, and transported to the field warm and also according to army food table 5. These were also parallel to ration US Army ration A (2). Although both genders were served adequate meals, KF and KM did not meet MDRI requirements for energy, calcium, vitamin D, and zinc. Assuming the servings indeed included all that they should, these findings indicate that serving a nutritionally complete menu does not necessarily mean that an adequate diet is being consumed (2). The fact that the nutrient intake of the males was similar to that of the females shows that most of the nutritional deficiencies were probably due to the low overall food intake of the subjects. It is reasonable to assume that females and males being together in the Karakal and dining at the same table contributed to similar eating habits and food consumption among subjects.
Our study experienced several shortcomings. The first was the use of the FFQ for assessing dietary intake. The FFQ is a dietary assessment tool that has been shown with a tendency for overestimating dietary intake (10) and is culture and population sensitive. Therefore, the estimation may be biased. In one validation study, there was a tendency for overestimating total caloric intake in the FFQ questionnaires (10) despite contradictory findings in other studies. However, it is important to note that the FFQ used was developed and validated specifically for the Israeli population and thus enables improved accuracy in reporting traditional and cultural foods (9). In addition, these questionnaires were completed by the soldiers in small groups to help complete them as correctly as possible. The detailed dietary analysis that was carried out allowed us to gather important information regarding eating patterns of female recruits in the Israeli army for the first time. Another shortcoming was that we did not present blood and bone measurements that may have helped us to look at health outcomes in relation to dietary intake. However, it would be of interest to look at diet in relation to health and performance outcomes in further studies.
During ABT, KF and KM reduced body fat by 2% and gained LBM at about 2 kg, whereas CF gained 2.5% and no specific change was seen in LBM (Table 1). Physical activity may have a positive effect on body composition during ABT and army service. The increased body weight among the KF was due to the increase in LBM as a result of the physical activity demands of the combat unit, whereas among the CF it was due to an increase in body fat.
It is clearly necessary to review the IDF food tables. Food table 2 is clearly deficient in calcium and is probably deficient in other nutrients as well. Energy expenditure according to work intensity is classified in the military as light, moderate, heavy, and exceptionally heavy activity (14). Food table 5 seems adequate for the Karakal soldiers with their light-to-moderate energy expenditure. However, this same diet is served on the same base to the infantry recruits whose energy expenditure is classified as heavy and have a daily requirement of 3950 kcal. If the infantry recruits use the food they are served similarly to that reported here, they must be severely nutritionally deficient. It is also recommended that the actual servings be examined because this is also a potential source of inadequate food supply. One method proposed for improving nutrient intake is nutritional education programs tailored to military women, emphasizing the importance of consuming adequate nutrients and varied types of food (2,12). There may not be time for an educational program during ABT, but steps can be taken to increase awareness of dietary intake among the soldiers during this period, such as having posters or information brochures set on the dining tables. The nutritional education program might be more relevant later in the service although preliminary data on these individuals show they were less deficient in nutrients 1 yr after induction. Furthermore, our data lean more toward the need to improve the food supplied, both in quantity and quality, instead of passing the responsibility to the recruits. The army must consider ways to provide more foods containing calcium as well as ways to encourage female soldiers to consume more of these and other dairy products because calcium is of great importance in developing peak bone mass (4,5) and in preventing stress fractures at a later stage. For women entering the 1-yr prearmy NS, a program should be prepared to improve their nutrition. These NS programs need more surveillance, particularly nutritional surveillance, as participants are in their late adolescent period and away from the protective influence of their parents. Our nutritional findings may only be the tip of the iceberg. There may also be other problems that need attention regarding these recruits and the matter should be further investigated.
When reading these data, it is important to keep in mind that most of the data are presented as means. After exclusion of outliers, the mean calcium consumption was 580 ± 224 mg·d−1, median 639. Therefore, 50% of the KF consumed less than 640 mg·d−1 and we may assume, based on statistical assumption, that 16% consumed less than 356 mg·d−1 and 2.5% consumed less than 132 mg·d−1. The same proportions are relevant for other nutrients. We therefore find in this population individuals who are extremely nutritionally deficient. This inadequate dietary intake can endanger the soldiers, lead to a decrease in physical performance, and create other health risks (8).
In conclusion, we have conducted for the first time a detailed survey of food intake patterns among army recruits, females as well as males. Our findings, by and large, confirm findings in other militaries of lower diet quality of soldiers in general and females in particular. These dietary deficiencies can cause lower performance in the field and nonfulfillment of army operational tasks (5). Future emphasis should be placed on the quality of the diet of soldiers in general and females in particular. High-quality nutrition will enhance the soldiers' quality of life (6) and must be made a priority for anyone responsible for the health of these recruits.
There is a need for future nutrition research in the IDF to learn the unique nutritional needs of the male and the female soldiers, particularly among female fighters. In addition, further nutrition studies should be done to determine the suitability of the US MDRI for use in the IDF. The IDF army tables must be examined to ensure they provide adequate nutrients for soldiers. Also, a nutrition education program should be developed and implemented for use in the military. Simple surveys could assess why recruits choose the foods they eat during ABT and the different food choices made by males and females. Additional nutrition studies to evaluate energy expenditure and hormone profiles among women will help the IDF establish and tailor nutrition standards for its soldiers and combat units.
The opinions and assertions in this article are those of the authors and do not necessarily represent official interpretation, policy, or views of the US Department of Defense or the Israeli Defense Forces.
1. Benell K, Grimson S. Factors associated with the development of stress fractures in women. In: Milgrom C, Burr DB, editors. Musculoskeletal Fatigue and Stress Fractures
. Boca Raton (FL): CRC Series in Exercise Physiology; 2001. p. 35-54.
2. King N. Perspectives on nutritional issues of military women. Womens Health Issues
3. King N, Fridlund KE, Askew EW. Nutrition issues of military women. J Am Coll Nutr
4. Klicka MV, King N. Assessment of dietary intakes of cadets at the U.S. military academy at West Point. J Am Coll Nutr
5. Marriott B. Nutrition and health in military women: a life long issue. Womens Health Issues
6. Novack V, Finestone AS, Constantini N, Shpilberg O, Weitzman S, Merkel D. The prevalence of low hemoglobin values among new infantry recruits and nonlinear relationship between hemoglobin concentration and physical fitness. Am J Hematol
7. Poos MI, Costello R, Carlson-Newberry SJ. U. S. Army Institute of Medicine, Food and Nutrition Board, Committee on Military Nutrition Research Activity Report
. Washington, (DC): National Academy Press; 1999. p. 46-9.
8. Rock KL, Lesher LL, Kramer FM, Aylward J, Harrington MS. Field acceptance and nutritional intake of the meal, ready-to-eat and heat and serve ration. U. S. Army Soldier Systems Command, Natick R. D. & E. Center, Natick, MA 01760, TR-98/022; 1998.
9. Shahar DR, Shai I, Vardi H, Brener-Azrad A, Fraser D. Development of a semi quantitative food frequency questionnaire (FFQ) to assess dietary intake of multiethnic populations. Eur J Epidemiol
10. Shai I, Rosner BA, Shahar DR, et al. Dietary evaluation and attenuation of relative risk- multiple comparisons between blood and urinary biomarkers, food frequency and 24-hour recall questionnaires: the DEARR study. J Nutr
11. Siri WE. The gross composition of the body. Adv Biol Med Phys
12. Skiller B, Booth C, Coad R, Forbes-Ewan C. Assessment of nutritional status and fatigue among army recruits during the army common recruit training course: part A. Catering services and diet. Australian Department of Defence, Defence Science and Technology Organisation, CBRN Defense Centre; 2005. p. 1-17.
13. Thairon WJ, Lieberman HR, Montain SJ, et al. Energy requirements of military personnel. Appetite
14. US Departments of the Army, Navy and Air Force HQ. Nutrition Standards and Education. Army Regulation 40-25. Medical Services
. Washington (DC): Department of the Army, Navy, and Air Force; 2001. p. 1-15. Available from: http://www.apd.army.mil/pdffiles/r40_25.pdf