You are here

Article Text

Article menu

Download PDFPDF

Original article
Injuries and injury risk factors among British army infantry soldiers during predeployment training
  1. David M Wilkinson1,
  2. Sam D Blacker1,
  3. Victoria L Richmond1,
  4. Fleur E Horner1,
  5. Mark P Rayson1,
  6. Anita Spiess2,
  7. Joseph J Knapik2
  1. 1Optimal Performance Limited, Bristol, UK
  2. 2United States Army Public Health Command, Aberdeen, USA
  1. Correspondence to David M Wilkinson, Optimal Performance Limited, The Coach House, 26B Oakfield Road, Clifton, Bristol BS8 2AT, UK; david{at}optimalperformance.co.uk

Abstract

Purpose This prospective cohort study examined injuries and injury risk factors in 660 British Army infantry soldiers during a predeployment training cycle.

Methods Soldiers completed a questionnaire concerning physical characteristics, occupational factors, lifestyle characteristics (including physical training time) and previous injury. Direct measurements included height, body mass, sit-ups, push-ups and run time. Electronic medical records were screened for injuries over a 1-year period before operational deployment. Backward-stepping Cox regression calculated HR and 95% CI to quantify independent injury risk factors.

Results One or more injuries were experienced by 58.5% of soldiers. The new injury diagnosis rate was 88 injuries/100 person-years. Most injuries involved the lower body (71%), especially the lower back (14%), knee (19%) and ankle (15%). Activities associated with injury included sports (22%), physical training (30%) and military training/work (26%). Traumatic injuries accounted for 83% of all injury diagnoses. Independent risk factors for any injury were younger age (17–19 years (HR 1.0), 20–24 years (HR 0.71, 95% CI 0.55 to 0.93), 25–29 years (HR 0.89, 95% CI 0.66 to 1.19) and 30–43 years (HR 0.41, 95% CI 0.27 to 0.63), previous lower limb injury (yes/no HR 1.49, 95% CI 1.19 to 1.87) and previous lower back injury (yes/no HR 1.30, 95% CI 1.03 to 1.63).

Conclusion British infantry injury rates were lower than those reported for US infantry (range 101–223 injuries/100 soldier-years), and younger age and previous injury were identified as independent risk factors. Future efforts should target reducing the incidence of traumatic injuries, especially those related to physical training and/or sports.

  • Injuries
  • military personnel
  • occupational
  • physical fitness
  • risk factors
  • sports, surveillance

https://doi.org/10.1136/ip.2010.028233

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    All UK soldiers are required to maintain a state of physical readiness that will allow them to meet the demands of military duty in any training and operational environment. Consequently, regular physical training forms an integral part of military life. However, military physical training is not without risk, and injuries to soldiers are common in both recruit basic training1 2 and regular military training.3–7 Injuries to soldiers reduce both training and operational effectiveness and increase the demands placed on associated medical care provision. Therefore, efforts to reduce injury incidence and injury rates during all types of military training have been a primary focus for many military organisations over recent years.8–16

    Injury prevention is frequently based on the well-established public health model, the first two steps of which require surveillance or survey (to determine the size of the injury problem) and research (to determine the causes of injury and quantify associated injury risk factors).17 In the UK, there is a dearth of robust quantitative data describing the scale of the injury problem among trained British Army personnel and the causes and associated risk factors for injury. Therefore, the purpose of this study was to determine injury rates and examine potential risk factors for injury in British Army infantry soldiers during their year-long predeployment training. Infantry soldiers were selected as they form the largest career employment group within the British Army (approximately 22%), and arguably perform the most physically demanding military roles.

    Methods

    Participants

    Following ethical approval from the Ministry of Defence Research Ethics Committee, soldiers from three British Army infantry battalions (1 Royal Anglian (1 RA), 1 Irish Guards (1 IG) and 4 Rifles) were approached to volunteer to take part in the study. Approximately 98% of 716 soldiers who attended one of eight organised briefing sessions volunteered to participate in the study, giving written informed consent. Soldiers were only included in the study if they were medically classified as employable in full combatant duties (in any area) in any part of the world as of 1 April 2008, reducing the final sample to 660 male soldiers. This number exceeded the minimum target sample size of 500, which would have allowed a difference in injury rates of 36% to be detected between quartiles (four groups of 125), with a power of 81% and p<0.05.

    Physical characteristics, lifestyle data and physical fitness data

    All 660 volunteers completed a lifestyle questionnaire before having their stature and body mass measured. The lifestyle questionnaire was modified from similar lifestyle questionnaires commonly used in other military investigation in the USA.18 19 Sections of the lifestyle questionnaire were designed to obtain occupational data, demographics, physical characteristics, tobacco and alcohol use, physical activity and injury history.

    The soldiers' bi-annual fitness scores (1.5 mile run-time, number of sit-ups in 2 min, number of push-ups in 2 min), performed within 3 months of the start of the study, were obtained from battalion physical training records.

    Injury data

    An injury case was defined as a soldier who sustained physical damage to his body and sought medical care one or more times,2 5 20 as recorded in the Defence Medical Information Capability Programme (DMICP) database, between 1 April 2008 and 31 March 2009. For each injury visit, the medical team completed an injury reporting questionnaire (IRQ), which was adapted from similar questionnaires used in US Army studies.8 19 21 The IRQ data included the date of the visit, type of visit (first entry or follow-up), timing of the injury, activity associated with the injury, anatomical location of the injury, recorded diagnosis and disposition (including number of days restricted duty, if any).

    To ensure complete reporting of injury visits on IRQ (and to complete IRQ with missing information), each soldier's DMICP records were re-screened for injury visits at the end of the study. For each new injury and related follow-ups, one IRQ was completed, with all possible data related to that injury extracted from the DMICP record. New injuries were defined as first entries into the DMICP database for a specific injury, when the DMICP database showed no previous entry for a similar injury (body location and diagnosis) in the preceding 6 months.5 22

    Under the recorded diagnosis section from the IRQ, injuries were grouped by ‘type’ into overuse, traumatic and other injuries. Overuse injuries were diagnosed as due or related to long-term, repetitive energy exchanges resulting in cumulative microtrauma.5 20 Traumatic injuries were diagnosed as due to sudden energy exchanges resulting in abrupt overload with tissue trauma.5 20 Other injury diagnoses included cold and heat injuries, bites/stings, neurological conditions and burns.

    From the IRQ disposition, two ‘levels’ of injury were identified, reflecting different degrees of injury severity.20 The first level included DMICP entries for all types of injury, regardless of whether restricted duty was prescribed. The second level included only those injuries that resulted in one or more days of restricted duty. By combining the ‘type’ and ‘level’ of injury, six different measures of injury were obtained (see table 1).

    View this table:
    Table 1

    Person-time injury incidence rates, new injury diagnosis rates and restricted duty rates by injury type

    Data analysis

    Descriptive statistics (mean±SD) were calculated for anthropometric, fitness and lifestyle characteristics. Frequencies were obtained for timing of and activities associated with injury, anatomical location, diagnosis and disposition. Time at risk for each soldier (days) within the study period was calculated as: 365 − number of days not serving in the army (eg, soldier left the army).

    The injury hazard (instantaneous risk per unit time) was estimated by fitting an exponential function to the Kaplan–Meier survival plot, forcing the intercept through one to give y=e−λx, where λ is the hazard rate.23 Injury incidence, injury incidence rate, new injury diagnosis rate and restricted duty injury rate were calculated using standard equations.5 20 24 25 Restricted duty days included all days in which some limited duty restrictions were imposed, or a soldier was signed off all daily duties.5 20

    Cox regression26 27 was used to estimate univariate HR for all potential risk factors for any injuries and any time-loss injuries. For Cox regression analysis, any injuries included both traumatic and overuse injuries, but not other injuries, so that risk factor analysis for injury effectively focused on musculoskeletal injuries.6 20 Multivariate Cox regression for any injuries and any time-loss injuries used a backward-stepping procedure with exclusion set at p>0.10 for the model. Only variables reaching p≤0.10 in the univariate analysis were entered into the multivariate model.27 Statistical analyses were performed on SPSS version 17.

    For all Cox regression analysis, continuous variables were converted into quartiles, or two, three, or four categories, based on balancing the number of cases in each group to maintain statistical power and/or logical cut scores for a particular variable.

    Results

    Physical, fitness and lifestyle characteristics

    Not all variables included data from all 660 soldiers due to omissions on the lifestyle questionnaire or in battalion physical training records, so sample sizes (n) are shown. The soldiers had an average±SD age, height, body mass, body mass index, 1.5 mile run-time, sit-ups performance and push-ups performance of 23.7±4.8 years (n=646), 1.79±0.07 m (n=660), 79.8±11.5 kg (n=660), 25.0±3.2 kg/m2 (n=660), 9.75±0.78 min (n=576), 64±11 repetitions (n=578) and 58±10 repetitions (n=578), respectively. Among the soldiers, 85% were white and 15% other ethnic origin. Fifty-three per cent of soldiers (341 of 646) smoked cigarettes (13±7 cigarettes/day) and 91% (598 of 655) drank alcohol (5.8±6.9 UK alcohol units/day). The soldiers (n=654) performed 305±174 min/week of non-military physical activity (eg, physical training, sports, recreational activities).

    Injury risk and injury rates

    Over the study period year, 386 soldiers sustained one or more injuries, giving a cumulative injury incidence of 58.5%. The Kaplan–Meier survival data were well described by an exponential function (r2=0.999), with a daily injury hazard of 0.27% (figure 1). There were 574 new injuries reported in total over the year, with 403 of these injuries resulting in soldiers receiving a total of 5179 days of restricted duties. As 18 soldiers were discharged from service before the end of the study, the total time at risk was 238 207 days or 652.6 person-years (including all leave time taken by soldiers and time spend on restricted duty). The injury incidence rates, new injury diagnosis rates and restricted duty rates, classified by injury type, are shown in table 1. During the year, 24 soldiers (3.6%) had their medical status downgraded (ie, were no longer fully deployable for operational duty).

    Figure 1
    Figure 1

    Cumulative probability of remaining injury free over the year (Kaplan–Meier plot). The dotted line represents a theoretical injury hazard of 0.0027 per day.

    Injury timing, associated activities, anatomical locations and diagnoses

    Approximately 75% of all injuries occurred during working hours (excluding travel to work). Physical training (30%), playing sport (22%), military operations/exercises (15%), and non-operation/exercise military work duties (10%) were the most common activities associated with injury. The five most common injury sites were the knee (19%), ankle (15%), lower back (14%), foot/toe (9%) and shin (8%), with injuries to the lower back and lower body accounting for 71% of all injuries. Traumatic, overuse and other injuries accounted for 83%, 13% and 4% of all injuries, respectively. Frequent injuries were traumatic sprains to the ankle (11%) and knee (7%), anterior knee pain (5%) and traumatic strains to the lower back (7%).

    Injury risk factors

    Table 2 shows the univariate Cox regressions for any injury (n=376) and any time-loss injury (n=293). As most injuries were traumatic (83%), traumatic injury risk factors were effectively equivalent to all injuries, and overuse injuries were too few in number to investigate as a separate injury type. For all injuries, higher injury risk was associated with younger age, shorter military service time, lower military rank, soldiers who reported themselves to be more active than their peers, more time performing physical activity, more time performing strength training and previous lower limb or lower back injury. For all time-loss injuries, higher injury risk was associated with younger age, shorter military service, lower military rank, regiment, soldiers who reported being more active than their peers and run training fewer than four times per week (table 2).

    View this table:
    Table 2

    Univariate cox regression results for any injury and any time-loss injury

    Table 3 shows the results of the backward-stepping multivariate Cox regression analysis, which included all variables with an overall significance level of p≤0.10 from the univariate analysis. Independent risk factors for all injuries included younger age and previous injury. Independent risk factors for all time-loss injuries included lower rank, soldiers who reported being more active than their peers, performing fewer than four run training sessions per week and previous lower limb injury (table 3).

    View this table:
    Table 3

    Multivariate Cox regression results for any injury and any time-loss injury

    Discussion

    The present study is the first to report detailed injury rates and injury risk factors in trained British Army infantry soldiers, and provides a baseline measure against which future injury prevention strategies can be evaluated. The overall injury incidence rate was 59 soldiers/100 person-years, with a new injury diagnosis rate of 88 injuries/100 person-years during a year long predeployment training cycle. Eighty-three per cent of all injuries were diagnosed as traumatic injuries. Younger age, lower military rank, soldiers who reported more physical activity than their peers, run training less than four times per week and previous injury were all independent risk factors for injury (table 3).

    Five studies have quantified injury risk and injury rates in trained US infantry soldiers, with study durations ranging from 56 to 395 days.3 7 20 22 28 The injury risk is difficult to compare between studies of different durations, as injury risk is a time-dependent measure.24 However, this study also quantified the injury hazard for British Army infantry soldiers (0.0027/day), which was relatively constant over the year-long study (figure 1). Note the three horizontal sections of the solid line in figure 1 (starting at ∼120, 255 and 320 days) correspond with soldiers' block leave (ie, a period in which all soldiers take leave together), which has been included in the calculation of the injury hazard (and total exposure time) in this study. Although a direct comparison of the injury hazards is not yet possible between infantry studies (as this is the first study to report the injury hazard in infantry soldiers), it does allow the quantification of injury risk (cumulative injury incidence) over any study duration.

    This comparison can be achieved by reading directly from figure 1 to compare the injury risk with studies of equal or shorter duration than this study, or estimating the injury risk using the derived exponential equation for comparison with longer duration studies. When these injury risks were converted to annualised injury incidence rates, the injury incidence rate from this study (59 soldiers/100 person-years) was similar to that reported by Reynolds et al28 of 56 soldiers/100 person-years and Smith et al22 of 58 soldiers/100 person-years. However, these injury incidence rates were much lower than those reported by Knapik et al,7 20 which ranged from 101 soldiers/100 person-years to 127 soldiers/100 person-years when using a similar definition for injury.

    The finding that 83% of all injuries were classified as traumatic injuries in this study was surprising, as previous studies on trained soldiers have consistently shown approximately half of all new diagnosed injuries to be traumatic in nature.5 20 28 However, the traumatic new injury diagnosis rate from this study (73 injuries/100 person-years) was very similar to that reported for US infantry (77 injuries/100 person-years) during similar pre-operational deployment training.20 This suggests that a large part of the difference in injury incidence rates between UK and US infantry during pre-operational deployment training may be explained by a much lower incidence of overuse injuries in British soldiers.

    The reason for this difference in overuse injury rates is not clear, but may be partly explained by the relatively homogenous distribution of 1.5 mile run-times in British infantry soldiers (9.75±0.78 min). Slower 2-mile run time is a consistently demonstrated injury risk factor among trained US infantry soldiers.6 7 20 28 Careful examination of the data in these studies shows that only the slowest categorical group had increased injury risk relative to the most fit group. When categorical boundary run times that define these slowest groups are reported, they start slower than 14 min6 or 15 min20 for the 2-mile run. Even when accounting for the half-mile difference in run distance, relatively few British infantry soldiers would be likely to fall into this slowest group category. This view is supported by the finding that 1.5 mile run-time was not a risk factor for injury in the present study (table 2). The soldiers also reported performing 5.1±2.9 h/week of total physical activity (sports and physical training), with over 40% of soldiers performing less than 2.5 h/week. Although this time excludes performing other physically demanding tasks that are carried out as part of day-to-day infantry duties, it is still a relatively modest total activity volume when compared with many athlete training regimes. In conjunction with an already good level of aerobic fitness (as measured by 1.5 mile run-time), this modest weekly total physical activity time may help explain the relatively low incidence of overuse injuries in this group of soldiers.

    Younger age, shorter service time in the army and lower military rank were risk factors for all injuries. These factors are related (collinear), probably accounting for the fact that only age was included in the final multivariate model. Younger soldiers are likely to be relatively new to army service (minimum joining age is approximately 17 years) and therefore be of low military rank (promotion is usually based on a minimum of 3 years service in each rank). Soldiers of lower rank are more likely to perform the physical tasks that put them at higher injury risk (eg, general section rifleman duties). Those of older age and higher rank are more likely to be in staff or supervisory positions, and therefore less exposed to the physical hazards that put them at risk of injury.7 29

    A link between the amount of physical activity performed and injury risk has been reported in a number of military training studies,3 30–32 and this higher incidence of injury is probably related to an increased exposure to physical hazards with increasing training time. Previous studies of military groups have also reported that previous injuries were associated with current injuries.4 33 Many injuries may be chronic or recurrent, accounting for at least part of this relationship.

    Those soldiers who performed run training for four or more times per week were 32% (HR 0.68) less likely to have a time-loss injury than those who only ran three times or less per week (table 3). This finding may at first seem counterintuitive, as high running volume is known to increase the risk for lower extremity injury substantially.32 34 However, a certain volume of run training may be necessary to promote the specific adaptations required to cope with the physical demands of an infantry role (ie, foot soldier). It is also important to remember that observational studies only identify associations between factors and the risk of injury.24 Further studies should examine the link between injury risk and the amount and type of physical training performed by infantry soldiers, to optimise the balance between reducing injury risk while optimising operational effectiveness.35

    In conclusion, the present study is the first to quantify injury rates and identify risk factors for injury in trained UK infantry soldiers. It provides baseline data for the first two of five necessary phases required for the implementation of an effective injury control programme (ie, surveillance and research).17 Future work to determine injury rates and injury risk factors in other military units and military services should be undertaken to place the results of this study in a wider context and provide quantitative information to support an effective injury control programme for all military personnel. Injury prevention strategies for UK infantry soldiers should target reducing the incidence of traumatic injuries, especially those related to physical training and/or sports.

    What is already known on this subject

    • Injury rates are high during military training for US soldiers preparing for operational deployment (range 101–223 injuries/100 soldier-years).

    • Independent risk factors for injury in US infantry soldiers included low aerobic fitness levels and a high body mass index.

    What this study adds

    • This is the first study to quantify injury rates and injury risk factors for British infantry soldiers (during a pre-operational deployment training cycle), and can be used to evaluate the effectiveness of future injury prevention initiatives.

    • Injury rates were lower for British infantry soldiers (88 injuries/100 person-years) than for US infantry training for operational deployment, due to lower overuse but similar traumatic injury rates.

    • Independent risk factors for new injuries were younger age and previous injury, but not aerobic fitness in these already well trained soldiers.

    References

    1. ↵
      1. Blacker SD,
      2. Wilkinson DM,
      3. Bilzon JL,
      4. et al
      . Risk factors for training injuries among British Army recruits. Mil Med 2008;173:278–86.
      OpenUrlPubMedWeb of Science
    2. ↵
      1. Knapik JJ,
      2. Sharp MA,
      3. Canham-Chervak M,
      4. et al
      . Risk factors for training-related injuries among men and women in basic combat training. Med Sci Sports Exerc 2001;33:946–54.
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Tomlinson JP,
      2. Lednar WM,
      3. Jackson JD
      . Risk of injury in soldiers. Mil Med 1987;152:60–4.
      OpenUrlPubMedWeb of Science
    4. ↵
      1. Schneider GA,
      2. Bigelow C,
      3. Amoroso PJ
      . Evaluating risk of re-injury among 1214 army airborne soldiers using a stratified survival model. Am J Prev Med 2000;18:156–63.
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Knapik JJ,
      2. Jones SB,
      3. Darakjy S,
      4. et al
      . Injury rates and injury risk factors among U.S. Army wheel vehicle mechanics. Mil Med 2007;172:988–96.
      OpenUrlPubMedWeb of Science
    6. ↵
      1. Reynolds K,
      2. Cosio-Lima L,
      3. Bovill M,
      4. et al
      . A comparison of injuries, limited-duty days, and injury risk factors in infantry, artillery, construction engineers, and special forces soldiers. Mil Med 2009;174:702–8.
      OpenUrlPubMedWeb of Science
    7. ↵
      1. Knapik J,
      2. Ang P,
      3. Reynolds K,
      4. et al
      . Physical fitness, age, and injury incidence in infantry soldiers. J Occup Med 1993;35:598–603.
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Knapik JJ,
      2. Bullock SH,
      3. Canada S,
      4. et al
      . Influence of an injury reduction program on injury and fitness outcomes among soldiers. Inj Prev 2004;10:37–42.
      OpenUrlAbstract/FREE Full Text
    9. ↵
      1. Knapik JJ,
      2. Spiess A,
      3. Swedler D,
      4. et al
      . Injury risk factors in parachuting and acceptability of the parachute ankle brace. Aviat Space Environ Med 2008;79:689–94.
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Knapik JJ,
      2. McCollam R,
      3. Canham-Chervak M,
      4. et al
      . Injuries and injury prevention among senior military officers at the Army War College. Mil Med 2002;167:593–9.
      OpenUrlPubMedWeb of Science
    11. ↵
      1. Pope RP
      . Injury surveillance and systematic investigation identify a rubber matting hazard for anterior cruciate ligament rupture on an obstacle course. Mil Med 2002;167:359–62.
      OpenUrlPubMedWeb of Science
    12. ↵
      1. Pope RP
      . Prevention of pelvic stress fractures in female army recruits. Mil Med 1999;164:370–3.
      OpenUrlPubMedWeb of Science
    13. ↵
      1. Rudzki SJ,
      2. Cunningham MJ
      . The effect of a modified physical training program in reducing injury and medical discharge rates in Australian Army recruits. Mil Med 1999;164:648–52.
      OpenUrlPubMedWeb of Science
    14. ↵
      1. Rudzki SJ,
      2. Pope R
      . Injury reductions seen in an infantry brigade using the Australian Defence Injury Prevention Programme (DIPP). Med Sci Sports Exerc 2006;38:S348.
      OpenUrl
    15. ↵
      1. Knapik J,
      2. Darakjy S,
      3. Scott SJ,
      4. et al
      . Evaluation of a standardized physical training program for basic combat training. J Strength Cond Res 2005;19:246–53.
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Jones BH,
      2. Canham-Chervak M,
      3. Sleet DA
      . An evidence-based public health approach to injury priorities and prevention: recommendations for the U.S. military. Am J Prev Med 2010;38:S1–10.
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Stout NA
      . The public health approach to occupational injury research: from surveillance to prevention. Saf Sci 2008;46:230–3.
      OpenUrlCrossRefWeb of Science
    18. ↵
      1. Knapik JJ,
      2. Swedler D,
      3. Grier T,
      4. et al
      . Injury reduction effectiveness of prescribing running shoes based on foot shape in basic combat training. Report no 12-MA-05SB-08. Aberdeen Proving Ground, MD, USA: US Army Center for Health Promotion and Preventive Medicine, 2008.
    19. ↵
      1. Knapik JJ,
      2. Bullock SH,
      3. Canada S,
      4. et al
      . The Aberdeen proving ground injury control project: influence of a multiple intervention programme on injuries and fitness among ordnance school soldiers in advanced individual training. Report no 12-HF-7990–03. Aberdeen Proving Ground, MD, USA: US Army Center for Health Promotion and Preventive Medicine, 2003.
    20. ↵
      1. Knapik JJ,
      2. Darakjy S,
      3. Jones SB,
      4. et al
      . Injuries and physical fitness before and after deployment by the 10th Mountain Division to Afghanistan for Operation Enduring Freedom. Report no 12-MA-05SD-07. Aberdeen Proving Ground, MD, USA: US Army Center for Health Promotion and Preventive Medicine, 2007.
    21. ↵
      1. Knapik JJ,
      2. Jones SB,
      3. Darakjy S,
      4. et al
      . Injuries among army light-wheel vehicle mechanics. Report no 12-MA-7193A-06. Aberdeen Proving Ground, MD, USA: US Army Center for Health Promotion and Preventive Medicine, 2006.
    22. ↵
      1. Smith TA,
      2. Cashman TM
      . The incidence of injury in light infantry soldiers. Mil Med 2002;167:104–8.
      OpenUrlPubMedWeb of Science
    23. ↵
      1. Clark TG,
      2. Bradburn MJ,
      3. Love SB,
      4. et al
      . Survival analysis part I: basic concepts and first analyses. Br J Cancer 2003;89:232–8.
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      1. Hopkins WG,
      2. Marshall SW,
      3. Quarrie KL,
      4. et al
      . Risk factors and risk statistics for sports injuries. Clin J Sport Med 2007;17:208–10.
      OpenUrlPubMedWeb of Science
    25. ↵
      1. Knowles SB,
      2. Marshall SW,
      3. Guskiewicz KM
      . Issues in estimating risks and rates in sports injury research. J Athl Train 2006;41:207–15.
      OpenUrlPubMedWeb of Science
    26. ↵
      1. Verhagen E,
      2. van Mechelen W
      1. Hopkins WG
      . Statistics used in observational studies. In: Verhagen E, van Mechelen W, eds. Sports injury research. Oxford, UK: Oxford University Press, 2010:69–98.
    27. ↵
      1. Machin D,
      2. Cheung YB,
      3. Parmar MKB
      . Survival analysis: a practical approach. 2nd edn. Chichester, UK: John Wiley and Sons Ltd, 2006.
    28. ↵
      1. Reynolds KL,
      2. Heckel HA,
      3. Witt CE,
      4. et al
      . Cigarette smoking, physical fitness, and injuries in infantry soldiers. Am J Prev Med 1994;10:145–50.
      OpenUrlPubMedWeb of Science
    29. ↵
      1. Darakjy S,
      2. Marin RE,
      3. Knapik JJ,
      4. et al
      . Injuries and illnesses among armor brigade soldiers during operational training. Mil Med 2006;171:1051–6.
      OpenUrlPubMedWeb of Science
    30. ↵
      1. Almeida SA,
      2. Williams KM,
      3. Shaffer RA,
      4. et al
      . Epidemiological patterns of musculoskeletal injuries and physical training. Med Sci Sports Exerc 1999;31:1176–82.
      OpenUrlPubMedWeb of Science
    31. ↵
      1. Jones BH,
      2. Knapik JJ
      . Physical training and exercise-related injuries. Surveillance, research and injury prevention in military populations. Sports Med 1999;27:111–25.
      OpenUrlCrossRefPubMedWeb of Science
    32. ↵
      1. Jones BH,
      2. Cowan DN,
      3. Knapik JJ
      . Exercise, training and injuries. Sports Med 1994;18:202–14.
      OpenUrlPubMedWeb of Science
    33. ↵
      1. Knapik JJ,
      2. McCollam R,
      3. Canham-Chervak M,
      4. et al
      . A second investigation of injuries among officers attending the US Army War College, academic year 2000. Report no 29-HE-2682-00. Aberdeen Proving Ground, MD, USA: US Army Center for Health Promotion and Preventive Medicine, 2000.
    34. ↵
      1. Bullock SH,
      2. Jones BH,
      3. Gilchrist J,
      4. et al
      . Prevention of physical training-related injuries: recommendations for the military and other active populations based on expedited systematic reviews. Am J Prev Med 2010;38:S156–81.
      OpenUrlCrossRefPubMedWeb of Science
    35. ↵
      1. Knapik JJ,
      2. Rieger W,
      3. Palkoska F,
      4. et al
      . United States Army physical readiness training: rationale and evaluation of the physical training doctrine. J Strength Cond Res 2009;23:1353–62.
      OpenUrlCrossRefPubMedWeb of Science

    Footnotes

    • Funding This work was funded by the Human Dimension and Medical Science domain of the Ministry of Defence scientific research programme.

    • Competing interests None.

    • Patient consent Obtained.

    • Ethics approval This study was conducted with the approval of the UK Ministry of Defence Research Ethics Committee, reference 0736/121, 20 July 2007.

    • Provenance and peer review Not commissioned; externally peer reviewed.