JRSM >
Volume 99, Number 1 >
Pp. 32-37
1 Regional Medical Centre & Regional Rehabilitation Unit, RAF Halton,
Aylesbury, Bucks HP22 5PG;
2 Department of Primary Care & General Practice, University of Birmingham,
UK
Correspondence to: Wing Commander Richard Withnall
E-mail:
richwithnall{at}doctors.org.uk
 | SUMMARY |
|---|
 TOP SUMMARY INTRODUCTIONMETHODSRESULTSCOMMENTREFERENCES |
|---|
Design: Randomized controlled trial. Setting: RAF Halton, UK. Site of all basic training for RAF personnel.
Participants: 1205 recruits participating in basic training between 17 September 2003 and 7 April 2004.
Interventions: Participants were randomized to receive either standard issue Saran non-shock absorbing insoles, or shock absorbing Sorbothane or Poron insoles, on a 1:1:1 basis.
Main Outcome Measures: The primary outcome measure was withdrawal from training for lower limb injury. The two primary comparisons were shock absorbing insole versus non-shock absorbing insole, and Sorbothane versus Poron (comparison of different shock absorbing insoles). Secondary outcomes were medical withdrawals for reasons other than those qualifying for the primary outcome measure.
Results: When comparing the non-shock absorbing insole to the shock absorbing insoles 72/401 participants (18.0%) allocated to Saran insoles were removed from training because of a qualifying lower limb injury, compared with 149/804 (18.5%) allocated to the shock absorbing insole (Sorbothane or Poron), odds ratio 1.04 (95% CI 0.75 to 1.44; P=0.87). When comparing the two shock absorbing insole 73/421 participants (17.3%) randomized to Sorbothane were removed from training because of a qualifying lower limb injury, compared with 76/383 for Poron (19.8%), odds ratio 0.85 (95% CI 0.58 to 1.23; P=0.37).
Conclusions: Similar rates of lower limb injuries were observed for all insoles (shock absorbing and non-shock absorbing) in the trial. The trial provides no support for a change in policy to the use of shock absorbing insoles for military recruits.
| INTRODUCTION |
|---|
|
TOPSUMMARYINTRODUCTIONMETHODSRESULTSCOMMENTREFERENCES |
|---|
Military personnel require high levels of fitness.2 RAF Halton is the site of Phase 1 training for all noncommissioned RAF (Royal Air Force) personnel. During the 9-week course, Phase 1 RAF recruits undergo standardized intensive physical activity including running, marching, recreational team sports, swimming and gymnasium-based activities such as circuit training. Military recruits are at significant risk of lower limb injury during their initial training.5 As injuries affect military training and operational output, and their treatment is often difficult, time-consuming and expensive, Parkkari et al.1 suggest that preventative strategies may be justified both on clinical and economical grounds.
Professional sporting organizations and the military have started to utilize shock absorbing insoles (SAIs) in an attempt to prevent overuse injury. SAIs may protect against injury by reducing the magnitude and rate of loading of the peak forces generated at foot strike during running and walking, reducing the ground reaction forces across the foot, and the loads transmitted to the skeleton.6
UK military personnel currently use three principal types of commercially produced insole:
Running produces a high proportion of lower back and lower limb injuries due to the velocity-dependent ground impact forces at each heel-strike which range from 1.2 times body weight when walking to 3.6 times body weight whilst sprinting.4 Biomechanics, such as joint angular velocities, influence impact forces more than anatomical factors.10,11 Heel-strike impact forces induce soft tissue resonance, however such vibrations are small or unrecordable.12 The amount of work done during a physical activity such as locomotion is dependent upon the ground-foot interface,13 and has a direct bearing upon performance.14 Different combinations of materials in shoes soles, insoles or sports surfaces dissipate different amounts of energy,13 change muscle activity, and affect the body's musculoskeletal work capability and subsequent injury rates.15 Nigg et al.12 demonstrated that wearing viscoelastic rather than solid heels during exercise reduced oxygen requirement (as a measure of work) by up to 2%.
Unpublished RAF data indicated that between 1996-1997 48% of recruits sustained injuries during phase 1 training: 80% of these injuries affected the lower limbs; 80% of all reported injuries occurred during high impact training in weeks 2 and 3, and of these, 35% of injuries affected the shin and foot (PL White personal communication). Each intake of approximately 70 personnel would lose 250 training days as a result of injury, representing 1.5% total training time available. Between 1997-1998, a local study of Sorbothane SAIs was undertaken, reporting a 28% reduction in the overall injury rate, with injuries occurring in weeks 2 and 3 reducing from 80% to 30%. A 38% reduction in training days lost was achieved. As a result, Sorbothane SAIs were issued to all RAF recruits undergoing phase 1 training. However, as the study was a non-randomized ‘before and after comparison’, the differences could be due to bias such as the Hawthorne (participation in a study) effect or the related placebo effect.
The relative merits of SAIs have usually been judged retrospectively. Systemic review provided insufficient evidence to suggest SAIs significantly reduce the incidence of overuse injury.17 As the review included studies of only ‘low to moderate’ quality, the authors acknowledged more data are required. Thacker et al.18 also proposed rigorous research to address common sports medicine problems, as the absolute and relative effectiveness of many interventions remain poorly understood.19
No RCTs have compared differing SAIs within a UK military clinical setting, and there is currently no consolidated UK military policy concerning the use of SAIs. As RAF recruits are drawn from a cross-section of the British population, males and females of differing socioeconomic class, ethnic background and levels of initial physical ability were included.
| METHODS |
|---|
|
TOPSUMMARYINTRODUCTIONMETHODSRESULTSCOMMENTREFERENCES |
|---|
Exclusion criteria were: failure to provided informed consent; past medical history of lower limb injury; any ongoing medical problems; current pregnancy.
Participants were included from 11 consecutive intakes at RAF Halton between 17 September 2003 and 4 February 2004. Follow up was conducted between 17 September 2003 and 7 April 2004.
Interventions
Participants were randomized to receive Saran (control), Sorbothane or
Poron SAIs by 1:1:1 proportion. Those meeting the inclusion and exclusion
criteria above were issued their randomized insole appropriate on the day of
basic kitting at the beginning of their basic training.
Objectives
The objective of the study was to estimate the beneficial effects, if any,
of the use of newer and more costly SAIs on the rate of lower limb injury in
participants undergoing high levels of physical activity associated with RAF
basic training. There were two primary analyses. First, to see if the use of
either Poron or Sorbothane SAIs would reduce the rate of qualifying lower limb
injury in comparison with Saran insoles. The second was to estimate the
difference, if any, in the rate of qualifying lower limb injuries in
participants using either Sorbothane or Poron SAIs. Secondary objectives
included examining the rate of other, non-qualifying injuries sustained during
basic training, and examining the influence of RAF Flight allocation (grouping
for basic training) on the protective effects, if any, of different
insoles.
Outcomes
The primary outcome measure was defined as any lower limb injury diagnosed
by the Regional Medical Centre doctors, nurses or physiotherapists from
participants who presented to the Medical Centre requesting withdrawal from
training on account of injury. Clinical staff at RAF Halton, very experienced
in dealing with lower limb injury, received additional training in the
classification of injury in the context of the trial. Secondary outcomes were
other debilitating injuries resulting in withdrawal from training.
Sample size
Separate sample size calculations were performed for each primary
comparison. Since there are two non-independent primary analyses in this
study, the available statistical power was divided between the two analyses to
preserve the overall critical 
level of 5%
(two-sided).20 For
the comparison of SAIs (Sorbothane or Poron) versus Saran, randomizing 400
patients to each group, and thus comparing 800 participants using SAIs with
400 participants using Saran insoles, was sufficient to detect a 11.5%
difference as statistically significant, with 90% power (17ß) at the
allocated critical level of 1.67%, making the conservative assumption
that 50% of participants would experience a primary outcome.
For the comparison of different SAIs (Sorbothane versus Poron) randomizing
400 patients to each group was sufficient to detect an 11.5% difference as
statistically significant, with 90% power (17ß) at the allocated critical
level of 3.33%, making the conservative assumption of a rate of
primary outcome in the less effective insole group of 43.5%.
Therefore the intention was to randomize 400 participants in each group. As each intake to RAF Halton for basic training includes about 100 potential participants, recruitment was continued offering participation in the study to all eligible participants in each intake until more than 1200 participants had been recruited.
Randomization
A randomization code was generated using a simple three-way randomization
algorithm programmed in Microsoft Access, and stratified by flight. Random
allocation was through a concealed process, in which details of all recruits
expected to attend RAF Halton were sent to the randomization centre at the
University of Birmingham during the week before their planned arrival.
Provisional allocation for each recruit was generated in Birmingham and
communicated with the study personnel at RAF Halton. Participants were
recruited in Halton by staff blinded to the allocation of participants. The
allocated insoles were supplied to participants during ‘kitting’
on the second day of their phase 1 training.
Blinding
The three insoles utilized in this study are not identical in design, and
thus blinding of investigators and participants was not possible.
Statistical methods
The primary analyses used Fisher's exact test, providing odds ratios and
95% confidence intervals. P values are also provided and were used to
assess statistical significance at the higher level required due to the alpha
spending procedure dividing available statistical power between related
comparisons in this three-arm trial. In addition, absolute risk differences
and 95% confidence intervals were provided to enable assessment of the
practical importance of the observed differences between the groups.
The primary analyses were conducted using the intention to treat principle.21 Participants who did not complete phase 1 training were included in the study until they left the service.
A generalized linear model with a logit link and binomial error was used to assess the affect of flight upon the primary outcome (analogous to the inclusion of treatment centre in a multi-centre trial). Overdispersion (extra binomial variability) was accounted for by inflating the scale factor by the ratio of the residual deviance and the degrees of freedom, on the appropriate stratum.
Ethical approval
Ethical approval for the study was provided by the RAF Experimental Medical
Ethics Committee.
| RESULTS |
|---|
|
TOPSUMMARYINTRODUCTIONMETHODSRESULTSCOMMENTREFERENCES |
|---|
|
Baseline characteristics
As anticipated, the participants in this trial were largely males in their
late teens and early twenties. The baseline characteristics by group are
described in Table 1.
|
Withdrawal from training for medical reasons
During the study there were 221 withdrawals from training that met the
criteria for primary outcome events. These are described in
Table 2. In addition, there
were 68 withdrawals from training for medical reasons other than those
qualifying for the primary endpoint in the study. These are described in
Table 3.
|
|
Saran versus shock absorbing insole
Primary outcome
Overall 72/401 participants (18.0%) allocated to Saran insoles were removed
from training because of a qualifying lower limb injury, compared with 149/805
(18.5%) allocated to a SAI (Sorbothane or Poron), odds ratio 1.04 (95% CI 0.75
to 1.44; P=0.87). The difference in withdrawal per group was small in
absolute terms, risk difference 70.6 (95% CI 75.1 to 4.2%), number needed to
treat (NNT) with a SAI to avoid a withdrawal is NNT=174_harm (95% CI 20_harm
to 24_benefit).
Affect of flight on primary comparison of Sorbothane and Poron versus Saran
The influence of flight membership on the outcome was examined using a
logistic model in which flight was included as a classification variable. This
provided similar results to the unstratified analysis: odds ratio 1.04 (95% CI
0.76 to 1.42; P=0.81).
Other non-qualifying withdrawals
Overall 14/401 participants (3.5%) allocated to Saran insoles withdrew from
training for medical reasons other than qualifying lower limb injuries. These
injuries included lower back pain, groin pain, hip pain and multiple
conditions. By contrast, 54/804 (6.7%) withdrew for these reasons in the SAIs
group, odds ratio 1.99 (1.07 to 3.93; P=0.024). In absolute terms,
this represented 3.2% more withdrawals with Sorbothane or Poron compared with
Saran (95% CI 0.5% to 5.7%), and number needed to treat with SAIs
(NNT)=32_harm (18_harm to 196_harm).
Sorbothane versus Poron
Primary outcome
Overall 73/421 participants (17.3%) randomized to Sorbothane were removed
from training because of a qualifying lower limb injury, compared with 76/383
for Poron (19.8%), odds ratio 0.85 (95% CI 0.58 to 1.23; P=0.37). The
difference in withdrawal per group was small in absolute terms, risk
difference for Poron minus Sorbothane is 72.5% (95% CI 2.9% to 77.9%). This
equates to a number needed to treat with Sorbothane rather than Poron of
40_benefit (95% CI 35_harm to 13_benefit).
Affect of flight on primary comparison of Sorbothane versus Poron
The influence of flight membership on the outcome was examined using a
logistic model in which flight was included as a classification variable. This
provided similar results to the unstratified analysis: odds ratio 0.85 (95% CI
0.59 to 1.21; P=0.36)
Other non-qualifying withdrawals
Overall 26/421 (6.2%) participants randomized to Sorbothane withdrew from
training for medical reasons other than qualifying lower limb injuries. By
contrast 28/383 (7.3%) allocated to Poron withdrew for these reasons, odds
ratio 0.85 (95% CI 0.47 to 1.52; P=0.58). In absolute terms, this
represents 1.1% fewer withdrawals due to reasons other than primary outcome in
the Sorbothane group (95% CI 72.4% to 4.8%).
| COMMENT |
|---|
|
TOPSUMMARYINTRODUCTIONMETHODSRESULTSCOMMENTREFERENCES |
|---|
This trial shows that SAIs do not lessen the risk of lower limb injury for military personnel. This result may be generalizable to civilian groups involved in physically demanding exercise, such as police forces, firemen, nurses, dancers, amateur recreational sportsmen and women; and professional individual athletes and sporting teams at local, regional, national and international level. Despite mechanistic theory that SAIs may augment innate muscle tuning in reducing heel-strike impact forces and lessening intra-compartmental vibrations, both Sorbothane and Poron have failed to influence injury rates in a military setting.
Our study was a large-scale randomized trial, providing good evidence on the relative effectiveness of the insoles examined. The study was conducted in military recruits and the extent to which the findings may be directly extrapolated to other potential users of shock absorbing insoles is unclear. Given that we found no evidence to support the use of shock absorbing insoles, it may seem unlikely that benefits would be observed in other similarly active populations. However there may be situations outside the scope of our work in which this may be the case.
Such a result carries financial implications for the military. This trial provides no justification for the continued funding of disproportionately expensive SAIs over standard insoles. In the military setting, the additional expenditure conveys no benefit to training or operational output, and should not be sustained. Additional savings may be achieved by extending such a policy across the wider RAF, Army and Royal Navy.
Investment in further research into risk management and injury reduction within the military training environment is proposed to augment the evidence base against which to target resources cost-effectively.
| Acknowledgments |
|---|
| Footnotes |
|---|
| REFERENCES |
|---|
|
TOPSUMMARYINTRODUCTIONMETHODSRESULTSCOMMENTREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Related articles in JRSM:
This article has been cited by other articles:
|
|
 |
|
  K. O'Leary, K. A. Vorpahl, and B. Heiderscheit Effect of Cushioned Insoles on Impact Forces During Running J Am Podiatr Med Assoc, January 1, 2008; 98(1): 36 - 41. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Aaltonen, H. Karjalainen, A. Heinonen, J. Parkkari, and U. M. Kujala Prevention of Sports Injuries: Systematic Review of Randomized Controlled Trials Arch Intern Med, August 13, 2007; 167(15): 1585 - 1592. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P McCrory Learning to love big brother Br. J. Sports Med., June 1, 2006; 40(6): 487 - 487. [Full Text] [PDF]
|
|
|
|
 |
|
 Minerva BMJ, January 14, 2006; 332(7533): 128 - 128. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
