Here’s an article from module 2 of my previous year on my Strength & Conditioning MSc at St Mary’s University. The brief for the work is below along with the article I chose to critique. Any feedback on the article is welcome as are any questions!!
You will be provided 3 articles (downloadable from SimmsCAPital). You will be required to read these articles and select one, on which you will proceed to provide a 2000 word critical analysis. Your analysis should consider key areas of research including; rationale for the hypotheses, research design, measures utilised, subject group, level of control, external validity and applicability of findings. Your work should be heavily supported with evidence from peer reviewed sources.
My chosen article –
Bennell, K., Wajswelner, P., Lew, P., Schall-Riacour, A., Leslie, S., Plant, D., & Cirone, J. (1998). Isokinetic strength testing does not predict hamstring injury in Australian Rules footballers. British Journal of Medicine, 32, 309-314.
This essay will critically analyse the research by Bennell et al. (1998) which assesses whether isokinetic strength testing at the commencement of the season can predict the occurrence of a hamstring strain injury in Australian Rules football. This analysis will consider the research design, subject group, level of control, external validity and applicability of any findings utilising relevant literature in this research field to conclude how accurate the research is and its level of importance. On conclusion an opinion on whether the author’s findings were correct and an assessment of the overall worth of research will be made.
The prospective cohort design implemented by Bennell et al. (1998) is a strength of the research as quality control is simplified and a high degree of completeness can be secured (Bahr & Holme, 2003). It also allows analysis on whether the injury was caused due to a prior lack of strength where as retrospective study designs are not able to distinguish between risk factors and injury sequelae (Bahr & Holme, 2003). A weakness with prospective research studies is insufficient study size (Crosier, Ganteaume, Binet, Genty & Ferret, 2008) resulting in a lack of statistical power. Bahr and Holme (2003) identify this issue recommending that due to the intraclass correlation coefficient of maximal hamstring strength cohort studies should use a minimum of one hundred and fifty injuries to detect small to moderate associations and thirty to fifty injuries to detect moderate to strong associations. With only twelve injury cases in the research by Bennell et al. (1998) the reliability of the results and validity of any findings is low as the authors were making recommendations from analysis of statistics low in power. Consequently any recommendations concluded from the research must be analysed with caution as the authors may be misled by the results. Therefore the conclusion by Bennell et al. (1998) that isokinetic muscle strength testing is not able to identify players most at risk of hamstring strain injury is potentially inaccurate. Furthermore a prospective study by Crosier et al. (2008) opposes the conclusion of Bennell et al. (1998) stating that isokinetic strength testing as a preseason screening tool is warranted. The research of Crosier et al. (2008) was performed with a research design matching that of Bennell et al. (1998) but analysed thirty five injuries in comparison to twelve which provides greater statistical power to their research. This leads to the view that the results and conclusions of Crosier et al. (2008) are more accurate and hold more importance than that of Bennell et al. (1998).
On analysis of the subject group used by Bennell et al. (1998) it is made up of differing levels of playing standard with a split between amateur and professional players. This non consistent variable negatively affects the research as a participant’s injury risk may be increased dependant on their playing level (Heidt, Sweeterman, Carlonas, Traub & Tekulve, 2000) often due to varied training intensities producing differences in fatigue levels (Worrell & Perrin, 1992). Prior, Guerin and Grimmer (2009) also concluded that higher levels of competition increases the risk of a hamstring strain. They concluded this on review of Verrall, Slavotinek, Barnes, Fon, and Spriggins (2001) study where it was concluded that Australian Football League (AFL) players sustained significantly more posterior thigh injuries than players performing in the lower level South Australian National Football League (SANFL). The subjects in the research of Bennell et al. (1998) were split similarly to that of Verrall et al. (2001) with six professional teams and four amateur teams used. Bennell et al. (1998) highlight themselves that different risk factors may be operating in different sports and individuals. They conclude that this makes it difficult to establish predictive relations. Further to this it is advised that variable sample groups may prevent generalisations to the wider sporting arena to be made (Foreman, Addy, Baker, Burns, Hill & Madden, 2006). It is therefore proposed that the differences in subject playing level reduces predictive relations and the transferability of the conclusion that isokinetic strength testing as a preseason screening tool is limited.
It is felt that the lower playing standard transfers to the skill level of the physiotherapy treatment available. This is both in the skill of the physiotherapists themselves and the resources and equipment available to correctly diagnose injuries. The measures set in place by Bennell et al. (1998) to diagnose a hamstring strain received prior to and during the study have potential to miss or misdiagnose strains. Only 57% of the recorded hamstring injuries in their study were confirmed as injuries to the hamstring through ultra sound scan. The remainder of the recorded hamstring strains were diagnosed by the medical staff at the participating clubs which on reading the editorial totalled fourteen different individuals at ten different clubs. This process of individual subjective interpretation of what is a hamstring strain and what is not could easily lead to misdiagnosis as the criteria used to diagnose could vary between clinicians (Gabbe, Bennell, Finch, Wajswelner & Orchard, 2006). The importance of accurate diagnostic protocols is further highlighted by Bahr and Holme (2003). They comment that the key is not to miss any injuries of interest by using appropriate diagnostic methods also ensuring no false positives are included. Bennell et al. (1998) seem to have a lack of control in diagnosing injuries with the internal consistency and consequently the reliability of the research in question.
Further evidence to support this is found on analysis of the diagnosis protocols and findings in the research by Verrall et al. (2001). In their study any athlete experiencing posterior thigh pain was given a magnetic resonance imaging (MRI) scan to establish the presence or absence of muscle strain injury. MRI scans have been shown to be sensitive at correctly diagnosing hamstring strains (Brandser, El-Khoury, Kathol, Callaghan & Tearse, 1995). Interestingly out of the thirty two athletes sent for an MRI in the research by Verrall et al. (2001) six athletes were deemed to have normal scans with their injury re-diagnosed as a referred pain posterior thigh injury. These findings highlight a weakness in the protocols used by Bennell et al. (1998) to diagnose hamstring strains. These thoughts are supported by Devlin (2000) who raises the fact that where diagnostic tests are not used the difference between a referred pain injury and an actual hamstring strain injury cannot be made. Implications of this is that injuries that weren’t actually hamstring strains but were referred pain posterior thigh injuries could have been included in the data collected. Ultimately if the data collected is not accurate then the accuracy of the eventual results is questioned along with the findings and recommendations made from analysis of the data.
It is felt that the effect of misdiagnosed hamstring injuries being included in the data collection in the research of Bennell et al. (1998) is large enough to have a material effect on the results. The research of Verrall et al. (2001) found 19% of initially diagnosed hamstring strains were not actually strains to the hamstring muscle. Furthermore 45% (14 total cases) of posterior thigh injuries were misdiagnosed as hamstring strains in research by Gibbs, Cross, Cameron and Houang (2004) and 31% (18 total cases) in a study by Schneider-Kolsky, Hoving, Warren and Connell (2006). MRI scans were again used in these studies to confirm that an actual hamstring injury was present. The implications of misdiagnosed strains entered into the wrong data group are statistically significant as the strength test results of these subjects would be significantly higher than the rest of the subjects in the injured hamstring group. This abnormal result would statistically alter the group mean and standard deviation leading to a wrong conclusion made by the authors.
On analysis of the isokinetic testing methods implemented by the Bennell et al. (1998) queries are raised over the validity and suitability of the testing procedures. The subjects warm up prior to the test involved a ten minute warm up and light stretching. Recommendations state that static stretching can inhibit a muscles maximum force output (Bacurau, Monteiro, Ugrinowitsch, Tricoli, Cabral & Aoki, 2009; Power, Behm, Cahill, Carroll & Young, 2004) highlighting the warm up protocols in place were not satisfactory. Although the warm up protocols used were unsatisfactory it is felt that the effect on the internal validity of the research is minimal as all athletes partaking in the study performed the test under the same protocols creating a level of internal consistency. The warm up protocols are also matched by that in the studies by Crosier et al. (2008), Askling, Karlsson and Thorstensson (2003) and Parcell, Sawyer, Trocoli and Chinevere (2002). On consideration of this it is felt that although the protocol set for warming up before taking the test has its limitations they will not have a significant material impact on the results of the research’s external validity when analysed on its own and when reviewed alongside other literature in this research area.
This is the same for the rest times given between repetitions. Rest has a critical influence on tension generating capabilities of muscle with adequate recovery important when evaluating muscle functional characteristics (Parcell et al., 2002). If rest periods between tests are not standardised and are not appropriate fatigue may develop in the musculature involved resulting in the test failing to actually measure maximum strength (Robinson, Stone, Johnson, Penland, Warren, & Lewis, 1995). Pincivero, Lephart and Karaunakara (1997) concluded that rest periods of one hundred and sixty seconds produced greater strength levels than a rest period of just forty seconds. Conflicting research by Parcell et al. (2002) advises that minimum rest periods of sixty seconds are adequate for isokinetic strength testing with Bottora, Russo and Oliveria (2005) concluding at least thirty seconds rest is sufficient. Although it is beyond the scope of this essay to fully review the current literature in this area it is identified that the recovery protocols set in place by Bennell et al. (1998) match that of other respected literature in the research area (Aagard, Simonsen, Magnusson, Larsson & Dyhre-Poulsen, 1998; Bottora et al., 2005; Crosier, Forthomme, Namurois, Vanderthommen & Crielaard, 2002; Parcell et al., 2002). This enables the results gathered to be transferable providing a level of external validity to this area of the study design.
Interestingly the Bennell et al. (1998) comment on how back related hamstring injuries were excluded from the study but provide the reader with no information as to why this was done and how these injuries were identified. There is also no comment on the number of injuries excluded because of this and no description of what the injury is. Verrall et al. (2001) report neuromeningeal structures may cause athletes referred pain in the posterior thigh which they categorize as back related posterior thigh pain. They continue to comment that prior to excluding players due to back related injuries further research should be undertaken to assess the validity of any assertions. It is felt that with the relatively small subject size any exclusion of participants or data should be explained and justified. As this was not done one questions the credibility of Bennell et al. (1998) as to whether they possess the relevant experience, expertise or understanding in this research area to provide the relevant information or justification. It is likely that again this information was provided to them from the participating clubs medical staff which carries the same subjective analysis issues which have been previously identified. This lack of control the authors seem to have continues from the diagnosis protocols set in place and is of concern on analysis of the accuracy of any findings.
Bennell et al. (1998) concluded that isokinetic strength testing was unable to discriminate between injured and uninjured players, injured and uninjured legs or predict hamstring injury. It is felt that a lack of validity in the results led them to conclude this when stricter and more accurate injury diagnosis protocols may have led them to a different conclusion. Although there were strengths in the prospective cohort design implemented the lack of statistical power due to inadequate subject numbers alongside the non-consistent variable that was the playing level of the subject group it is felt that the research design was ultimately weak. A lack of control in the data collection and injury diagnosis is a theme that reduces the statistical value of the study and leads to the conclusion of inaccuracy in the research. In comparison to the work by Crosier et al. (2008) where MRI scans were used to identify injuries and the sample size was greater producing a stronger level of statistical power the research by Bennell et al. (1998) fails to provide the same level of external validity and applicability. Ultimately it is felt that these weaknesses identified have a negative impact on the message delivered in the paper. It is felt that the conclusion made by Bennell et al. (1998) is inaccurate leading to the conclusion that the research should hold minimal importance to the research area with there being a lack of overall worth to the research.
References
Aagaard, P., Simonsen, E.B., Magnusson S.P., Larsson, B., & Dhyre-Poulsen, P. (1998). A new concept for isokinetic hamstring: quadriceps muscle strength ratio. The American Journal of Sports Medicine, 26, 231-237.
Askling, C., Karlsson, J., & Thorstensson, A. (2003). Hamstring injury occurrence in elite soccer players after preseason strength training with eccentric overload. Scandinavian Journal of Medicine & Science in Sports, 13, 244-250.
Bacurau, R.F., Monteiro, G.A., Ugrinowitsch, C., Tricoli, V., Cabral, L.F., & Aoki, M.S. (2009). Acute effects of ballistic and a static stretching exercise bout on flexibility and maximal strength. Journal of Strength and Conditioning Research, 23, 304-308.
Bahr, R., & Holme I. (2003). Risk factors for sports injuries – a methodological approach. British Journal of Sports Medicine, 37, 384-392.
Bennell, K., Wajswelner, P., Lew, P., Schall-Riacour, A., Leslie, S., Plant, D., & Cirone, J. (1998). Isokinetic strength testing does not predict hamstring injury in Australian Rules footballers. British Journal of Medicine, 32, 309-314.
Bottaro, M., Russo, A., & Oliveira, R.J. (2005). The effects of rest interval on quadriceps torque during an isokinetic testing protocol in elderly. Journal of Sports Science and Medicine, 4, 285-290.
Brandser E.A., El-Khoury G.Y., Kathol M.H., Callagn, J.J., & Tearse, D.S. (1995). Hamstring injuries: radiographic, conventional tomographic, CT and MR imaging characteristics. Radiology, 197, 257–262.
Crosier, J.L., Forthomme, B., Namurois, M.H., Vanderthommen, M., & Crielaard, J.M. (2002). Hamstring muscle strain recurrence and strength performance disorders. American Journal of Sports Medicine, 30, 199-203.
Crosier, J.L., Ganteaume, S., Binet, J., Genty, M., & Ferret JM. (2008). Strength imbalances and prevention of hamstring injury in professional soccer players. The American Journal of Sports Medicine, 36, 1469-1475.
Devlin, L. (2000). Recurrent posterior thigh symptoms detrimental to performance in rugby union. Predisposing factors. Sports Medicine, 29, 273–287.
Foreman, T.K., Addy, T., Baker, S., Burns, J., Hill, N., & Madden, T. (2006). Prospective studies into the causation of hamstring injuries in sport: A systematic review. Physical Therapy in Sport, 7, 101-109.
Gabbe, B.J., Bennell, K.L., Finch, C.F., Wajswelner, H., & Orchard, J.W. (2006). Predictors of hamstring injury at the elite level of Australian Football. Scandinavian Journal of Medicine & Science In Sports, 16, 7-13.
Gibbs, N,J., Cross, T.M., Cameron, M., & Houang, M.T. (2004). The accuracy of MRI at predicting recovery and recurrence of acute grade one hamstring muscle strains within the same season in Australian Rules football. Journal of Science and Medicine in Sport, 7, 248-258.
Heidt, R.S., Sweeterman, L.M., Carlonas, R.L., Traub, J.A., & Tekulve, F.X. (2000). Avoidance of soccer injuries with preseason training. The American Journal of Sports Medicine, 28, 659-662.
Parcell, A.C., Sawyer, R.D., Tricoli, V,A., & Chinevere, T.D. (2002). Minimum rest period for strength recovery during common isokinetic testing protocol. Medicine and Science in Sport and Exercise,34, 1018-1022.
Pincivero, D.M., Lephart, S.M., & Karanukara, R.G. (1997). Effects of rest interval on isokinetic strength and functional performance after short-term high intensity training. British Journal of Sports Medicine, 31, 229-234.
Power, K., Behm, D., Cahill, F., Carroll, M., & Young, W. (2004). An acute bout of static stretching: Effects on force and jumping performance. Medicine and Science in Sport and Exercise, 36, 1389-1396.
Prior, M.B., Guerin, M., & Grimmer, K. (2009). An evidence based approach to hamstring strain injury: A systematic review of the literature. Sports Health A Multidisciplinary Approach, 1, 154-164.
Robinson, J.M., Stone, M.H., Johnson, R.L., Penland, C.M., Warren, B.J., & Lewis, R.D. (1995). Effects of different weight training exercise/rest intervals on strength, power, and high intensity exercise endurance. Journal of Strength and Conditioning Research, 9, 216-221.
Schnieder-Kolsky, M.E., Hoving, J.L., Warren, P., & Connell, D.A. (2006). A comparison between clinical assessment and magnetic resonance imaging of acute hamstring injuries. The American Journal of Sports Medicine, 34, 1008-1015.
Verrall, G.M., Slavotinek, J.P., Barnes, P.G., Fon, G.T., & Spriggins, A.J. (2001). Clinical risk factors for hamstring muscle strain injury: A prospective study with correlation of injury with magnetic resonance imaging. British Journal of Sports Medicine, 35, 435-440.
Worrell, T.W., & Perrin, D.H. (1992). Hamstring muscle injury: The influence of strength, flexibility, warm up and fatigue. Journal of Orthopaedic Sports Therapy, 16, 12-18.
