You don't need to be signed in to read BMJ Blogs, but you can register here to receive updates about other BMJ products and services via our site.


The Premier League and the Professional Footballers Association must censure Chelsea: by Professor William Tormey

15 Aug, 15 | by Karim Khan


Medical ethics clashes with team priorities

Jose Mourinho’s actions in criticising Dr Eva Carneiro and physiotherapist Jon Fearn for attending an injured player on the pitch at Chelsea must be vigorously challenged. The General Medical Council’s ‘Good Medical Practice’ states that the doctor must make the care of the patient the first concern and take prompt action if the doctor thinks that patient safety, dignity or comfort is being compromised.

The treatment of Dr Eva Carneiro by Chelsea was unprofessional, carried out in public and should have consequences. Her judgement was denigrated despite her obvious concern for the injured player. Public humiliation is no way to treat any club doctor.

The attendant massive publicity demands that the British Medical Association and the professional bodies involved in sports medicine insure that medical ethics are respected. The Premier League and the Professional Footballers Association must censure Chelsea and publicly assert support for good professional standards in their medical services. There should be no equivocation.


Professor William Tormey

Biomedical Sciences

Ulster University


Northern Ireland

Phone 00353872544646


There are no financial associations and no conflicts of interests

Guest post by @DrJohnOrchard. On Andre Villas-Boas, the unreasonable pressure on coaches/managers, and why player health should be in clinicians’ hands

18 Dec, 13 | by Karim Khan


A month is a long time in football

OrchardPICOn November 21st, I was one of three sports physicians who wrote a Blog at BJSM on the topic of concussions in football & managerial interference in medical decisions. I tried to assess the risks involved for all of the participants in the Hugo Lloris concussion incident. Perhaps controversially, I estimated that the (then) Tottenham manager, Andre Villas-Boas, had between a 1 in 3 and 1 in 5 chance of being sacked this season. Well as it turns out if I had have offered to hold bets at these apparently meagre odds I would have been taken to the cleaners, as he didn’t survive the calendar year, let alone the remainder of the football season.

Which begs the question, why should someone (an EPL manager) with a job expectancy of roughly a year – give or take – have any role in decisions which may have an impact on the health of the player 20 years down the track? The answer is that of course they shouldn’t, but of course they do. If there was one thing that AVB made very clear in his short tenure, it was that he and he alone decided when players were substituted off the field. Other managers have said that they respect the opinion of their medical staff, but those at the coalface know of pressure to not be “too conservative” in a cut-throat world with limited substitutions.

It’s not surprising that a manager would put ‘team performance’ ahead of ‘long-term player welfare’. (We are not pointing any fingers – we are just drawing a logical conclusion).

Did AVB’s stance on concussion have a role in his downfall? I suspect not; my experience in professional sport is that managers are judged primarily on (poor) results. If Tottenham were leading the EPL then he would have been getting praised for being a strong leader who made tough decisions. Since my November blog, the Australian cricket coach Darren Lehmann has talked about batting on after being knocked unconscious by a ball the first time he batted at the WACA. No one talked about this being an inappropriate thing to say, possibly because cricket has fewer incidences of concussion but – more pragmatically – because coaches are fair game for criticism when they are losing but almost immune to criticism when they are winning. Darren Lehmann has just presided over a 3-0 Ashes win for Australia that – like AVB getting the sack – would have seemed impossible a month ago.

The ‘must win’ culture for coaches is unfair

The deal which coaches get – “win or else” – is unfair, but all in sport need to understand this deal and then question whether those under such ridiculous pressure to win should have the health and welfare of players in their hands? How can AVB be asked to think about Hugo Lloris’ health 20 years hence when the coach might only be in the job another month? Given the manager is NOT well placed to consider a player’s long-term health, how are sports administrators redressing the imbalance of power on match day between the coaching and medical staff? Note that the NFL paid out close to 800 million $US to football players who felt their long-term health was not a club priority during their playing years. NHL players are now seeking a similar payout. (Of course the NFL did not acknowledge ‘guilt’ of any kind).

Is there time for doctors to make an accurate concussion diagnosis pitchside?

A further development from my Blog, but relating to a different game (i.e. NOT the Tottenham doctors) is that a team doctor who has been accused – by the press – of allowing a concussed player to stay on the field wrote to say that in the incident in question he didn’t believe the player to have been concussed (despite what the press wrote). He admitted that the rules of football meant that his assessment was unfortunately a brief one on the pitch and that he supported a rule where he could more thoroughly assess the player on the sideline. Rugby Union has introduced Pitchside Concussion Medical Assessment. Doctors are thus under conflicting pressure from their teams (to err on the side of leaving the player on the field) and their medical colleagues (to err on the side of taking the player off, permanently if this is all the rules allow). It is a hard time to be a team doctor.

Legislate to be allow doctors to make additional concussion assessments and require player substitution as needed.

The bottom line is that in almost every professional sport the decisions on which players to substitute are primarily controlled by coaching staff who are forced by the nature of the job to think in the short term. Witness AVB’s sacking. Doctors are in a position to think longer term with respect to a player’s health later in life. The rules of sport need to change to allow doctors to have the power to make (additional) assessments and substitutes in the case of potentially concussed players.


Dr. John Orchard @DrJohnOrchard is an Australian sports physician, injury prevention researcher, Cricket NSW doctor, and BJSM Associate Editor.


7 Nov, 13 | by Karim Khan

(A full version of the Daily Mail publication, page 75, November 7). The Daily Mail @DailyMailUK is doing a tremendous service to improve concussion awareness and player management. Kudos Daily Mail.

Now that the dust (if not Hugo Lloris’ scrambled brain) has settled on the Spurs keeper’s knock to the head on Sunday, let’s review the situation and ask what we can learn. Let’s remember that the focus must be on what is best for this player, and sportsmen and women the world over


  • Lloris was clearly concussed. He was knocked out, was wobbly on his feet and in his manager’s own words after the game “Hugo still doesn’t recall everything about the incident”
  • The Spurs doctor (who is highly regarded and was commended for his work on resuscitating Fabrice Muamba) wanted the player removed from the pitch. I have looked at the TV coverage numerous times and he clearly signals that the player should go off. In fact Villas-Boas admitted as much after the match when he said “the medical department was giving me signs that the player couldn’t carry on because he couldn’t remember where he was” and that “he went against medical guidelines to keep the goalkeeper on the pitch”.
  • The decision to keep the player on the pitch was solely the Managers. He admitted that after the game “’It was my call to delay the substitution, you have to make a decision in situations like this”
  • Loris had a CT scan performed after the game and the club’s website said “The Club can confirm that Hugo Lloris underwent a precautionary CT scan and was given the all-clear and travelled back to London last night”
  • A CT scan is performed to rule out more serious head injury. It cannot exclude concussion which probably explains why the Spurs statement said given “the all-clear” which in reality was from serious head injury but the media interpreted as from concussion
  • Despite the clear cut evidence of concussion and the Manager’s admission that he was responsible for the decision, Spurs changed their tune the following day presumably on advice from the club’s PR department (otherwise known as the “Protect the manager at all cost department”)
  • Their Head of Sports Medicine, physiotherapist Wayne Diesel was quoted as saying “Once the relevant tests and assessments were carried out we were totally satisfied that he was fit to continue playing.”
  • Spurs have a Europa League game on Thursday, 5 days after the Everton game
  • The most recent World Concussion meeting was co-sponsored by FIFA and held at the FIFA headquarters in Zurich last November. The Consensus Statement from that conference published in March this year is quite clear on management guidelines for concussion
  • Regarding return to play (RTP) on the same day, it states it was unanimously agreed that no RTP on the day of concussive injury should occur”.
  • Regarding a graduated RTP following concussion “RTP protocol following a concussion follows a stepwise process as outlined in table 1. With this stepwise progression, the athlete should continue to proceed to the next level if asymptomatic at the current level. Generally, each step should take 24 h so that an athlete would take approximately 1 week to proceed through the full rehabilitation protocol once they are asymptomatic at rest and with provocative exercise. If any post-concussion symptoms occur while in the stepwise program, then the patient should drop back to the previous asymptomatic level and try to progress again after a further 24 h period of rest has passed”.





  • Following the recommended protocol, the minimum time before returning to play is 6 days assuming that the player is totally asymptomatic (no headaches, nausea, “foggy” feeling etc) the day after the incident and right through the rehabilitation
  • Most Premier League clubs would also perform a computerised neuropychological test at the end of the rehabilitation process to confirm full recovery
  • The Spurs Manager explained his decision to over-rule the club doctor on this basis “I made the call to keep him on the pitch because of the signs he was giving. When you see this kind of assertiveness from the player it means that he is able to carry on. He was determined to continue and looked concentrated, driven and focused enough for me not to make the call to replace him. The saves he made after the incident proved that right”.
  • The Manager was overly influenced by the player, rather than the expert medical opinion. The fact that Lloris made some good saves after continuing is not relevant, as it is the long term effects of playing concussed that are a concern. There are plenty of historical precedents for players playing quite effectively immediately after a concussion. It does not justify the decision.
  • To be fair to the Manager, the medical profession’s stand on the management of concussion has changed over the past few years and the Manager may not be aware of this
  • Previously concussion was thought to be a self-limiting relatively benign condition. In the past few years there is increasing evidence of long term brain problems in retired footballers. Most of the research has come from the NFL who recently settled a lawsuit form a large group of retired players for $750 million (without admitting any guilt).
  • Clearly we as a profession have not succeeded in educating football club managers as to the change of attitude and the new protocols
  • Football in the UK would be wise to follow the lead of the English Rugby Union who have summoned all their coaches to Twickenham this Thursday to hear the latest on the management of concussion.
  • When a player is suspected of being concussed, he should be immediately removed from the field of play and assessed to determine whether he indeed has concussion. This assessment, which should ideally be done in the medical room, takes approximately 5 minutes. See SCAT3 (Free). As a result the other football codes have introduced a temporary substitution which can be made while the player is being assessed. In rugby this is a 5 minute period, in Aussie Rules football it is 20 minutes. Soccer needs to consider something similar.

What now?

The short term dilemma for Spurs is whether Lloris plays tonight (Thursday night). They have put themselves in a difficult position. If he plays, then in addition to their breach of concussion protocol on Sunday, they will be breaching the RTP protocol which requires a minimum of 6 days graduated rehabilitation

  • If they rule him out, then they are admitting that he was concussed and that they were wrong to allow him to continue playing
  • If that PR department had been doing its job they would have said that they had always planned to play their No 2 goalkeeper on Thursday and got out of it that way!!
  • The team doctor has the expert knowledge and is the one person who has the player’s health as his/her primary responsibility and therefore should be the sole arbiter of whether a player is concussed.

[BJSM Editor’s note: Credit to Tottenham for clearly following the Zurich Concussion Guidelines here: The Manager is quoted as resting Lloris as a result of Sunday’s concussion. For non-expert readers, when Lloris returns to play should depend on his symptoms (and potentially neuropsychological tests, not a specific ‘time’ . One week is a minimum to progress through the stages (above) but it can take longer if symptoms (headache, unusual tiredness, dizziness) persist. ] Posted on Thursday Nov 7th after the Europa Cup game.

How should this have been handled?

  • It would have been nice yesterday instead of Spurs trying to shift the blame to their (absolutely innocent) medical staff, to hear the Manager publicly state that he had made a mistake, that he was not up-to-date regarding the changes in guidelines for the management of concussion, state his total support for the club’s medical team, and state clearly that he will not interfere in the future.
  • That would have made a positive out of a negative.
  • Instead Villas-Boas has come out and abused those of us who have expressed concern calling us “incompetent”.
  • Sadly he had missed a wonderful opportunity to get the message out there that concussion must be taken seriously.



A responsible key-holder: is it time for legislation on medical provision for athletes at the sports ground?

9 Jul, 13 | by Karim Khan

Sport and Exercise Medicine: The UK trainee perspective (A BJSM blog series)

By Stefan Kluzek and Rose Penfold

Rx bottlesUK sports grounds typically have a stock of emergency medications, supplied and prescribed by a club doctor.  Recently, a physiotherapy colleague told me that Professional Sport Association pitch- side trainers informed her that medications, especially emergency medications, should always be readily accessible, especially when the doctor is not available. Since this raises issues regarding professional standards and duty of care, I resolved to investigate current legal frameworks protecting both medical personnel and athlete in the athletic arena.

Preliminary research highlights that there is no specific guidance on medical provision at sports grounds, especially during training.  I proceeded to contact the Care Quality Commission (CQC), Sports Grounds Safety Authority (SGSA), UK Anti-Doping (UKAD), General Medical Council (GMC) and the Medical Protection Society (MPS). The Health and Social Care Act 2008, which established the CQC, does not extend to cover sports training facilities.  The SGSA, whilst playing a critical role in overseeing the safety of spectators, has no remit concerning athletes. UKAD does not have a policy on this and merely recommends that doctors follow existing guidance regarding access to controlled substances. The GMC have not published any specific guidelines nor (as far as I could determine) articles; however, they suggest that similar rules apply to medications kept at any facility and referred me to their following core guidance1:

  • Good medical practice (2013), paragraph 12: “You must…follow the law, our guidance and other regulations relevant to your work.”
  • Good practice in prescribing and managing medicines and devices (2013), paragraph 6:  “You must maintain and develop knowledge and skills in…prescribing and medicines management relevant to your role.”
  • And paragraph 51: “…you must make sure that arrangements are in place for monitoring, follow-up and review, taking account of patients’ needs and any risks arising from the medicines.”

Assuming sports grounds are no different to other healthcare facilities, if we do not uphold these standards we risk compromising the quality of pitch-side care. Security and record-keeping should be monitored to ensure all drugs are in-date, correctly stored and appropriately administered by qualified professionals. A risk assessment for the sports ground environment should be conducted every time one supplies new medication. The CQC have published on-line a self-assessment tool for primary care providers and acute NHS trusts or independent hospitals – perhaps a means by which club doctors can monitor their current practice2.

Legal frameworks do exist allowing physiotherapists to prescribe medication: namely, Patient Group Directions (PGDs), Patient Specific Directions and Supplementary Prescribing. These legislations are “reserved solely for situations where they offer an advantage for patient care (without compromising safety), and where it is consistent with appropriate professional relationships and accountability”. Whether this applies here is controversial. PGDs, for instance, cannot be written in the private healthcare sector unless the organisation is CQC-registered. Supplementary prescribing is undertaken in NHS environments where a multidisciplinary team develops a Therapeutic Care Plan for a patient; again, not really applicable here. Interestingly, last year the Department of Health published a Summary of Public Consultation on Proposals to Introduce Independent Prescribing by Physiotherapists3 – of 689 total responses, 99% were in support. If there is a move towards independent prescribing by physiotherapists in the future, this may resolve the dilemma as responsibility for prescription and use of medication will fall to each individual.

To conclude, this remains a legislative grey area and, from personal experience, this extends into clinical practice. Whilst it may not be a problem for Premiership clubs (where medically-qualified personnel are always present), lower-division clubs may find themselves in difficult situations. Approved and accredited pitch-side trauma courses have been improving pitch-side standards of emergency care in recent times. It seems logical, therefore, that adequately trained and competency-assessed individuals should have access to medications to use in the situations they have been trained for. By law4, employers are obligated to protect the health, safety and welfare of employees; this includes athletes in their occupational environment. As a group, are we prepared to self-regulate by enforcing the standards of other healthcare facilities, or is it time for professional bodies to produce some official, pragmatic guidelines? Ultimately, protection of athlete safety at the sports ground is our moral and professional responsibility.

I eagerly await opinions on this issue.


1General Medical Council

2 Care Quality Commission

3Department of Health

4Health and Safety Executive, Health and Safety Law


Dr Stefan Kluzek MRCP (UK) MRCP(London) DipSEM is a Sport and Exercise Medicine Registrar in the Oxford Deanery.  He is a Clinical Research Fellow in Sport and Osteoarthritis at Oxford University.  He is also the Medical Officer to Oxford University Rugby FC and Medical Officer at Oxford RL.

Rose Penfold is a 4th year Medical Student at Oxford University.

Dr James Thing co-ordinates “Sport and Exercise Medicine: The UK trainee perspective” monthly blog series.

Applicability of muscle injury classification system in The Munich Consensus Statement? (Letter to the Editor)

31 Jan, 13 | by Karim Khan

 By Drs. Del Buono, Best and Maffulli 

Letter to the Editor

In response to: Terminology and classification of muscle injuries in sport: The Munich consensus statement. Mueller-Wohlfahrt HW, Haensel L, Mithoefer K, Ekstrand J, English B, McNally S, Orchard J, van Dijk CN, Kerkhoffs GM, Schamasch P, Blottner D, Swaerd L, Goedhart E, Ueblacker P.  Br J Sports Med, 2012.

 Dear Prof Khan,muscles of the back


“The soul in darkness sins, but the real sinner is he who caused the darkness”

Victor Hugo, Les Miserables


We commend the authors for their efforts to provide a much needed new classification of muscle injuries in their article Terminology and classification of muscle injuries in sport: The Munich consensus statement (open access); the fact that highly experienced clinicians became involved in this field increases the potential impact upon both daily practice and future research fields 1. However, we wish to express our concerns regarding the actual applicability of this proposed classification system in clinical practice, especially for what concerns the translation to management of these injuries.

Notably, criticizing traditional terminology, as ‘passè and confusing’, the authors highlight the need for a more standardized definition and universal classification that reflects both functional and structural features of muscle injuries. We respectfully point out that the term ‘functional’ is not well defined, and it is used with various ambiguous and at times inconsistent meanings, such as fatigue-induced muscle disorder or delayed-onset muscle soreness (DOMS). We underline that both functional and structural disorders may lead to functional limitations in athletes, and these latter may also hide misunderstood structural changes of the muscle. Therefore, from a diagnostic view point, both the terminology and the ability to distinguish these two entities pose challenges. For structural changes, the authors also suggest that the term “tear” better reflects the structure of the muscle, recommending not to use the term “strain”. The latter, conversely, implies the biomechanics of the injury 1. We point out that the term “strain” reflects some radiological features (MRI and US). In fact, a strain is defined as a Grade I injury, in which less than 5 % of muscle fibers are disrupted, with a feathery oedema-like pattern, and intramuscular high signal on the fluid-sensitive sequences at MRI. This condition is well differentiated from grade II (partial tear) and III (complete tear) lesions 2. The proposed distinction is further complicated by the fact that it is unlikely that MRI is sensitive enough to detect the presence of microscopic disruptions, which may be decisive in differentiating ‘functional’ from ‘structural’ muscle injuries.

We recently proposed that a novel anatomic system was to classify acute muscle strains 2. Realistically, we suggest that the proper identification and description of the injury site could be prognostic for muscle recovery. Arising from the traditional imaging classification (MRI and US), we have simply considered the anatomy of the muscle, and classified the lesion as type I, when involving the proximal MTJ, type 2, for muscle belly injuries, and type 3, when the distal MTJ is torn. We accept that this classification system relies on clinical findings but also MRI (or ultrasound) scans. Specifically, to subcategorize muscle belly injuries, both coronal and axial views have to be taken into account. On coronal and sagittal imaging scans, the muscle belly may be injured proximally, in the middle, or distally. On axial sequences, we defined the injuries as intramuscular, myofascial, myofascial/perifascial, myotendinous, and combined.

Therefore, different from the classification system presented, which describes structural and functional muscle disorders, we propose a novel classification based on anatomical and imaging features: these aspects have a higher impact on diagnosis and management of these injuries. Imaging (US and MRI) assessment is not only helpful to help management 3, but it would also be used for assessment of injury severity and to predict the time of return to sport activity 2. Both classification systems likely have strengths and limitations, perhaps the next step is the validation of these systems and their ability to predict convalescence from sport and time for return to play.



1          Mueller-Wohlfahrt HW, Haensel L, Mithoefer K, Ekstrand J, English B, McNally S, Orchard J, van Dijk CN, Kerkhoffs GM, Schamasch P, Blottner D, Swaerd L, Goedhart E, Ueblacker P. Terminology and classification of muscle injuries in sport: The Munich consensus statement. Br J Sports Med In press.

2          Chan O, Del Buono A, Best TM, Maffulli N. Acute muscle strain injuries: a proposed new classification system. Knee Surg Sports Traumatol Arthrosc 2012;20:2356-2362.

3          Boutin RD, Fritz RC, Steinbach LS. Imaging of sports-related muscle injuries. Radiol Clin North Am 2002; 40: 333-362.


Angelo Del Buono 1, Thomas M. Best 2, Nicola Maffulli 3

1 Department of Orthopaedic and Trauma Surgery, Campus Biomedico University, Via Alvaro del Portillo, 200, 00128 Trigoria, Rome, Italy

2 Division of Sports Medicine, Department of Family Medicine The OSU Sports Medicine Center, The Ohio State University, Columbus, OH, USA

3 Centre for Sports and Exercise Medicine, Barts and The London School of Medicine and Dentistry, Mile End Hospital, 275 Bancroft Road, London E1 4DG, England.



BJSM Social Media Contributes to Health Policy Re-Think: A success story in Hertfordshire

2 Oct, 12 | by Karim Khan

By Christine Neyndorff and Dr. Richard Weiler

There was an unprecedented response following our last BJSM blog to the draft consultation for the Hertfordshire Health & Wellbeing Board strategy for the coming years.

The remarkable response following this BJSM blog deserves an update (thanks Ed.). We also wanted to thank everyone for their contributions.

The original draft strategy for health and wellbeing included 8 priorities – increasing levels of physical activity was not one of them.

98.4% of BJSM online poll voters felt that Hertfordshire should tackle the problem of physical inactivity as a top priority for health and wellbeing.

The formal draft consultation response feedback:

  • 2747 responses were received (we are told this demonstrates an unprecedented level of interest)
    • 165 responses received from groups or organisations
    • 2582 from individuals

The majority of responses felt that there should be greater reference to healthy lifestyle. In particular, most of these recommended that physical activity be included as its own priority.

“Promoting physical activity for all ages should be a priority” East and North Hertfordshire NHS Trust

“Increasing levels of physical activity has many benefits, and can contribute to several of the proposed priorities…District and Borough Councils have a major role in providing and promoting physical activities, and could make a big contribution to delivery of a priority centred on this issue” Broxbourne Borough Council.

In addition to the written consultation there was also considerable debate on twitter about Hertfordshire’s public health strategy.

Moving forward…the draft strategy is being revised taking this feedback into account.  We understand the Health & Wellbeing board is considering including the following priority in the revised strategy,  “Increasing physical activity and promoting a healthy weight.” 

[Despite 83% of BMJ readers believing that the public health evidence still supports cardiorespiratory fitness gained through physical activity over weight loss (see poll results here), we would be very satisfied with this inclusion demonstrating evidence based progress in recognition of the importance of physical activity in health and wellbeing.]

The final draft of the strategy will be considered by the Shadow Health & Wellbeing board in January 2013.

It is vital that this joint priority remains included and further correspondence may help to achieve this.

Please consider sending any supporting comments to Fiona Deans who will be able to collate them and forward them to the County Council.



Thanks again for you ongoing support,

Christine Neyndorff,  Director of HSP and Dr. Richard Weiler, “HSP Physical Activity Ambassador”

The Herts Sports and Physical Activity Partnership, known as HSP, was established in 2003 and our vision has remained simple and constant “Working together to encourage more people to be more active more often” . 

HSP is a voluntary organisation with a strategic board including representatives of local agencies and partners committed to working together to increase participation in sport and physical activity and promote the many benefits. The Board is supported by a small team of staff providing leadership and co-ordination to work more effectively together at county level.

CT scans and X-rays increase risk of cancer – changing the goal posts in sports medicine

19 Sep, 12 | by Karim Khan

By John Orchard and Jessica Orchard

Two jaw-dropping papers from The Lancet 1 and BMJ 2 published in the past month should have a major effect on the practice of sports medicine.  They have clearly demonstrated that radiating scans in young people actually do lead to an increased risk of cancer later in life. Perhaps until 2012 this was a theoretical risk, but as of the publication of these landmark papers 1-3 we can be certain that the increased risk is not zero. There will be much more to come in this field over the next few years and it will dramatically change the landscape of radiology and all medical practice.

Pearce and colleagues’ study in The Lancet looked at the excess risk of leukaemia and brain tumours for children and young people exposed to CT scans. They found that children exposed to cumulative doses of 50mGy in CT scans may have triple the risk of leukaemia, and doses of 60mGy may have almost triple the risk of brain tumours1. Though this appears to be a massive increase in risk, the authors point out that these cancers are still relatively rare, causing an estimated one excess case of leukaemia and one excess brain tumour per 10,000 head CT scans. They are clearly cause for concern, as indicated by the fact that 12 other groups from 15 countries are studying the risk of scans on children3.

These Lancet findings are more striking when combined with the findings of Pijpe and colleagues’ GEN-RAD-RISK paper published last week in the BMJ2. This study showed that when women who carry a specific mutation associated with breast cancer (BRCA1/2), and who  were exposed to diagnostic radiation before the age of 30, had almost twice the risk of breast cancer (with a dose-response pattern). This study involved lower doses which we have previously considered fairly ‘safe’ (e.g. 4mGy from a single mammogram or shoulder x-ray). Therefore, BRCA1/2 carriers, with an already increased risk of a very common cancer, would be particularly at risk from exposure from radiating scans at a young age.

Why does this matter for sports and exercise medicine?

Sports and exercise medicine is a field in which most patients have many years of life expectancy remaining; it is also a field in which diagnostic imaging is very common. Imaging is often confined to the limbs but also involves the spine.  Importantly,  the GEN-RAD-RISK paper showed, for example, that shoulder X-rays in women with the BRCA1/2 mutation can increase the risk of breast cancer. This does not prove that a shoulder X-ray is unsafe for the entire female population, but because it is quite plausible, we need to reassess the use of radiating scans. The authors of this study have already recommended that women with the BRCA1/2 mutations should not get mammograms and it is hard to see how this recommendation will not soon be extended to all younger women, as mammograms are meant to be preventing deaths from cancer, not causing them.

Studies have not been published to look at, for example, the risk of  cancers in the abdominal cavity (e.g. bowel, ovarian) after lumbar spine CT scan, but again we have to presume from the existing knowledge that the increased cancer risk is not zero. In this case,  the unknown is the size of the increase in cancer risk (and not whether there actually is one). All tests (and treatments) in medicine need to consider benefits, risks and costs. On the benefit side, the test which gives the best information relevant to management needs to be identified. This can’t be done in isolation of the increased cancer risk of radiating scans, particularly in young or middle aged patients.

There will still be cases where a test that involves radiation is going to give preferred information to a non-radiating one – a classic example being in the knee of a 70 year old, where X-ray will tell what needs to be known in 95% of cases and MRI scan is generally an excessive use of imaging. However in scenarios where we used to recommend radiating tests (e.g. CT and bone scan to investigate for suspected pars stress fracture in an adolescent) we may need to quickly change to a recommendation of first line MRI scan to avoid increasing the risk of cancer. Health systems are going to need to change in scenarios where radiating tests are funded but non-radiating tests aren’t, because clearly it would raise ethical questions for a health system to be funding (offering a financial incentive) to have a test which can increase a patient’s risk of cancer when a non-risky test is available but unfunded.

Up to fifty years ago, some shoe stores used to perform X-rays on the spot to show whether a kid’s shoe was fitting well4 – this practice is now considered archaic.  Sensibly there is now an attitude in medicine that a pregnant woman should not receive an X-ray or CT scan if the information could be obtained in any other fashion. We are probably heading into an era where the same attitude needs to apply to all children and young people, for CT scan and even X-ray. Modalities such as MRI and ultrasound (and good old-fashioned clinical examination) will need to become more prominent in sports and exercise medicine, at the expense of radiating examinations. These studies highlighted in the blog will generate a demand for consensus meetings involving sports physicians, radiologists, radiation physicists, and epidemiologists among others to provide guidance for clinicians, professional bodies and patients. Depending on the recommendations made at consensus meetings, there should ideally also be a review of government/insurance funding arrangements to remove any financial incentives towards the inappropriate use of radiating scans.


John Orchard is an Australian sports physician who has worked with numerous professional team sports. His sometimes controversial views are personal and not necessarily representative of organisations he is affiliated with. You can read more at www.johnorchard.comand/or follow @DrJohnOrchard on Twitter

Jessica Orchard is an Australian lawyer with qualifications in economics and public health, currently employed at the NSW Cancer Council. Her views in blogs are also personal and not necessarily representative of her affiliated organisations.


  1. Pearce MS et al. Radiation exposure from CT scans in childhood and subsequent risk of leukaemia and brain tumours: a retrospective cohort study. Lancet. 2012 Aug 4;380(9840):499-505.
  2. Pijpe A at al. Exposure to diagnostic radiation and risk of breast cancer among carriers of BRCA1/2 mutations:  retrospective cohort study (GENE-RAD-RISK).BMJ. 2012 Sep 6;345:e5660.
  3. Einstein AJ. Beyond the bombs: cancer risks of low-dose medical radiation. Lancet. 2012 Aug 4;380(9840):455-7.
  4. Bowden T. Frying one’s gonads for shoes.

Return to play decision making – Reducing clinician ‘personality bias’ with a call for objective clinical testing

12 Sep, 12 | by Karim Khan

Guest blog by Phil Coles (@PhilColesPhysio)

Making the decision of when an athlete should return to play after an injury is one of the most challenging parts of a sports clinician’s role. This is especially so when working with professional sporting teams, where the pressures can be immense. Ideally, a clear decision making process should be combined with reliable clinical objective markers to reduce the potential for the ‘personality bias’ of the clinician leading to error in these decisions.

Being aware of personality bias

Rehabilitators working with elite athletes may have their own ‘personality bias’ that can expose them to the risk of two opposing yet equally significant errors.

On the one hand, the clinician may tend to be overly aggressive. This could be an internal compulsion (the ‘Gambler’ clinician) or may be the result of external pressures leading them to rush a player back in to competition before it is reasonably safe for them to do so (the ‘Weak’ clinician).  Premature return to competition can lead to athletes breaking down with re-injury or simply performing below expectations. If however, an injury does recur in the early stages of a return to competition, then it (perhaps reasonably) exposes that clinician to direct blame for a poor outcome. Any poor performance related to physical deficits may also negatively affect the clinician’s relationship with the player and or their coach / manager. Both of these outcomes may in fact put that clinician’s career at a club in jeopardy, and this is a fact of which most clinicians are well aware.

On the other hand, the clinician may tend towards to being overly conservative. This may also be due to internal compulsion (the ‘Conservative’ clinician) or because of fear of the consequences described above (the ‘Cynical’ clinician). If athletes are kept out for longer than necessary to reduce the risk they might break down or perform below expectations on their return, it will mean they miss valuable competition time. This second type of error is not as immediately obvious to the coach / athlete and therefore it is less likely to bring direct blame to the clinician. Naturally if there is a pattern of consistently delayed recovery over a long period of time then it may reflect poorly on those involved. However, it is much more difficult to blame them directly, as it is never really clear as to when any individual athlete could have returned from a particular injury.

Perhaps contrary to expectations then, it is likely many clinicians default to the position of being overly conservative. Unfortunately this means some will make a conscious choice to be overly conservative, not in the best interests of the player or the team, but rather in the hope of reducing any risk of being held liable for the more ‘obvious’ poor result (re-injury or poor performance).

What is the cost of ‘delayed’ return to play?

To highlight the true cost of unnecessary conservatism from the cynical clinician to a club, consider the following example. If a football team that plays once a week was to have 30 injuries over the course of a season and all those injuries were given just one extra week of rehabilitation more than was really necessary, that would cost the club a total of an extra 30 missed games. Consider then if all those players actually came back 1 week earlier as perhaps many of them could, but this caused 5 players (17% of injuries) to re-injure the same area. In the case of each of those 5 re-injuries, if the athlete missed a further 4 weeks, the club would lose players for an extra 20 games. This means that despite those recurrences, the club would have cut their total number of games lost to injury by 30% (from 30 to 20 games lost). Dr John Orchard (@DrJohnOrchard) made this point in his review of injuries and re-injuries in AFL in 2005. From his analysis at the time he concluded ‘at this stage it may be a sensible strategy to allow earlier return to play in team sports and accept a low-moderate re-injury rate’ after having seen such a pattern in the reported data.

In reality, of course, there any many modifying factors that would need to be considered in each individual case. For example, is there extra risk of recurrence with an earlier return for that particular injury, and how long recovery likely will take after re-injury?  Also, other factors such as the point of the season, the particular game, and the position being played etc. will influence whether or not a risk is worth contemplating for an individual athlete at a particular time.

What happens in real life?

The reality is that clinicians working with high level athletes must recognise that it may be equally as negative to have a bias towards conservatism as it is to have bias towards aggression in rehabilitation. Although most clinicians working in an elite environment would probably deny that they ever knowingly act overly conservatively, in reality most would (if being honest) admit there have been times when they have taken longer to return a player to competition than was perhaps essential because they feared the repercussions of any re-injury. Conversely, most would also accept that there have probably been times when they allowed issues that don’t directly relate to the injury into the thought process that ultimately allowed a player to return prematurely. Sports clinicians must be brave in the sense that they must be able to withstand the outside forces which might encourage a rushed return to play, but they must be equally brave in backing their own ability and judgement in getting a player back when the relative risk is reasonable, rather than waiting for the risk to be nil, which of course it will never actually be.

The judgement of what is a reasonable risk is where the real skill of a sports rehabilitator lies. The ability to make this judgment correctly in a more consistent fashion, relies firstly utilising a clearly defined decision making process. The actual process of that return to play decision making was well outlined by Matheson et al (2011). They described a thorough model for considering all the factors that may affect our clinical judgement when deciding on the return to play of an athlete. The first two steps are to evaluate the athlete and the risk of returning to sport. This involves assessing the health status of the athlete and then considering that against risks particular to that sport and in that athlete. This is where improved objective markers would be particularly useful. Having decided that someone may return based on these principles, they acknowledge that there are still many ‘modifiers’ to your final decision which must be considered, and so ultimately clinical reasoning remains paramount. These modifiers would include consideration of issues such as the timing and season, the stage of an athlete’s career, the importance of athlete to the team, the importance of a particular game to the athlete, any conflicts of interest at play (such as financial reward to the player or therapist), any chance of masking occurring, and risks of litigation etc.

What objective measures are there?

The ability to make return to play decisions objectively will help to decrease the potential of clinician personality bias to lead to error. For this reason I contend that developing improved objective markers that may predict a safe return to play is perhaps the greatest research need for rehabilitators working in high level sport.

Unfortunately there are not yet many proven objective markers for sport specific return to play, but there are certainly some clinical tests that may be considered to reliably assess for known risk factors to injuries. Consider these examples.

(1)The Hamstring active flexibility and apprehension test developed by Asking et al (2010) is a reliable test which is more sensitive to picking up on-going Hamstring deficit than traditional assessment methods. (Click here to listen to a podcast with Carl Askling about hamstring management). Considering that hamstring recurrences are such a problem in the football codes it would be reasonable to suggest a normalisation on that test along with all other traditional clinical signs is essential before endorsing a return to play.

(2) A decrease in adduction power as measured by a squeeze test may predict the onset of groin pain in AFL players (Crow 2010). Perhaps therefore after any groin injury a reasonable objective milestone that must be met during rehabilitation before being allowed to progress to full loading is that an athlete must have reached at least their pre-morbidity levels on that squeeze test. (Per Holmich’s podcast on groin pain is here; his short YouTube video is here).

In summary, to clear an athlete to return to play there needs to be confidence that the rehabilitation has been complete, and that a clear decision making process was followed. You must be aware of the dangers of ‘personality bias’ among clinicians and we should attempt to minimise this through the use of objective clinical testing wherever possible. Perfect judgement is impossible but clinicians and managers should appreciate that being overly conservative can be an equally significant and perhaps more common error as being overly aggressive. They should also accept that using objective markers is the way to minimise this. If the current markers fail us or do not exist in the sport specific detail we would like, it does not mean we should shy away from using objective markers, it means we should dedicate time to developing more accurate ones.


Orchard et al. Return to Play following muscle strains. Clin J Sport Med. 2005;15:436–441.

Matheson et al. Return to play decisions- are they the team physician’s responsibility? Clin J Sport Med 2011;21:25–30.

Crow et al. Hip adductor muscle strength is reduced preceding and during the onset of groin pain in elite junior Australian football players. Journal of Science and Medicine in Sport. 13 2010; 202–204.

Askling et al . A new hamstring test to complement the common clinical examination before return to sport after injury. Knee Surg Sports Traumatol Arthrosc. 2010;18:1798803.

Stay tuned for the BJSM podcast on this topic with Liverpool FC’s Darren Burgess 


For the past 2.5 years Phil Coles has been head of Physical Therapies at Liverpool Football club. Prior to this he was head physiotherapist for the Socceroos including the 2010 World Cup, and an associate lecturer at the University of Sydney. You can follow Phil on twitter @PhilColesPhysio.


Closing soft tissue wounds rapidly at pitchside – A role for metal skin staples without anaesthesia?

17 Aug, 12 | by Karim Khan

Guest Blog by Drs Christopher Fowell and Phillip Earl

*Please also see the commentary from @DrJohnOrchard on this blog here

* You  can vote on whether you think metal staples are a good idea via the BJSM ‘Current Issue’ page here:


You are working pitchside and your player receives a facial laceration. What are your options for fast wound closure? Are metal staples an option?


Head and facial soft tissue injuries occur frequently during contact sports, especially those in which headgear is not routinely used.  Professional sport is continuous in its nature; hence a prompt return to competitive action is usually desired when rules require a bleeding player to leave the field.

Different methods of wound closure have been described and reviewed extensively in both the surgical and traumatic settings1,2.  Although skin staples are superior to sutures in the surgical head and neck setting1, their use has gained little popularity in the sports medicine setting of traumatic wounds to the head and neck.

In this BJSM Blog, we describe a case series of professional and semi-professional footballers sustaining lacerations to the face and scalp. Since 2004, all players returning to action were managed rapidly at the pitchside using metal skin staples, without complication.

Subjects and Methods

Using a physician’s clinical database, we undertook a retrospective cohort study of patients treated for soft tissue lacerations sustained during competitive play between 1987 and 2012.  All injuries were treated by the same clinician.


Sixty-four patients had sustained 64 separate lacerations.  Fifty-seven (89.0%) of lacerations were sustained to the head and face (Table); most commonly to the scalp and supra-orbital tissues.

Table.  Site of lacerations.


Number (%)

Head and Face

57 (89.0)

   – Head / scalp

20 (31.3)

   – Periorbital

18 (28.1)

   – Other face

7 (10.9)

   – Intra-oral

4 (6.3)

   – Lip

4 (6.3)

   – Ear

3 (4.7)

   – Tongue

1 (1.6)

Other (all lower limb)

7 (10.9)


Prior to 2004, all patients were treated under local anaesthetic, using a combination of resorbable, synthetic, braided (Vicryl®, Ethicon Inc.) sutures and a non-resorbable, synthetic, monofilament (Ethilon®, Ethicon Inc.) sutures for wound closure.  Since 2004, 11 patients who were planning to return immediately to play had lacerations to the scalp, forehead and supra-orbital rim closed using metal skin staples (Weck Visistat® 35W disposable skin stapler, Teleflex Medical).   No local anaesthetic was infiltrated prior to skin closure for these patients.  Patients withdrawn from the match were managed using sutures as previously.


Soft tissue injuries occur frequently during athletic activities3, ranging from lacerations to superficial abrasions. Following any orofacial injury, initial structured assessment using the ABCDE approach of Advanced Trauma Life Support is required.  Once significant injury has been excluded or treated appropriately, prompt, thorough debridement and surgical repair of lacerations is required to restore function and anatomical form.  Players who sustain injuries that cause bleeding are required to leave the field of play immediately in most sports.  Players may return following definitive closure, when the risk of transmission of blood-borne infections decreases.

Surgical staples have been a common method of wound closure in the surgical setting since the 1990s, and are gaining popularity in the traumatic setting.  Do they have a role for pitchside closure of lacerations sustained during sport? Dr John Orchard (@DrJohnOrchard) reported a case of a patient sustaining an eyebrow laceration during an Australian rugby league game in 20044.  He closed the wound directly on the touchline using metal skin staples, a technique he reported having used previously. The case courted controversy due to the televised nature of the game.

An overwhelming advantage of skin staples is very speedy wound closure. That stapling is faster than suturing has been reported in 5  randomised controlled trials1.  Prompt return to competitive play in these circumstances is beneficial to both the individual and the team’s performances.  In the one previous reported case of skin staples used in closing wounds on the touchline, the player returned to play within 80 seconds of injury, and touching the ball within 40 seconds of having the final staple placed.  The reported case and the author’s direct observation of patients tolerating wound closure using staples in the absence of local anaesthetic further decreases the time away from competitive play.

RCT evidence indicates that skin staples have a lower wound infection rate than sutures1.  With regard to cosmetic outcome, studies have shown staples to be comparable to, and in some studies better than sutures at long term follow up.  Previous reports have described the removal of staples at the end of a game, followed by definitive wound closure under local anaesthetic with sutures.  We believe this is unnecessary, assuming the wound was cleaned appropriately and adequate wound edge approximation was achieved at initial closure.

Metal staples are more prominent from a wound than sutures and present a theoretical risk of causing damage to other players.  Wound coverage with a simple soft bandage eliminates this risk.   No players in the study have suffered further injury, or inflicted injury on another competitor, through having staples in situ and returning to competitive action.

Staples are not appropriate for lacerations on all sites of the head and neck.  They are contra-indicated on mucosal areas, and hence should not be used on intra-oral wounds.  Other sites which staples should not be used include the eye-lids, lips, ears and nose.

What about tissue glues?

Tissue adhesives are a effective and rapid method of closing traumatic wounds to the head and face2.  They are comparable with other methods of closure with regards to cosmesis, pain and procedure time, their disadvantage is a their greater rate of wound dehiscence.  Following repair of lacerations on the touchline with sutures or tissue adhesive, further collisions have caused wounds to re-open.  It is felt the extra strength of staples helps to prevent this4.

In summary, metal skin staples are a safe, effective and rapid method of achieving closure of traumatic wounds in the touchline environment.  This allows very prompt return to competitive play following blood injury.


  1. Iavazzo C, Gkegekes I, Vouloumanou EF et al. Sutures versus staples for the management of surgical wounds: a meta-analysis of randomised controlled trials. Am Surg 2011;77(9):1206-1221
  2. Farion KJ, Russell KF, Osmond MH et al. Tissue adhesives for traumatic lacerations in children and adults.  Cochrane database of systematic reviews  2002; Issue 3. DOI:10.1002/14651858.CD003326
  3. Ranalli D, Demas P. Orofacial injuries from sport, preventative measures for sports medicine.  Sports Med 2002;32(7):409-418
  4. Orchard JW. Video illustration of staple gun to rapidly repair on-field head laceration.  Br J Sports Med 2004;38(4):e7

Dr Fowell is at the University Hospitals Coventry & Warwickshire NHS Trust – Oral & Maxillofacial Surgery
University Hospital Clifford Bridge Road Walsgrave, Coventry CV2 2DX, United Kingdom

Dr Earl is at the Worcestershire Acute Hospitals NHS Trust – Oral & Maxillofacial Surgery, Worcester, United Kingdom

No conflicts declared and specifically there was no support for the authors’ research or publication from the makers of Weck Visistat® 35W disposable skin stapler, Teleflex Medical.

BJSM encourages Guest Blogs and these can be submitted directly (ideally with artwork attached as a separate file) to


The legality of Pistorius: why ethics is more relevant than biomechanics. Guest blog @DrJohnOrchard

5 Aug, 12 | by Karim Khan

by @DrJohnOrchard

I’m pleased to see Professor Lippi’s opinion piece on Oscar Pistorius in BJSM’s Online first [1], as it is a very important topic and the BJSM is a very appropriate forum to publish on this debate. Much of the article is a good neutral overview of the parameters of this debate. However I disagree very strongly with some of the conclusions made. In particular this section:

“If we all agree—as we do, indeed—that whatever artificial addition on athlete’s body shall be considered unfair or even illicit (the ban of the bathing suits that enhanced swimmers performance is a paradigmatic case), then, prosthetic technology should follow the same route. Beside the fact that Pistorious’s running performance may be higher, the basic dynamics has been definitely proven to be grossly different from that of intact-limb sprinters, and he should not be allowed to race in the Olympics, whereby his natural field remains the Paralympics.” [1]

Firstly, I don’t think that there is universal agreement that “all artificial aids should be illicit”. What is a running shoe other than an artificial aid? It is simply an artificial aid that everyone is allowed to use (although different brands, which surely have different biomechanics, are allowed and chosen). Equestrians are allowed saddles, cyclists are allowed helmets that reduce drag and footballers can wear studs on their boots to improve grip on grass. Artificial aids are available in many sports and we debate and regulate depending on a combination of scientific argument and consensus opinion. We also debate whether caffeine, pseudoephedrine and salbutamol should be on the banned substances list and sometimes change our minds. Lippi points out that the decision was made to ban ‘fast swimsuits’ as if this was the only decision available, when of course it is easy to envisage a scenario where this decision could have been determined with the opposite outcome and we all just accepted better technology. We accept that modern golf balls and clubs allow the ball to be hit further than previous versions, even though many have made the argument to limit this technology. These are all decisions on artificial aids, not automatic choices where we have only had one option.

I don’t think it is an established ‘fact’ that Pistorius has biomechanical advantages over able-bodied runners which outweigh his disadvantages. Obviously there are respected biomechanics experts who have quantified advantages that he does have, but there remain multiple unknowns with respect to the disadvantages. The counter-argument that Pistorius and his supporters (including myself) make is: you can have as much ‘in vitro’ science as you like, but why do able-bodied runners post faster times in every discipline than amputees using artificial limbs over the same distance? In vitro science is fallible. I imagine that a motivated biomechanist could present an in vitro study suggesting that a running shoe would make you run slower compared to bare feet or a physiologist similarly that women had a theoretical advantage in the marathon than men. You wouldn’t need better science to mount a powerful counter-argument – why don’t barefeet athletes (since Abebe Bakila) win running events or women beat men? If amputee runners were consistently beating able-bodied runners then the science alleging an unfair advantage to Pistorius would have a lot more weight. Let’s face it, science can’t yet tell us whether Nike shoes lead to more injuries than Asics shoes or even lead to faster running (even though we could actually do RCTs on these hypotheses, which is not available in the case of amputee athletes) and we need to be humble about what the limits of scientific analysis are.

If the jury is still out on whether Pistorius has an unfair advantage then he deserves the benefit of the doubt. If he was a completely crazy second tier able-bodied athlete who had cut off his own legs in order to try to improve his times, then you could mount a very good ethical argument that he should be excluded (in order to discourage others from following suit). He is in fact the opposite – one of the most inspirational athletes of all time. Where biomechanics can’t give us a foolproof answer, we need to judge this based on our ethical preferences, just as it was decided to ban fast swimsuits, but to keep caffeine legal. Just as the golf authorities will decide whether or not long putters stay legal or become illicit. Just as we decide whether drug cheats should get a 1, 2, 4 year or life suspension. The key question is “what do we want the Olympics to look like?” We decide that you can’t compete in the Olympic marathon in a wheelchair because we don’t want the Gold, Silver and Bronze medals all going to wheelchair athletes. That is a value judgement. If amputee runners were winning every medal at the Olympics, I would be comfortable with a decision that banned them from the events before we did start to get lunatics chopping their legs off to compete. At the moment we have a single amputee runner (Oscar Pistorius) who is internationally competitive in the able-bodied 400m but nowhere near as fast as Michael Johnson, the world record holder. Do we want this sort of athlete in the Olympics? I can’t comprehend an ethical world where it could be determined, ethically, that LaShawn Merritt could return from a drug suspension in time to compete in the 400m at the Olympics, but that we decided to exclude Pistorius from the same event because we thought he had an unfair advantage that we weren’t comfortable with. I am very relieved that the IOC didn’t exclude him. It has already been shown, however, in the Pistorius case, that it is possible to change the rules (from Pistorius being ineligible in Beijing to eligible in London). The “thin edge of the wedge” argument can be countered with the obvious – if Pistorius, or any other amputee athlete, starts beating world records by huge margins, there is every opportunity to change the rules once again.

Personally I would rank Oscar Pistorius amongst the most significant Olympic athletes of all time, alongside Paavo Nurmi, Jesse Owens, Dawn Fraser, Abebe Bakila, Bob Beamon, Mark Spitz, Nadia Comanici, Cathy Freeman, Steven Redgrave, Michael Phelps and Usain Bolt.

All of these athletes make the list because of the Gold medals performance that they have put in. Pistorius is possibly the only non-Gold medallist who belongs in such an esteemed list. Most importantly I believe he will have a greater impact on the world than any of the other legends, in that he may lead to a completely different vision we have of ‘disability’. I will explore this possibility in my upcoming Dr J. column in Sport Medicine Australia’s magazine Sport Health and co-publish it on the BJSM Blog in the near future.

Lippi G. Pistorious at the Olympics: the saga continues. Br J Sports Med doi:10.1136/bjsports-2012-091545

See also medical student Abhishek Chitnis’ BJSM Blog on this topic. (Retweeted 21 times in first hour it was up)


John Orchard is an Australian sports physician who has worked with numerous professional team sports. His sometimes controversial views are personal and not necessarily representative of organisations he is affiliated with. You can read more at and/or follow @DrJohnOrchard on Twitter


Professor Lippi whose article in Italian can be found here

BJSM blog homepage


A peer review journal for health professionals and researchers in sport and exercise medicine. Visit site

Creative Comms logo

Latest from British Journal of Sports Medicine

Latest from British Journal of Sports Medicine