Article Text

Download PDFPDF

Illnesses during 11 international athletics championships between 2009 and 2017: incidence, characteristics and sex-specific and discipline-specific differences
  1. Pascal Edouard1,2,3,4,5,
  2. Astrid Junge6,7,
  3. Marine Sorg2,
  4. Toomas Timpka8,
  5. Pedro Branco5,9
  1. 1 Inter-university Laboratory of Human Movement Science (LIBM EA 7424), University of Lyon, University Jean Monnet, Saint Etienne, France
  2. 2 Department of Clinical and Exercise Physiology, Sports Medicine Unit, University Hospital of Saint-Etienne, Saint-Etienne, France
  3. 3 Medical Commission, French Athletics Federation (FFA), Paris, France
  4. 4 Swiss Olympic Medical Center, Centre de Médecine Du Sport , Divisionde Médecine Physique et Réadaptation, Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland
  5. 5 European Athletics Medical & Anti Doping Commission, European Athletics Association (EAA), Lausanne, Switzerland
  6. 6 Medical School Hamburg (MSH), Hamburg, Germany
  7. 7 Swiss Concussion Center and Schulthess Clinic Zürich, Zurich, Switzerland
  8. 8 Athletics Research Center, Linkoping University, Linkoping, Sweden
  9. 9 Health and ScienceCommission, InternationalAssociation of Athletics Federations (IAAF), Monaco
  1. Correspondence to Dr Pascal Edouard, Department of Clinical and Exercise Physiology, Sports Medicine Unit, IRMIS, Campus Santé Innovations, University Hospital of Saint-Etienne, Saint-Etienne 42 055, France; Pascal.Edouard42{at}


Background Illnesses impair athletes’ participation and performance. The epidemiology of illness in athletics is limited.

Objective To describe the occurrence and characteristics of illnesses during international athletics championships (indoor and outdoor), and to analyse differences with regards to athletes’ sex and participation in explosive and endurance disciplines.

Methods During 11 international championships held between 2009 and 2017, physicians from both national medical teams and the local organising committees reported daily on all athlete illnesses using a standardised report form. Illness frequencies, incidence proportions (IPs) and rates (IRs), and relative risks (RR) with 95% CIs were calculated.

Results During the 59 days of the 11 athletics championships, 546 illnesses were recorded in the 12 594 registered athletes equivalent to IP of 43.4 illnesses per 1000 registered athletes (95% CI 39.8 to 46.9) or IR of 1.2 per 1000 registered athlete days (95% CI 1.1 to 1.2). The most frequently reported illnesses were upper respiratory tract infections (18.7%), exercise-induced fatigue/hypotension/collapse (15.4%) and gastroenteritis (13.2%). No myocardial infarction was recorded. A total of 28.8% of illnesses were expected to lead to time loss from sport. The illness IP was similar in male and female athletes, with few differences in illness characteristics. During outdoor championships, the illness IP was higher in endurance than explosive disciplines (RR=1.87; 95% CI 1.58 to 2.23), with a considerably higher IP of exercise-induced illness in endurance disciplines, but a similar upper respiratory tract infection IP in both discipline groups.

Conclusions Illness prevention strategies during international athletics championships should be focused on the most frequent diagnoses in each discipline group.

  • illness
  • prevention
  • risk factor
  • athlete
  • athletics

Statistics from


Illnesses can impair sports participation and performance. High illness rates (IRs) were reported to be associated with the Norwegian team’s poor performance during the 2006 Winter Olympics.1 To compete in international championships, athletes are predisposed to illnesses through travel, jet lag, changes in climatic and environmental conditions, and changes in food habits,2–4 in addition to the competition per se.5 The incidence proportion (IP) of illnesses per 1000 registered athletes was 72 during the 2012 Summer Olympics6 and the 2010 Winter Olympics,7 54 per 1000 during the 2016 Summer Olympics,8 89 per 1000 during the 2014 Winter Olympics,9 135 per 1000 during the 2010 FIFA World Cup,10 and varied between 71 and 129 per 1000 during the 2009, 2013 and 2015 Fédération Internationale de Natation (FINA) World Aquatics Championships.11 These IPs from different events and over many years highlight the extent of the illness problem during major competitions.

During international athletics championships, illness IPs between 20 and 68 per 1000 registered athletes have been reported, with 17% to 34% of illnesses leading to time loss from sport (4 to 20 time loss illnesses per 1000 registered athletes).2 3 12–14 However, these publications focused mainly on injury epidemiology, and illness was not the subject of extensive descriptive and comparative analyses.2 3 12–14 Only two studies have reported on illnesses during indoor championships.13 14 Data on sex-related differences in illness IP are contradictory.2 3 12 Some studies reported that endurance athletes had a higher illness IP than those in explosive disciplines,2 15 though this result was not confirmed in other studies.3 12 A study of illness epidemiology based on large samples of top-level athletes competing at international competitions is needed and aligns with the health protection objectives of the IOC16 and international federations.17

Therefore, we aimed to: (1) Describe the occurrence and characteristics of illnesses at international athletics championships (indoor and outdoor), and (2) analyse differences with regards to an athletes’ sex and their participation in explosive and endurance disciplines.


Illness data recorded at 11 international athletics championships held between 2009 and 2017 were analysed: 2009,12 20112 and 2013 World Outdoor Championships, 2012,3 2014 and 2016 European Outdoor Championships, 2014 World Indoor Championships, and 2011,13 2013,14 2015 and 2017 European Indoor Championships.

The methods (study design, illness definitions and data collection procedures) were identical for all championships and have been previously described in detail.2 3 12–14 Any illnesses newly occurring during the championships (also including acute/symptomatic episodes of chronic conditions) were reported daily by the national medical teams (physicians or physiotherapists) and/or by the local organising committee’s physicians. Illness data were reported anonymously. Illness was defined as any physical complaint unrelated to an injury and occurring during the championships.2 3 12–14 18 We used the classification system described in the consensus statement for epidemiological studies in athletics that included: (1) Systems affected, (2) main symptoms, (3) causes (without distinction between sudden and gradual onset), (4) severity, (5) disciplines (events) (for details see online  supplementary file).18 Illnesses leading to 1 day or more of absence from sport were defined as time loss illnesses. Disciplines were grouped into explosive disciplines (sprints, hurdles, throws, jumps, combined events) and endurance disciplines (middle and long distance running, marathon, race walking).15

Supplementary file 1

Illness IPs were calculated as the number of illnesses per 1000 registered athletes, and the IRs were calculated as the number of illnesses per 1000 athlete days. The International Association of Athletics Federations (IAAF) or the European Athletics Association provided the list of registered athletes for each championship. The total number of registered athletes was calculated by totalling the athlete registrations at each of the 11 championships, that is, if an athlete registered for more than 1 championship he/she counted for each championship. Athlete days were calculated by multiplying the number of athletes registered for a championship by the number of days of the respective championship.8 19

The data analysis was performed in three steps. Data were first presented as frequencies, IPs and IRs, in total and separately for the indoor and outdoor championships, for female and male athletes, as well as for endurance and explosive disciplines. Thereafter, χ2-tests were computed to analyse differences in the distribution of illness characteristics between categories of athletes (male vs female athletes and explosive vs endurance disciplines). In the final step, the relative risk (RR) of illness with regards to sex and participation in the explosive and endurance disciplines, accounting for the type of championships (indoor or outdoor), were analysed with 95% CIs and application of magnitude-based inferences.20 Magnitude thresholds for RRs were reported as trivial, small, moderate, large and very large for RR values of <1.2, <1.9, <3.0, <5.7 and <19, and the inverse for reductions of >0.9, >0.7, >0.5, >0.3 and >0.1, respectively.20 A Bonferroni correction was used to control for multiple tests. All data were processed using spreadsheet calculations in Microsoft Excel. Significance was accepted at p<0.05.


National medical team participation and response rate, and athlete coverage

On average, 87.5% of all national medical teams, covering 82.0% of registered athletes, participated in the illness surveillance project and returned on average 93.5% of the expected report forms. The completeness of illness data in the report forms averaged 96.8%. No athlete refused to allow his/her data to be used for scientific research.

Duration of championships and athlete registrations

The 11 championships comprised a total of 59 days, 16 for indoor and 43 for outdoor championships. The mean event duration differed between indoor (mean=3.2 days, range: 3–4) and outdoor championships (mean=7.2 days, range: 5–9) (p=0.002).

A total of 12 594 athletes registered in these championships. The average number of athletes per championship was higher for the outdoor (mean=1614, range: 1352–1896) than the indoor championships (mean=583, range: 539–643) (p=0.002) (table 1).

Table 1

Number and percentage of registered athletes according to type of championships, sex and disciplines

Overall illness IPs and rates, and their variations with championship type, sex and discipline

A total of 546 illnesses were recorded at the 11 championships, 60 during indoor and 486 during outdoor championships. Overall 43.4 illnesses per 1000 registered athletes (95% CI 39.8 to 46.9) (table 2), or 1.2 illnesses per 1000 registered athlete days (95% CI 1.1 to 1.2) were reported.

Table 2

Illness incidence proportions per 1000 registered athletes with 95% CI during 11 international athletics championships according to type of championships, sex and disciplines

The illness IP per 1000 registered athletes was moderately higher during outdoor than indoor championships (RR=2.44; 95% CI 1.87 to 3.18) (table 2), but similar when considering athlete days (RR=0.91; 95% CI 0.69 to 1.19).

The overall illness IP did not differ between male and female athletes (RR=0.85; 95% CI 0.72 to 1.01) (table 2).

The IP was moderately higher in the endurance than the explosive disciplines (RR=1.94; 95% CI 1.65 to 2.29) (table 2). The illness IP also differed between single disciplines (χ2=206.0, p<0.001); it was highest in race walking and marathons, and lowest in throws and jumps (table 2 and figure 1A,B).

Figure 1

Illness incidence proportions per 1000 registered athletes (A) according to the type of championship, discipline (explosive/endurance) and sex, and (B) for each discipline and sex in 11 international athletics championships for all illnesses, illnesses with estimated time loss ≥1 day and illnesses with estimated time loss >7 days. TLI, time loss illness.

Overall illness characteristics

The most common illness location was the upper respiratory tract (30.0%), followed by the gastrointestinal tract (20.0%), the cardiovascular system (15.4%) and the neurological system (9.9%) (table 3). The main symptoms were pain (28.8%), nausea/vomiting/diarrhoea (13.9%) and dehydration/weight loss (12.1%). The most frequent causes were infections (30.8%) and exercise (27.1%); 16.8% were classified as ‘others’ (table 3).

Table 3

Number and percentage of affected systems, causes, estimated duration of time loss, and main diagnoses of illnesses during 11 international athletics championships according to type of championships and sex

Almost a third of all illnesses (28.8%) were expected to result in time loss from sport (information missing for 47 illnesses), corresponding to 12.5 time loss illnesses per 1000 registered athletes (95% CI 10.5 to 14.4), or 0.3 time loss illnesses per 1000 registered athlete days (95% CI 0.3 to 0.4). In 94.9% of the time loss illnesses, the estimated time loss from sport was 1–7 days (minor severity), and in 5.1%  8–28 days (moderate severity). The most common time loss illnesses were exercise-induced fatigue/hypotension/collapse (22.9%), gastroenteritis (22.3%) and upper respiratory tract infections (19.1%). The eight moderate severity illnesses were: infections affecting the upper respiratory tract (n=2), gastrointestinal (n=1) and dermatological (n=1) infections, as well as cardiovascular dehydration/hyperthermia problems (n=2), urogenital problem (n=1) and psychiatric disorder (n=1).

Illnesses during indoor championships

During indoor championships 20.6 illnesses per 1000 registered athletes (95% CI 15.4 to 25.8) or 1.3 illnesses per 1000 registered athlete days (95% CI 1.0 to 1.6) were reported.

The illness IP did not differ between male and female athletes (RR=0.94; 95% CI 0.57 to 1.56) (table 2), and no sex-related difference in the distribution of illness characteristics was observed regarding the system affected, main symptoms, cause, severity or disciplines (endurance/explosive or all disciplines) (tables 2 and 3).

Neither was there any difference of illness IP between the explosive and endurance disciplines (RR=1.62; 95% CI 0.95 to 2.78) (table 2 and figure 1A), nor discipline-related differences in the distribution of illness characteristics with regard to the system, main symptoms, cause, or severity.

Illnesses during outdoor championships

During the outdoor championships 50.2 illnesses per 1000 registered athletes (95% CI 45.9 to 54.6), or 1.2 illnesses per 1000 registered athlete days (95% CI 1.1 to 1.3) were reported.

The IP did not differ between male and female athletes (RR=0.84; 95% CI 0.71 to 1.00) (table 2). The male and female athletes differed regarding the distribution of affected systems (χ2=33.4, p<0.001) (table 3), and the IP of gastrointestinal illnesses (13.3 vs 7.2; RR=1.84; 95% CI 1.23 to 2.76) and urogenital illnesses (3.6 vs 0.0) was slightly higher in women. The distribution of illnesses also varied between the nine specific disciplines (χ2=25.4, p<0.001), mainly due to a higher illness frequency among male race walking athletes than their female peers (table 2). No sex-related difference in illness was observed for main symptoms, cause, severity and explosive/endurance disciplines (tables 2 and 3).

The illness IP was slightly higher in the endurance than explosive disciplines (RR=1.87; 95% CI 1.58 to 2.23) (table 2 and figure 1A,B). The difference increased with illness severity: a small difference for no time loss illnesses (RR=1.37; 95% CI 1.10 to 1.72), a moderate difference for minor illnesses (RR=2.73; 95% CI 1.96 to 3.81) and a very large difference for moderate illnesses (RR=13.2; 95% CI 1.58 to 109.2) (table 4). Endurance and explosive disciplines also differed in the distribution of illness characteristics (ie, affected system [χ2=106.4, p<0.001], main symptom [χ2=112.6, p<0.001], cause [χ2=118.5, p<0.001] and severity [χ2=16.0, p<0.001]), and in the IPs of specific affected systems, symptoms or diagnoses (table 4). The IP of cardiovascular (RR=28.1; 95% CI 12.3 to 64.5, very large) and psychiatric/psychological (RR=4.66; 95% CI 2.01 to 10.8, large) illnesses was higher in endurance than explosive disciplines, while the IP of upper respiratory tract illnesses was similar (RR=0.93; 95% CI 0.65 to 1.33) (table 4). The IP of symptoms like weight loss, fatigue, palpitations and vertigo was higher in endurance than explosive disciplines (35.9 vs 2.7; RR=13.3; 95% CI 8.1 to 21.8, very large). The IP of illnesses with exercise-related (RR=9.25; 95% CI 6.10 to 14.0, very large) and environmental (RR=2.73; 95% CI 1.60 to 4.68, moderate) causes was higher in endurance than explosive disciplines, while infectious causes were higher in explosive than endurance disciplines (RR=1.49; 95% CI 1.02 to 2.18, moderate) (table 4). The IP of exercise-induced dehydration/fatigue/hypotension/collapse was higher in endurance than explosive disciplines (RR=24.1; 95% CI 11.1 to 52.2, very large), and similar for upper respiratory tract infections (RR=1.48; 95% CI 0.91 to 2.41) (table 4).

Table 4

Percentage of illnesses and illness incidence proportions per 1000 registered athletes with 95% CI, for affected systems, causes, estimated duration of time loss and main diagnoses during outdoor championships according to the discipline category and the sex


This analysis of illnesses during international athletics championships based on a large sample of top-level athletes extends results from previous studies based on only one championship.2 3 6 8 12–14

Overall IPs, rates and characteristics of illnesses

The overall illness IP was 43 per 1000 registered athletes and the IR 1.2 per 1000 registered athlete days. In comparison to other sports, the illness IP during international athletics championships was similar to that during the 2016 Olympic Games,8 but lower than during the 2012 Olympic Games,6 the FINA World Aquatics Championships11 21 22 and the 2010 FIFA World Cup.10 Longer championship durations can lead to a higher illness IP.10 13 Climatic and environmental conditions can also play a role in these differences, for instance heat conditions could predispose for exertional heat illnesses especially in endurance athletes.15 23 Methodological aspects regarding the illness surveillance system24 25 could also explain the differences, or any combination of all of these arguments.

The overall characteristics of illnesses seen during athletics championships were similar to those observed in other sports: the respiratory tract was the most frequently affected body system followed by the digestive system, pain was the most frequent symptom, and infection the most frequent cause.6 8 10 21 22 26 27 However, our results show significant differences in illness characteristics according to athletic discipline (ie, endurance vs explosive disciplines), and thus, we suggest caution when considering and analysing sports that may include several different sports, especially including both explosive and endurance elements, as often seen during the Olympic Games and in athletics.

The illness IP per 1000 registered athletes was higher during outdoor than indoor championships but the IR per athlete days was similar for indoor and outdoor championships. This difference can be explained by the longer duration of outdoor championships, as previously hypothesised.10 13 Other major contributors to the observed difference are the marathon and race walking; obviously both are included only in outdoor championships, and both were found to have the highest illness IPs. Due to the long duration, these athletes are exposed to the climatic and environmental conditions for a longer time, with corresponding effects on illness patterns. Therefore, we suggest that data should be presented separately for indoor and outdoor championships.

Sex-related differences

Our study revealed a similar illness IP in female and male athletes. This result is in agreement with those from the FINA World Aquatics Championships in 200921 and 2013.22 However, it is different from the 20126 and 20168 Olympic Games during which, for all sports, female athletes had significantly higher illness IPs than male athletes, that is, 86 vs 536 and 57 vs 408 illnesses per 1000 registered athletes, respectively.

No sex-related differences in the characteristics of illnesses were observed during indoor championships, but female athletes had higher IPs of gastrointestinal illnesses than men during outdoor championships. This is in agreement with results by Pugh et al 28 reporting that elite female athletes were more likely to suffer from gastrointestinal symptoms, while Schwabe et al 29 did not report higher gastrointestinal complications in women during a 21 km race. Further research is needed to better understand gastrointestinal illnesses in athletes and potential sex-related risk differences.28 30 In addition, a higher frequency of urogenital problems in women was observed. These were primarily urogenital pain (88%), with about a third caused by menstruation and almost half by undetermined causes. Menstrual pain was not mentioned in the consensus statement for team physicians,31 although it obviously causes specific problems for female athletes during international championships. A more precise registration by medical teams could help a better understanding of this problem, and propose specific prevention measures.

Differences among disciplines

During outdoor championships, endurance athletes were almost twice as likely to sustain an illness than athletes in explosive disciplines. This finding is consistent with the results from the 2015 IAAF World Championships reporting an almost 10 times higher risk of illnesses in endurance disciplines.15 This could be explained by the longer duration of endurance events increasing the exposure problem, and could especially be due to variations in climatic condition and heat stress. This is borne out by the number of cases of exercise-induced dehydration/fatigue/hypotension/collapse reported in endurance disciplines.

Respiratory tract infections were the most frequent illness reported in explosive disciplines, ranking second in endurance disciplines, with similar IPs in the two discipline groups. The higher risk of upper respiratory tract infections in elite versus recreational athletes has been discussed previously,32 33 but the surprising result here is that it was not the most frequent problem in endurance disciplines. The exercise-induced illnesses were the most frequent illness in endurance disciplines, therefore it is of interest to improve knowledge and prevention of this problem (eg, more precise diagnosis, acclimation, adapting hydration…).15

Based on these results, and in agreement with Timpka et al,15 we suggest specific attention be paid to endurance athletes. Furthermore, we propose that epidemiological data from different disciplines should be analysed separately.

Methodological considerations

A strength of the present study is the use of a standardised methodology for data collection during 11 championships, fulfilling the criteria recommended by Edouard et al,24 with regards to team participation, response rates and completeness of data. Another strength is the large sample size of 12 594 registered athletes and 546 illnesses, which provides a foundation for reliable approximations of the true illness problem,11 allowing more representative results and in-depth analyses, especially for indoor championships.

However, a limitation is that the classification of affected systems, main symptoms and causes, changed slightly in 2013 according to the consensus on athletics' injury and illness surveillance (eg, in 2013 for affected system ‘respiratory tract’ was divided into upper and lower parts and ‘ophthalmological’ category was added; for main symptoms ‘congestion, hypersecretion or discharge’, ‘numbness, weakness or tingling’ and ‘mood/sleep disturbance, anxious or depressed’ categories were added; and for causes ‘nutritional, endocrine or metabolic disturbance’ and ‘psychiatric’ categories were added).18 This classification of the causes of illness was not always suitable as 16.8% of them were reported as ‘others’. The coding system could be improved to allow physicians to report illnesses more accurately.2 3 Furthermore, only the main symptom was recorded, although athletes can suffer with more than one symptom (such as pain and fever with an infection). The calculation of IR using athlete days could also lead to an underestimation of the incidence, since some individual athletes may not have been present during the entire championships. In addition, the current calculation of IP does not exactly represent the number of ill athletes, since some athletes could have presented with more than one illness during a single championship. Finally, although we applied stratified analyses and a Bonferroni correction, some confounding factors may have been missed and effects of multiple comparisons may still remain.

Illness prevention measures

The present results could help (national and local) medical teams, when preparing for championships, to anticipate the most frequent or severe illnesses. The results support the previously published recommendations for preventing illnesses during championships: educate athletes and their entourage on prevention of infectious diseases, dehydration and heat stress; implement simple prevention strategies, like proper hygiene measures, drinking regularly and only bottled water, eating only ‘safe’ food, regular hand washing with alcohol gel, surveillance of special indoor air cleaning systems, decrease contact with people outside the team, and be aware of changes in temperature and climatic conditions.1 2 7 15 23 Screening tests for airway problems and suitable strategies for athletes with a heavy competition load should be offered to all athletes at risk.1 Although prevention strategies have been promoted since many years, the illness risk remained almost unchanged and warrants further study. Don’t athletes adopt these measures? Are these measures inefficient? Should there be more? These questions should be answered by further studies.

Given the difference in illness IPs and characteristics among disciplines observed in this study, we suggest that illness prevention measures should be discipline-specific. Prevention measures for explosive disciplines should focus on upper respiratory tract infections and pre-existing pathologies. For endurance disciplines, these should take into account environmental factors, such as weather conditions to prevent exercise-induced dehydration/fatigue/hypotension/collapse, all illnesses related to the cardiovascular system.

Prevention of illnesses is relevant in a win-win performance-health protection strategy as reported by Hanstad et al.1 The implementation and compliance with illness prevention measures can lead to a decrease in IRs and an increase in medals won. Key to the success of this research was the trust and good relationships between the competitors and their health teams.1

We conclude that illness prevention strategies during athletics championships should be tailored for endurance and explosive disciplines. The focus should be on upper respiratory tract infections, exercise-induced dehydration/fatigue/hypotension/collapse and gastroenteritis/diarrhoea.

What are the findings?

  • The overall incidence rate during 11 international athletics championships was 1.2 illnesses per 1000 registered athlete days (95% CI 1.1 to 1.2).

  • Infection was the most common cause of illness, with upper respiratory tract and gastrointestinal infections being the most frequent.

  • Illness characteristics differed between endurance and explosive disciplines.

  • The illness incidence proportion (IP) during outdoor championships was higher for endurance athletes than for athletes in the explosive disciplines, mainly due to an increased IP of exercise-induced and environment-related disorders.

  • No difference in illness IP and few differences in illness characteristics were observed between male and female athletes during athletics championships.

How might it impact on clinical practice in the future?

  • Illness prevention strategies during athletics championships should at all times focus on infectious disease control, and for endurance disciplines on exercise-induced and environment-related disorders. Clinicians should take into account championship type (indoor or outdoor) and discipline category (endurance or explosive).


The authors are highly appreciative of the cooperation given them by the medical staff of national teams and the physicians attached to competition organising committees who volunteered their time to collect the data for this project.



  • Contributors PE: substantial contributions to the conception and design of the study; collection, analysis and interpretation of data; drafting, writing and revising of the manuscript; and approval of the final version to be published. AJ: substantial contributions to the conception and design of the project; analysis and interpretation of data; writing and revision of the manuscript; and approval of the final version to be published. MS, TT: substantial contributions to the analysis and interpretation of data; revision of the manuscript; and approval of the final version to be published. PB: substantial contributions to the conception and design of the study; collection and interpretation of data; revision of the manuscript; and approval of the final version to be published.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Ethics approval The injury and illness surveillance project (eg,. data collection) was reviewed and approved by the Oslo University School of Medicine Ethical Committee (for athletics championships from 2009 to 2011) and by the Saint-Etienne University Hospital Ethical Committee (for athletics championships from 2012 to 2017), and the analysis of illnesses using all data was reviewed and approved by the Saint-Etienne University Hospital Ethical Committee.

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

  • Patient consent for publication Not required.

Request Permissions

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