Musculoskeletal Injuries and Pain in Dancers: A Systematic Review

1819

REVIEW ARTICLE

Musculoskeletal Injuries and Pain in Dancers: A Systematic Review Cesar A. Hincapié, DC, MHSc, Emily J. Morton, DC, FCCS(C), J. David Cassidy, PhD, DrMedSc ABSTRACT. Hincapié CA, Morton EJ, Cassidy JD. Musculoskeletal injuries and pain in dancers: a systematic review. Arch Phys Med Rehabil 2008;89:1819-29. Objective: To assemble and synthesize the best evidence on the epidemiology, diagnosis, prognosis, treatment, and prevention of musculoskeletal injuries and pain in dancers. Data Sources: Medline, CINAHL, PsycINFO, Embase, and other electronic databases were searched from 1966 to 2004 using key words such as dance, dancer, dancing, athletic injuries, occupational injuries, sprains and strains, and musculoskeletal diseases. In addition, the reference lists of relevant studies were examined, specialized journals were handsearched, and the websites of major dance associations were scanned for relevant information. Study Selection: Citations were screened for relevance using a priori criteria, and relevant studies were critically reviewed for scientific merit by the best evidence synthesis method. After 1865 abstracts were screened, 103 articles were reviewed, and 32 (31%) of these were accepted as scientifically admissible (representing 29 unique studies). Data Extraction: Data from accepted studies were abstracted into evidence tables relating to the prevalence and associated factors, incidence and risk factors, diagnosis, treatment, economic costs, and prevention of musculoskeletal injuries and pain in dancers. Data Synthesis: The scientifically admissible studies consisted of 15 (52%) cohort studies, 13 (45%) cross-sectional studies, and 1 (3%) validation study of a diagnostic assessment tool. There is a high prevalence and incidence of lower extremity and back injuries, with soft tissue and overuse injuries predominating. For example, lifetime prevalence estimates for injury in professional ballet dancers ranged between 40% and 84%, while the point prevalence of minor injury in a diverse group of university and professional ballet and modern dancers was 74%. Several potential risk factors for injury are suggested by the literature, but conclusive evidence for any of these is

From the Artists’ Health Centre Research Program, Toronto Western Hospital (Hincapié, Cassidy); Division of Health Care and Outcomes Research, Toronto Western Research Institute (Hincapié, Cassidy) and Centre of Research Expertise in Improved Disability Outcomes, University Health Network (Hincapié, Cassidy); Dalla Lana School of Public Health, University of Toronto (Hincapié, Cassidy); and Division of Clinical Education, Canadian Memorial Chiropractic College (Morton), Toronto, ON, Canada. Presented in part as a poster to the International Association for Dance Medicine and Science, October 19 –21, 2006, West Palm Beach, FL. Supported by the University Health Network Artists’ Health Centre Foundation, the Ontario Workers’ Safety and Insurance Board through the Centre of Research Expertise in Improved Disability Outcomes, and the Canadian Memorial Chiropractic College. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors or upon any organization with which the authors are associated. Reprint requests to Cesar A. Hincapié, DC, MHSc, Toronto Western Hospital, 399 Bathurst St, Fell Pavilion 4-144, Toronto, ON M5T 2S8, Canada, e-mail: [email protected]. 0003-9993/08/8909-00970$34.00/0 doi:10.1016/j.apmr.2008.02.020

lacking. There is preliminary evidence that comprehensive injury prevention and management strategies may help decrease the incidence of future injury. Conclusions: The dance medicine literature is young and heterogeneous, limiting our ability to draw consistent conclusions. Nonetheless, the best available evidence suggests that musculoskeletal injury is an important health issue for dancers at all skill levels. Better quality research is needed in this specialized area. Future research would benefit from clear and relevant research questions being addressed with appropriate study designs, use of conceptually valid and clinically meaningful case definitions of injury and pain, and better reporting of studies in line with current scientific standards. Key Words Dancing; Injuries; Musculoskeletal system; Pain; Rehabilitation; Systematic review. © 2008 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation ANCERS ARE A UNIQUE blend of artist and athlete particularly susceptible to musculoskeletal injuries and D pain. The health problems of dancers are worthy of attention for several reasons. First, because most dancers begin training at a young age, there is potential for a great impact on their future health. Second, the interplay of physical and aesthetic demands in dance may lead to various health issues especially relevant to dancers. For example, a variety of musculoskeletal, metabolic, and nutritional disorders have been described among dancers,1-4 which may significantly impact on their health-related quality of life. Finally, as an occupational group, dancers have received little attention in the health literature.3 Dance-related injuries are of wide interest, and several reviews have been previously published.5-12 However, these reviews are methodologically weak and limited because they do not meet current scientific standards for reviews of the literature.13 For instance, they typically do not report a literature search strategy, include no quality appraisal of studies from which findings are derived, are based on expert opinion, and are not reproducible. The rationale for systematic reviews of the literature is well established,14 and this approach to dancerelated research has been recommended within the dance medicine and science community.15 The general aim of our literature search was to identify all relevant literature on the health problems of dancers. Our initial screening of the literature revealed 2 substantive and related themes: musculoskeletal injuries and pain, and metabolic and nutritional disorders. This report presents a best evidence synthesis16,17 of the epidemiology, diagnosis, prognosis, treatment, and prevention of musculoskeletal injuries and pain in dancers. Our purpose was to create a baseline of the best scientific

List of Abbreviations SEFIP

self-estimated functional inability because of pain

Arch Phys Med Rehabil Vol 89, September 2008

1820

MUSCULOSKELETAL INJURY AND PAIN IN DANCERS, Hincapié

evidence that can inform clinicians, researchers, dancers, and other stakeholders about the health problems of dancers. METHODS Literature Search We systematically searched the scientific literature published between 1966 and 2004. The primary sources of literature were the electronic databases Medline (1966 to October 2004), PsycINFO and Embase (both 1980 to October 2004), and CINAHL (1982 to October 2004). Additional electronic databases, including the Cochrane Central Trials Registry (CENTRAL), MANTIS, and the Index to Chiropractic Literature, and several others available through EBSCO (ie, AltHealthWatch, AMED, Biomedical Reference Collection, Psychology and Behavioral Sciences Collection) were also searched (dates ranging between 1985 and 1990, to October 2004). Indexed terms and text words such as dance, dancer, dancing, athletic injuries, occupational injuries, sprains and strains, musculoskeletal diseases, bone density, menstruation disturbances, eating disorders, and others were used to search the databases. Additionally, we examined the reference lists of all relevant studies, hand-searched specialized journals (ie, the Journal of Dance Medicine and Science and Medical Problems of Performing Artists), and scanned the websites of major dance associations for unpublished literature. Screening for Relevance to Best Evidence Synthesis Objectives We screened all citations identified through our search strategy and included English-language reports; published reports of original research, systematic reviews, conference proceedings, government reports, guidelines, or unpublished grey literature manuscripts; studies containing original raw data on at least 20 human research participants, including a control group if present; studies examining the prevalence, incidence, associated factors, risk factors, diagnosis, interventions, economic costs, prognosis, or other aspects of musculoskeletal injury and pain, and metabolic and nutritional disorders in dancers; and studies of dancers in any form of artistic dance such as ballet, modern, tap, theatrical, folk, flamenco, break-dancing, ballroom dancing, and ice dancing. We excluded studies on recreational or exercise forms of dance such as aerobic dancing or social dancing in clubs, parties or raves; studies on the cognitive, behavioral, or learning aspects of dance; narrative, editorial, or clinical reviews, opinion papers, letters to the editor, and editorials; studies of conditions with questionable clinical relevance or asymptomatic presentation; studies using cadavers or nonhuman subjects; and studies reporting findings not specific to dancers (eg, studies in which dancers’ information was combined with other athletes’ information and results could not be evaluated specifically for dancers). On the basis of our inclusion and exclusion criteria, 2 reviewers independently appraised the relevance of each citation found in the electronic search through a 2-level screening process, with disagreements resolved by consensus. In the first-level screening, reviewers categorized citations as probably relevant, of unknown relevance, or irrelevant. For each citation rated as probably relevant or of unknown relevance, the entire paper was obtained and, in the second-level screening, these were deemed to be either relevant or irrelevant to the systematic review. Critical Review of the Literature All studies judged relevant were critically appraised. The appraisal assessed scientific merit and clinical relevance by Arch Phys Med Rehabil Vol 89, September 2008

using a priori criteria and electronic critical review forms.18 These forms prompted the reviewer to focus on issues of study design, study population, issues related to conduct of the study, participation rates, follow-up rates (where relevant), measurement issues, and analysis. Quality appraisal criteria were derived from fundamental principals of epidemiologic methods, measurement, and design.19,20 Our initial screening of this literature revealed 2 substantive and related themes: (1) musculoskeletal injuries and pain and (2) metabolic and nutritional disorders. In this review, we report on the first of these themes. Two reviewers performed independent in-depth reviews, and the remaining reviewer read the studies. Decisions about a study’s methodologic quality and scientific merit were made by consensus of the authors after full discussion. Studies were considered scientifically admissible, which could include accepting part or all of the findings, or scientifically inadmissible because of fatal biases and methodologic flaws. Data Abstraction Into Evidence Tables Studies relevant to musculoskeletal injuries and pain in dancers were identified as relating to prevalence and associated factors, incidence and risk factors, diagnosis, treatment, economic costs, or prevention. We abstracted data from admissible studies into evidence tables relating to these topic areas and facilitating synthesis of the information according to the best evidence synthesis method.16,17 These evidence tables summarize our findings and form the basis of our recommendations. RESULTS After applying our inclusion and exclusion criteria to 1865 identified abstracts to assess their relevance to the health problems of dancers, 191 articles were judged to be relevant and critically reviewed. Of these, 103 dealt with musculoskeletal injuries and pain in dancers. We accepted 32 (31%) of these 103 articles (representing 29 unique studies) as scientifically admissible after critical scientific appraisal (fig 1). These studies are the basis for our findings and consist of 15 cohort studies, 13 cross-sectional studies, and 1 validation study of a diagnostic assessment tool. Characteristics of Musculoskeletal Injury in Dancers The fully detailed versions of tables 1 and 2 (supplemental tables 1, 2, available online at http://www.archives-pmr.org) show that most musculoskeletal injuries are soft tissue injuries such as sprains, strains, and tendinopathies, although stress fractures have also been the focus of some of the literature. Overuse and chronic injuries are commonly reported, and most injuries affect the lower extremities and back (see supplemental tables 1, 2). Despite limitations in the quality of the admissible literature, the evidence suggests that most dancers’ injuries are mild or minor and require minimal time off (see supplemental tables 1, 2). Prevalence and Associated Factors of Musculoskeletal Injury and Pain in Dancers We accepted 13 cross-sectional studies (findings reported in 14 publications) on the prevalence and associated factors of musculoskeletal injury and pain (outlined in table 1 and fully detailed in supplemental table 1). The source populations for the studies vary geographically, with 3 studies from North America, 8 from European countries, 1 from China, and 1 from Australia (see table 1). The inclusion and exclusion criteria also varied considerably across the studies. Six studies focused solely on professional

MUSCULOSKELETAL INJURY AND PAIN IN DANCERS, Hincapié

Potentially relevant publications identified and screened for retrieval (N=1865) Articles excluded on basis of title and abstract or brief screen of complete publication (n=1674) Articles retrieved for in-depth, critical review (n=191)

Articles relevant to metabolic and nutritional disorders in dancers theme (n=97)*

Articles relevant to musculoskeletal injuries and pain in dancers theme (n=103)*

Inadmissible articles due to fatal methodologic flaws (n=71)

Scientifically admissible articles† (n=32, representing 29 studies) Topic areas: • 14 articles/13 studies on prevalence and associated factors • 17 articles/15 studies on incidence and risk factors • 1 article/1 study on diagnosis • 4 articles/2 studies on interventions and economic costs • 6 articles/4 studies on care-seeking and health care services Fig 1. Flow diagram of systematic review inclusion and exclusion. * Some articles are relevant to both themes. †Some articles relate to more than 1 topic area.

dancers,21-27 4 on university or preprofessional dancers,28-31 and 3 on a mix of professional, preprofessional, and/or university-level dancers.32-34 Most studies included ballet dancers only.23-28,33,34 One study focused solely on theatrical dancers22 and one on ice dancers,29 and 3 included a variety of dance styles such as ballet, modern, theatrical, and Chinese traditional dance.21,31,32 Case definitions for injury also varied considerably. For example, 1 study defined injury as an incident that resulted in time off dancing or required medical consultation,31 while another used reported injuries that had affected dancing “in any way.”21 Modifications of the Nordic Musculoskeletal Questionnaire,35 which defines “musculoskeletal trouble” as pain, aching, or discomfort in various body regions, were used by 1 group of authors.25-27 Three studies did not provide any explicit definition of injury.23,29,34 Many studies divided injuries into acute, chronic, overuse, or major versus minor; however, these categories were rarely well defined. Given the different periods for prevalence estimates, diverse source populations, varying inclusion and exclusion criteria, and assortment of case definitions, it is not surprising that the reported prevalence estimates for injury or pain vary from a low of 3% to a high of 95% (see table 1). Given this heterogeneity, it is not possible to report a summary

1821

proportion. Nonetheless, as a whole, the literature suggests that the prevalence of musculoskeletal injury and pain in dancers is high. The point prevalence of minor injury in a diverse group of university and professional ballet, modern, and theatrical dancers was 74%,32 and that of pain related to chronic injuries in professional ballet and modern dancers was 48%.21 Lifetime prevalence estimates for injury in professional ballet dancers ranged between 40% and 84%, with most dancers reporting more than 1 dance-related injury (see table 1). In university and preprofessional dancers, lifetime prevalence of musculoskeletal injury ranged between 26% and 51% (see table 1, excluding prevalence estimates in ice dancers). In 2 better quality studies, the 1-year period prevalence of musculoskeletal pain in Swedish professional ballet dancers was reported to be about 95%,25,26 and 90% of those followed up 6 years later reported recurrent pain.27 Eleven of the 13 prevalence studies reported on factors associated with musculoskeletal injury or pain in dancers. Table 1 lists these factors (full detail in supplemental table 1), including age,23 sex,26,33 years of dance experience,23,34 performance level,26,34 style of dance,31 hours of training a day,24 dance setting (eg, performance, rehearsal, dance class),21 prior injury,22 personality traits,23 work dissatisfaction,26 menstrual dysfunction,24,31 muscular tension before performing,26 joint hypermobility,33 sensorimotor control,30 choreography demands,22 stretching and flexibility training,28 and performing on a raked or angled stage.22 Overall, these reported associations are preliminary in nature, and only 1 study22 made use of multivariable analysis to adjust for important covariates in assessing associated factors. Incidence of and Risk Factors for Musculoskeletal Injury and Pain in Dancers We accepted 15 cohort studies (findings reported in 17 publications) on the incidence of and risk factors for musculoskeletal injury and pain (outlined in table 2 and fully detailed in supplemental table 2). Most cohorts are from dance schools or companies in the United States, with the exception of cohorts from South Africa,36 Korea,37 Sweden,38 Norway,39 and Switzerland.40 These studies cover inception periods from 1973 to 2001 (see table 2). Seven studies focused on professional dancers, including 5 ballet company cohorts,36,38-43 1 modern dance company cohort,44 and 1 Broadway show cohort45 in which ballet predominated. Six studies included preprofessional dancers (3 ballet cohorts,46-48 1 theatrical dance cohort,49 1 mixed ballet and modern dance cohort,50 1 ice dance cohort51), and 2 studies included a mix of professional, university, and secondary school–level dancers.37,52 All studies included both female and male subjects except one46 that examined the attrition rate of female ballet dancers from a preprofessional school of ballet. Case definitions for injury varied considerably. Three studies did not explicitly define injury.37,38,46 Five studies defined injury as any dance-related problem that warranted attention by, referral to, or treatment by a health care professional39,48-51; another 3 used definitions of injury based on being unable to dance or time loss from dance class, rehearsal, or performance40,45,52; 1 study defined injury as any musculoskeletal complaint resulting in financial outlay by the dance company44; another used all workers’ compensation–reported injuries36; and yet another used any dance-related medical problem selfreported on a standardized form to the company financial officer.41-43 Incidence estimates were reported in the accepted studies as either cumulative incidence proportions (ie, number of injury Arch Phys Med Rehabil Vol 89, September 2008

1822

Dance Setting Study

Country 32

Chmelar et al Hamilton et al23 Bowling21 McNeal et al34

Style

Level

Study Size (N); Response Rate

Prevalence Estimates† Age (y)

Point

6 Months

12 Months

Lifetime

Associated Factors

NA Age, years of dance, personality traits Dance setting Years of dance, performance level Hours of training per day, menstrual dysfunction Menstrual dysfunction, dance style Prior injury, choreography demands, performing on a raked stage Sex, performance level, muscular tension before performing, work dissatisfaction Nonassociated factors: age, sex, workload

USA USA

B, M B

P, U P

39; 64% 29; 64%

18–37 22–41

74% ND

ND ND

23% ND

ND 20%–79%

B, M B

P P, PP, U, YS P

141; 75% 350; 80%–100%

42% ND

ND ND

84% 18%–80%

54; 55%

⬎18 ⬍13 and ⱖ13 NA

48% ND

Kadel et al24

U.K. USA and Canada Sweden

ND

ND

ND

32%

To et al31

China

U

98; 95%

17–33

51%–85%

7%–39%

ND

ND

Evans et al22

England

B, M, T, CT T

P

58; 71%

NA

NA (no dance-specific prevalence estimates given)

Ramel and Moritz26

Sweden

B

P

64; 84%

17–47

ND

ND

69%–94%

ND

Ramel and Moritz25 and Ramel et al27 Askling et al28

Sweden

B

P

128; 87% 51; 60%

17–47

ND

ND

61%–95%

ND

Sweden

B

PP

98; 99%

17–25

ND

ND

ND

51%

DubravcicSimunjak et al29 McCormack et al33 Hiller et al30

Croatia

I

PP

469; 82% 136 dancers

13–20

ND

ND

ND

3%–15%

England

B

P, PP

NA

NA (study focused on hypermobility and BJHS)

Sex, joint hypermobility

Australia

NA

U

287; NA 220 dancers 21; NA

22⫾7

NA (study focused on factors associated with functional ankle instability)

Sensorimotor control

B

Stretching and flexibility training NA

Abbreviations: B, ballet; BJHS, benign joint hypermobility syndrome; CT, Chinese traditional; I, ice dance; M, modern; NA, not available; ND, no data; P, professional; PP, preprofessional; T, theatrical; U, university; YS, young student. *See supplemental table 1 for full details at http://www.archives-pmr.org. † Point prevalence is the proportion of people in a population who have a disease or condition at a particular time, such as a particular date. Period prevalence is the proportion of people in a population who have a disease or condition over a specific period, such as 6 or 12 months. Lifetime prevalence is the proportion of people in a population who at some point in their lives (up to the time of study) have experienced an injury, compared with the total number of people at risk in the population.

MUSCULOSKELETAL INJURY AND PAIN IN DANCERS, Hincapié

Arch Phys Med Rehabil Vol 89, September 2008

Table 1: Cross-Sectional Studies of Prevalence and Associated Factors of Musculoskeletal Injury and Pain in Dancers*

Table 2: Cohort Studies of Incidence and Risk Factors for Musculoskeletal Injury and Pain in Dancers* Dance Setting Study Size (N)

Incidence Estimates†

Basis for Injury Case Definition

Cumulative Incidence

Country

Style

Level

Inception Period

USA South Africa

T B

PP P

9mo 10y

218 47

Medical attention WCB report

85% ND

Wiesler et al50

USA

B, M

PP

9mo

148

Medical attention

64%

ND .65 Injuries per dancer-year ND

Bronner and Brownstein45 Hamilton et al46 Miller and Moa48 Ménétrey and Fritschy40 Solomon et al,41 Solomon et al,42 and Solomon et al43 Liederbach and Compagno52

USA

B, T

P

7wk

30

Time-loss

40%

ND

USA USA Switzerland USA

B B B B

PP PP P P

4y 10mo 1y 5y

NA Medical attention Time-loss Self-report

55% (dropout rate) 43% 42% 77%–94%

ND ND ND ND

USA

B

P, U

2y

644

Time-loss

NA

NA

Nilsson et al38

Sweden

B

P

5y

98

NA

ND

Byhring and Bo39

Norway

B

P

5mo

41

Medical attention

76%

0.62 Injuries per 1000 dance-hours 3.2 Injuries per dancer

Luke et al47

USA

B

PP

9mo

39

Mixed: self-report and medical attention

90% self-report; 77% medical attention

Bronner et al44

USA

M

P

5y

42

Financial

17%–81%

Fortin and Roberts51 Noh and Morris37

USA Korea

I B

PP P, U, HS

1 competition 10mo

Medical attention NA

17% NA

40 210 (42 dancers) 60 59 to 68

208 (58 dancers) 105

Incidence Density

4.7 Injuries per 1000 dance-hours; 2.9 Injuries per 1000 dance-hours 0.18-0.57 Injuries per 1000 dance-hours ND NA

Risk Factors

NA Sex Prior injury, dance style Nonassociated factors: sex, BMI, years of training, ankle ROM NA Musculoskeletal injury Dance Prior injury Seasonal timing, sex Nonassociated factors: age, rank Fatigue, psychologic characteristics associated with eating disorders, dieting, frequency/ intensity of training Age, sex Seasonal timing Nonassociated factors: stress/tension, feeling of having influence on working conditions NA

MUSCULOSKELETAL INJURY AND PAIN IN DANCERS, Hincapié

Nonassociated factors: sex NA Psychosocial coping

Abbreviations: B, ballet; BMI, body mass index; HS, high school; I, ice dance; NA, not available; ND no data; M, modern; P, professional; PP, preprofessional; ROM, range of motion; T, theatrical; U, university; WCB, workers’ compensation board. *See supplemental table 2 for full details at http://www.archives-pmr.org. † Cumulative incidence is the number of new injury cases during a specified period divided by the number of people at risk in the population at the beginning of the study. Incidence density is the number of new injury cases during a specified period divided by the total time at which individuals in the population are at risk.

1823

Arch Phys Med Rehabil Vol 89, September 2008

Study

Rovere et al49 Klemp and Learmonth36

1824

MUSCULOSKELETAL INJURY AND PAIN IN DANCERS, Hincapié

cases/population at risk for injury) or incidence density rates (ie, number of injury cases/population-time at risk for injury). The cumulative incidence of musculoskeletal injury ranged from 40% to 94% in professional dancers and from 17% to 90% in preprofessional and school-level dancers (see table 2). Finally, incidence density rates for injury varied across the accepted studies from a low of 0.18/1000 dance-hours in professional ballet dancers44 to a high of 4.7/1000 dance-hours in preprofessional high school ballet dancers.47 Given the heterogeneity in inception periods, diverse source populations, varied case definitions for injury, and the different methods used to report incidence in the accepted studies of musculoskeletal injury in dancers, the variability in incidence estimates shown in table 2 is not surprising, and an overall summary incidence estimate would be meaningless. Nine of the 15 incidence studies explored risk factors for injury in dancers (see table 2), 1 study focused on dance as a risk factor for injury in performing arts students,48 and 1 additional study investigated risk factors, including injury, for attrition from ballet school.46 Most are descriptive studies that explored crude associations between potential determinants and the onset of injury or pain. Few studies37,44,50 used stratified and/or multivariable analyses to determine which factors have important independent associations with incident musculoskeletal injuries or pain. Table 2 lists the risk factors that have received attention in the literature (full detail in supplemental table 2), including age,38,41-43 sex,36,38,41-44,50 number of years of training,50 dancer’s rank,41-43 prior injury,40,50 seasonal timing,39,41-43 frequency or intensity of training,52 fatigue,52 psychologic characteristics associated with eating disorders (ie, drive for thinness, bulimic tendencies, perfectionism, body dissatisfaction–– domains of the Eating Disorders Inventory–252,53), dieting,52 psychosocial coping,37 dance as a risk factor for injury in performing arts students,48 and injury as a risk factor for dropping out of ballet school.46 Overall, these reported risks are modest and preliminary in nature, and there are no confirmatory studies54 of risk factors for musculoskeletal injury or pain in dancers. Diagnostic Procedures and Assessment Tools for Musculoskeletal Injury and Pain in Dancers One cross-sectional study of a diagnostic and assessment method for musculoskeletal pain and functional limitation in dancers55 was scientifically admissible (table 3). This study suggests that the SEFIP questionnaire—a modified version of the Nordic Musculoskeletal Questionnaire35—shows good agreement with pain and dysfunction elicited on physical examination and may be useful as a screening instrument in dancers. Interventions and Economic Costs for Musculoskeletal Injury and Pain in Dancers We accepted 2 cohort studies (findings reported in 4 publications) of interventions and costs for musculoskeletal injury (table 4). Both studies focused on similar interventions: injury prevention and management programs implementing a policy of “self-insurance” against injury-related expenses.41-44 However, they used uncontrolled, observational study designs in varying populations of dancers and at different times (see table 4). As such, these studies provide only weak, preliminary evidence concerning intervention effectiveness and economic costs. Their findings suggest that comprehensive injury prevention and management programs may reduce the annual incidence of dance injuries and new workers’ compensation cases, Arch Phys Med Rehabil Vol 89, September 2008

decrease the total number of lost work days, improve dance company morale, and result in reduced health care insurance costs for dance-related injuries (see table 4). Health Care Provider Use and Care-Seeking Patterns for Musculoskeletal Injury and Pain in Dancers Among the studies we accepted as scientifically admissible, 4 studies (findings reported in 6 publications) addressed the care-seeking habits or health care provider choices of injured dancers (table 5). Overall, these studies suggest that most injured dancers seek treatment for their injuries and receive care from various health care practitioners such as physiotherapists, primary care physicians or specialists, massage therapists, chiropractors, osteopaths, and acupuncturists (see table 5). These findings are only descriptive in nature: none clarify why certain practitioners are preferred over others or address the effectiveness of one type of practitioner over another. Of note, 1 study found that 15% to 30% of young and preprofessional dancers did not seek medical attention for their injuries.34 However, it is not known whether the younger dancers’ injuries were severe enough to warrant medical attention, the dancers misjudged the severity of their injuries, the dancers “danced through” their injuries, or these nonprofessional dancers experienced financial barriers to seeking treatment. These issues warrant further study. DISCUSSION This systematic review is an important contribution to the world of dance medicine and science. It is the first attempt to collect and appraise systematically all of the literature on musculoskeletal injuries and pain in dancers. It is striking that over two thirds (69%) of the 103 articles that we reviewed were judged as scientifically inadmissible. Many were small case series; used a study design unsuited to the research question; had too vague a case definition; provided insufficient information about the source population, the at-risk population, or sampling methods; or involved unrepresentative, highly selected study populations. We accepted 32 (31%) of the 103 articles that we reviewed. This literature has many limitations, and it is difficult to draw consistent conclusions from it. Some of these problems are inherent in all studies of musculoskeletal injury. For example, there is no universally accepted definition of musculoskeletal injury, and the studies are so heterogeneous that it is difficult to compare prevalence estimates, incidence rates, and risk factors. Another obvious problem is that many studies do not identify the population at risk that should form the denominator in any prevalence or incidence calculation. The admissible studies also have a wide range of inclusion and exclusion criteria that affect the interpretation and comparability of the findings. For instance, study populations vary geographically in the performance levels of dance as well as in the styles of dance examined. Despite the foregoing limitations, there are some important conclusions that we can make. There is evidence that musculoskeletal injury is an important health issue for dancers at all skill levels. There is a high prevalence and incidence of lower extremity and back injuries, with soft tissue and overuse injuries predominating. Several potential risk factors for injury are suggested by the literature, but conclusive evidence for any of these is lacking. In addition, there is preliminary evidence that comprehensive injury prevention and management strategies may help decrease the incidence of injury and contain health care insurance costs for dance-related injuries. It is difficult to draw any conclusions from this literature regarding the severity or prognosis of dance injury. Few studies

Findings Outcome Measures

SEFIP: Intensity of present pain in relation to ability to dance self-assessed on a 5-point scale for 14 body regions. Test battery: present pain, marked differences between sides, or muscular weakness blindly and independently assessed in examination of flexibility, strength, and coordination. Tool: SEFIP questionnaire, based on the Nordic Musculoskeletal Questionnaire. Reference: test battery for pain and/or muscular dysfunction designed for dancers and done through physical examination. Professional ballet dancers (17 females/ 11 males; mean age, 27.6y; range, 19 – 43y) from 2 major dance companies. (n⫽28) Ramel et al55 Validation: construct validity, crosssectional. Sweden

Diagnostic Tool and Reference Standard Subjects and Setting Authors, Study Design, and Country

Table 3: Study of Diagnostic Procedure and Assessment Tool for Musculoskeletal Injury and Pain in Dancers

SEFIP: mean sensitivity and specificity for all 14 body regions was 78% and 89%, respectively. When shins, elbows, and wrists were excluded (because of ⬍5 dancers reporting pain), sensitivity was 85% and specificity was 88%. Mean agreement between SEFIP and test battery was 88% (range, 75% [for hips] to 96% [for neck]), and the mean ␬ was .69 (␬ range, .48 [for knees and hips] to .92 [for neck]).

MUSCULOSKELETAL INJURY AND PAIN IN DANCERS, Hincapié

1825

explicitly define the severity of injury, and among those that do, the definitions are inconsistent. From these heterogeneous data, we may tentatively conclude that most dancers’ injuries are mild or minor and require minimal time off (see supplemental tables 1, 2). Although severe injuries are less common, these understandably account for most time loss from dancing.32,36,39,44 The extent of disability resulting from dance injuries, however, remains unclear. Review Strengths and Limitations Our review has several strengths that are worth noting. We systematically gathered, appraised, and synthesized the best available evidence. Some readers, unfamiliar with the best evidence synthesis approach to reviewing the literature, may not appreciate its value. However, we believe that limiting our findings to studies that are of better methodologic quality is a notable strength. We also reported all steps of our comprehensive review explicitly, thereby making our process more transparent and reproducible. Several limitations also merit discussion. Like all reviews of the literature, our systematic review is limited by the quantity and quality of the available evidence. We note especially the lack of studies in young dancers and in dance styles other than ballet. We were also disappointed in the limited quality of studies; for example, few studies used multivariable analysis to assess the independence of potential risk factors for injury. Finally, our review is limited to the literature available as of October 2004. Future Recommendations There are several topics identified in this literature with purported importance to dancers’ health for which no scientifically admissible evidence was found. For example, it has been suggested that dancers are prone to impingement syndromes of the ankle56-59; however, we are unable to draw any conclusions regarding the prevalence or incidence of impingement syndromes because of fatal methodologic flaws in the available literature. Few studies have addressed the long-term effect of injury on a dancer’s career and the impact of injury on healthrelated quality of life. The possibility that exposure to intensive dance training at a young age may lead to long-term musculoskeletal health consequences remains unanswered. While we accepted several studies that estimate the prevalence or incidence of stress fractures in dancers, risk factors for stress fractures in dancers have not been investigated using appropriate research designs. This is an area requiring further study, particularly with respect to possible risk factors such as menstrual disturbances, eating disorders, and the female athlete triad.60 Finally, there is no scientifically admissible evidence on the effectiveness of preparticipation screening in identifying and modifying risk factors for musculoskeletal injury in dancers. We recommend that several questions be addressed by future research to inform dancers, dance companies, clinicians, and researchers about musculoskeletal injury and pain (appendix 1). Several methodologic issues are important for future research. It is difficult to form consistent conclusions about the prevalence, incidence, risk, and prevention of musculoskeletal injury in dancers because of the heterogeneity in the accepted studies. An important source of this heterogeneity is variation in case definitions of injury. We recommend that future authors consider the validity of their case definitions of injury and discuss this issue in their publications. Case definitions should incorporate information about injury severity, type, and location, and delineate acute traumatic Arch Phys Med Rehabil Vol 89, September 2008

1826

Subjects and Setting

Intervention Evaluated

Outcomes and Follow-Up

Key Findings and Limitations

Solomon et al, Solomon et al,42 and Solomon et al43 Series of five 1-year prospective single cohorts USA

Professional ballet dancers of all ranks, using a variety of choreographic styles, in a ballet company from 1993–1998. (n range, 59 – 68, depending on year of study)

Injury incidence proportions, direct costs in U.S. dollars, and health care services use. 5-year follow-up.

Injury incidence declined slightly during the 5-year period: 94%, 87%, 82%, 77%, and 81%, in years 1 through 5, respectively. The average annual health care costs were $258,286. The estimated direct cost savings to the company were $1,210,994 over the 5-year period. An improvement in company morale over the study period was noted. Nonrandomized, descriptive design limits the conclusions.

Bronner et al44 5-Year single cohort with before-after intervention USA

Professional modern dancers age 19 – 40 years from 2 dance companies in 1996 – 2001. Data from years 1 and 2 (ie, before intervention) were compiled retrospectively, and data from years 3–5 (ie, after intervention) were collected prospectively. (n⫽42)

Injury prevention and management program including injury screening, transitional classes for injured dancers, an injury prevention focus group, and a policy of “self-insurance” against injury-related expenses: the company paid health care providers directly for “minor” injuries involving no lost time and no single billing over $500, and claimed only “major” injuries against workers’ compensation insurance. Comprehensive management program implemented in years 3–5, involving primary prevention (ie, dance-specific annual screenings, technique modification, cross-training, and treatment of minor complaints) and secondary prevention, including on-site case management and intervention.

Incidence proportions of workers’ compensation cases, incidence density rates of injury, and number of dance days missed because of workers’ compensation injury. 5-year follow-up.

Yearly incidence of workers’ compensation cases ranged between 79% and 81% in years 1 through 3 and dropped to 24% and 17% in years 4 and 5, respectively. Incidence of injury per 1000 dance-hours was .51, .48, .57, .29, and .18 for years 1 through 5, respectively. Total numbers of lost work days caused by new workers’ compensation injuries were 230, 135, 70, 58, and 87 for years 1 through 5, respectively. Nonrandomized, descriptive design limits the conclusions.

41

MUSCULOSKELETAL INJURY AND PAIN IN DANCERS, Hincapié

Arch Phys Med Rehabil Vol 89, September 2008

Table 4: Studies of Interventions and Economic Costs for Musculoskeletal Injury and Pain in Dancers Authors, Study Design, and Country

Table 5: Studies of Health Care Provider Utilization and Care-Seeking Patterns for Musculoskeletal Injury and Pain in Dancers Authors, Study Design, and Country

Key Health Services Findings

Professional ballet and modern dancers, age 18 years or more, from 7 professional dance companies in 1987. (n⫽141)

Physiotherapists, medically qualified practitioners (general practitioners or specialists), massage therapists, acupuncturists, and osteopaths.

Primary: prevalence of dancerelated injuries. Secondary: dancers’ perceptions of causes of injuries, availability and take up of treatment from professional therapists and others.

McNeal et al34 Cross-sectional USA/Canada

Professional, preprofessional, college, and young student ballet dancers from 5 ballet schools and preprofessional companies, 3 colleges, and 4 professional companies. (n⫽350) Professional ballet dancers of all ranks, using a variety of choreographic styles, in the Boston Ballet Company from 1993–1994 to 1997–1998. (n range, 59 – 68, depending on year of study)

Not specified. Includes expert medical advice and/or treatment.

Primary: lifetime prevalence of ankle, knee, and foot injury in each of the study’s dance groups. Secondary: percentage of injured dancers seeking medical advice or treatment for injuries. Primary: injury incidence and direct health care–related costs in U.S. dollars. Secondary: health care services use.

Of the dancers injured in the past 6 months, 100% consulted a health professional, and 63% saw ⬎1 professional. The most commonly seen health professionals were physiotherapists (76%), general practitioners or specialists (47%), and massage therapists (29%). Most dancers who saw medical doctors did so privately and had medical insurance coverage. Some dancers had insurance for complementary and alternative health practitioners or physiotherapists but not for medical doctors. Of injured preprofessional dancers, 73% sought medical advice or treatment for their injuries, compared with 97% of injured professional dancers. Except for injured professional dancers, 15%–30% of injured dancers did not seek medical attention for their injuries. Physical therapists earned the greatest percentage of dollars for services provided (mean, 48.6%), followed by massage therapists (18.1%) and chiropractors (10.9%). Financial statistics suggested that the income derived from contracting to provide health care services to a dance company is unlikely in itself to produce a lucrative health care practice. Health care providers, most commonly nonmedical, were consulted for 92% of injuries to dancers. Some dancers consulted 1 or more providers. The most commonly seen health care providers were physiotherapists (57%), massage therapists (9%), osteopaths, chiropractors (5%), and acupuncturists.

Solomon et al,41 Solomon et al,42 and Solomon et al43 Series of five 1-year prospective singlesample cohorts USA

Evans et al22 Cross-sectional England

Subjects and Setting

Professional theatrical dancers and actors from 20 West End theater productions. (n⫽269)

Health Care Providers Involved

Physiotherapists, massage therapists, chiropractors, physicians, Pilates-based providers, acupuncturists, and podiatrists.

Physiotherapists, massage therapists, general practitioners, osteopaths, chiropractors, acupuncturists, podiatrists, and medical specialists.

Primary: total number of injuries, sex-specific number of injuries in current production, and factors associated with prevalent musculoskeletal injury. Secondary: percentage of injuries for which medical attention was sought and types of health professionals seen.

MUSCULOSKELETAL INJURY AND PAIN IN DANCERS, Hincapié

1827

Arch Phys Med Rehabil Vol 89, September 2008

Outcome Measures

Bowling21 Cross-sectional United Kingdom

1828

MUSCULOSKELETAL INJURY AND PAIN IN DANCERS, Hincapié

injuries from overuse injuries, as well as chronic and recurrent injuries.61,62 It would also be useful to distinguish between injuries occurring during training and those occurring during performance. Producing valid prevalence and incidence estimates requires not only accurate and complete ascertainment of cases but also accurate and complete ascertainment of the population at risk. Cases must be drawn from that population at risk. We recommend that authors clearly state their source populations and discuss issues around their sampling frames for inclusion into their studies. In calculating prevalence proportions, injury recall periods should be limited to a short time (eg, up to 6mo) to reduce memory bias. In addition, incidence density rates are preferable to cumulative incidence proportions because they are a more accurate measure of the population-time at risk. More meticulous attention to issues of bias and research methodology is critical to understanding the epidemiology, diagnosis, prognosis, treatment, and prevention of musculoskeletal injuries and pain in dancers. Risk factors should be studied in cohort or case-control study designs employing multivariable statistical analysis that allows for the identification of independent risk factors after adjustment for other important covariates. Finally, rigorous statistical methods are essential. CONCLUSIONS The literature on the epidemiology, diagnosis, prognosis, treatment, and prevention of musculoskeletal injuries and pain in dancers is young and developing. Nonetheless, there is evidence that musculoskeletal injury is very common and an important health issue for dancers at all skill levels. However, the resultant disability and impact on healthrelated quality of life remain unclear. Better quality research is needed in this specialized area. Future research would benefit from clear and relevant research questions being addressed with appropriate study designs, use of conceptually valid and clinically meaningful case definitions of injury and pain, and better reporting of studies in line with current scientific standards. We have summarized the literature, highlighted its strengths and weaknesses, and provided suggestions for future research. Acknowledgments: We thank Anne Taylor-Vaisey MLS, Heather Hanson DC, Angela Pucci MSc, and Deanne Collier from the Canadian Memorial Chiropractic College for their assistance with searching and retrieving the literature.

References 1. Warren MP, Brooks-Gunn J, Hamilton LH. Scoliosis and fractures in young ballet dancers: relation to delayed menarche and secondary amenorrhea. N Engl J Med 1986;314:1348-53. 2. Dhuper S, Warren MP, Brooks-Gunn J, Fox RP. Effects of hormonal status on bone density in adolescent girls. J Clin Endocrinol Metab 1990;71:1083-8. 3. Bronner S, Ojofeitimi S, Spriggs J. Occupational musculoskeletal disorders in dancers. Phys Ther Rev 2003;8:57-68. 4. Warren MP, Brooks-Gunn J, Fox RP, et al. Persistent osteopenia in ballet dancers with amenorrhea and delayed menarche despite hormone therapy: a longitudinal study. Fertil Steril 2003;80:398404. 5. Sohl P, Bowling A. Injuries to dancers: prevalence, treatment and prevention. Sports Med 1990;9:317-22. 6. Greer JM, Panush RS. Musculoskeletal problems of performing artists. Baillieres Clin Rheumatol 1994;8:103-5. 7. Schon LC, Weinfeld SB. Lower extremity musculoskeletal problems in dancers. Curr Opin Rheumatol 1996;8:130-42. 8. Garrick JG, Lewis SL. Career hazards for the dancer. Occup Med 2001;16:609-18. 9. Hansen PA, Reed K. Common musculoskeletal problems in the performing artist. Phys Med Rehabil Clin N Am 2006;17:789-801. 10. Kadel NJ. Foot and ankle injuries in dance. Phys Med Rehabil Clin N Am 2006;17:813-26. 11. Miller C. Dance medicine: current concepts. Phys Med Rehabil Clin N Am 2006;17:803-11. 12. Motta-Valencia K. Dance-related injury. Phys Med Rehabil Clin N Am 2006;17:697-723. 13. Oxman AD, Guyatt GH. Guidelines for reading literature reviews. CMAJ 1988;138:697-703. 14. Mulrow CD. Systematic reviews: rationale for systematic reviews. BMJ 1994;309:597-9. 15. Caine CG, Garrick JG. Dance. In: Caine DJ, Lindner KJ, Caine CG, editors. Epidemiology of sports injuries. Champaign: Human Kinetics; 1996. p 124-60. 16. Slavin RE. Best evidence synthesis: an alternative to metaanalytic and traditional reviews. Edu Researcher 1986;15:5-11. 17. Slavin RE. Best evidence synthesis: an intelligent alternative to meta-analysis. J Clin Epidemiol 1995;48:9-18. 18. Von Holst H, Nygren A, Schubert J, et al. Best evidence synthesis on mild traumatic brain injury: results of the WHO Collaborating Centre for Neurotrauma, Prevention, Management and Rehabilitation Task Force on Mild Traumatic Brain Injury. J Rehabil Med 2004;43(Suppl):8-144.

APPENDIX 1: FUTURE RESEARCH PRIORITIES FOR MUSCULOSKELETAL INJURY AND PAIN IN DANCERS What are the long-term effects of intensive dance training on young dancers? What is the long-term impact of musculoskeletal injury and pain on the health-related quality of life of dancers? What is the related disability associated with prevalent and incident dance injuries? What are the independent risk factors for lower extremity and back injuries in dancers? Will interventions to prevent dance injuries prove effective in randomized controlled trials? Will epidemiologic, intervention, or prognosis studies focused on other styles of dance generate different results than studies focused on ballet or modern dance? What, if any, measures that would warrant specific injury prevention actions can be used to screen before a dance season? What clinical indicators can be used to help dance companies and dancers determine when a dancer can return to performing or training without increased risk of reinjury? What health insurance benefits are available for injured dancers? Would the provision of benefits or onsite injury management prove effective in improving injury recovery? How commonly is musculoskeletal injury not reported by dancers to their dance company, and what are the reasons for not reporting injuries? How do dancers who do not report manage their injuries, and how safe and effective are these methods?

Arch Phys Med Rehabil Vol 89, September 2008

MUSCULOSKELETAL INJURY AND PAIN IN DANCERS, Hincapié

19. Fletcher RH, Fletcher SW. Clinical epidemiology: the essentials. 4th ed. Baltimore: Lippincott Williams & Wilkins; 2005. 20. Rothman KJ, Greenland S. Modern epidemiology. 2nd ed. Philadelphia: Lippincott-Raven; 1998. 21. Bowling A. Injuries to dancers: prevalence, treatment, and perceptions of causes. BMJ 1989;298:731-4. 22. Evans RW, Evans RI, Carvajal S. Survey of injuries among West End performers. Occup Environ Med 1998;55:585-93. 23. Hamilton LH, Hamilton WG, Meltzer JD, Marshall P, Molnar M. Personality, stress, and injuries in professional ballet dancers. Am J Sports Med 1989;17:263-7. 24. Kadel NJ, Teitz CC, Kronmal RA. Stress fractures in ballet dancers. Am J Sports Med 1992;20:445-9. 25. Ramel E, Moritz U. Self-reported musculoskeletal pain and discomfort in professional ballet dancers in Sweden. Scand J Rehabil Med 1994;26:11-6. 26. Ramel EM, Moritz U. Psychosocial factors at work and their association with professional ballet dancers’ musculoskeletal disorders. Med Probl Perform Art 1998;13:66-74. 27. Ramel EM, Moritz U, Jarnlo G. Recurrent musculoskeletal pain in professional ballet dancers in Sweden: a six-year follow-up. J Dance Med Sci 1999;3:93-100. 28. Askling C, Lund H, Saartok T, Thorstensson A. Self-reported hamstring injuries in student-dancers. Scand J Med Sci Sports 2002;12:230-5. 29. Dubravcic-Simunjak S, Pecina M, Kuipers H, Moran J, Haspl M. The incidence of injuries in elite junior figure skaters. Am J Sports Med 2003;31:511-7. 30. Hiller CE, Refshauge KM, Beard DJ. Sensorimotor control is impaired in dancers with functional ankle instability. Am J Sports Med 2004;32:216-23. 31. To WW, Wong MW, Chan KM. The effect of dance training on menstrual function in collegiate dancing students. Aust N Z J Obstet Gynaecol 1995;35:304-9. 32. Chmelar R, Fitt SS, Shultz BB, Ruhling RO, Zupan MF. A survey of health, training, and injuries in different levels and styles of dancers. Med Probl Perform Art 1987;616. 33. McCormack M, Briggs J, Hakim A, Grahame R. Joint laxity and the benign joint hypermobility syndrome in student and professional ballet dancers. J Rheumatol 2004;31:173-8. 34. McNeal AP, Watkins A, Clarkson PM, Tremblay I. Lower extremity alignment and injury in young, preprofessional, college, and professional dancers, part II: dancer-reported injuries. Med Probl Perform Art 1990;5:83-8. 35. Kuorinka I, Jonsson B, Kilbom A, et al. Standardised Nordic questionnaires for the analysis of musculoskeletal symptoms. Appl Ergon 1987;18:233-7. 36. Klemp P, Learmonth ID. Hypermobility and injuries in a professional ballet company. Br J Sports Med 1984;18:143-8. 37. Noh YE, Morris T. Designing research-based interventions for the prevention of injury in dance. Med Probl Perform Art 2004;19: 82-9. 38. Nilsson C, Leanderson J, Wykman A, Strender LE. The injury panorama in a Swedish professional ballet company. Knee Surg Sports Traumatol Arthrosc 2001;9:242-6. 39. Byhring S, Bo K. Musculoskeletal injuries in the Norwegian National Ballet: a prospective cohort study. Scand J Med Sci Sports 2002;12:365-70. 40. Ménétrey J, Fritschy D. Subtalar subluxation in ballet dancers. Am J Sports Med 1999;27:143-9.

1829

41. Solomon R, Micheli LJ, Solomon J, Kelley T. The “cost” of injuries in a professional ballet company: anatomy of a season. Med Probl Perform Art 1995;10:3-10. 42. Solomon R, Micheli LJ, Solomon J, Kelley T. The “cost” of injuries in a professional ballet company: a three-year perspective. Med Probl Perform Art 1996;11:67-74. 43. Solomon R, Solomon J, Micheli LJ, McCray E Jr. The “cost” of injuries in a professional ballet company: a five-year study. Med Probl Perform Art 1999;14:164-9. 44. Bronner S, Ojofeitimi S, Rose D. Injuries in a modern dance company: effect of comprehensive management on injury incidence and time loss. Am J Sports Med 2003;31:365-73. 45. Bronner S, Brownstein B. Profile of dance injuries in a Broadway show: a discussion of issues in dance medicine epidemiology. J Orthop Sports Phys Ther 1997;26:87-94. 46. Hamilton LH, Hamilton WG, Warren MP, Keller K, Molnar M. Factors contributing to the attrition rate in elite ballet students. J Dance Med Sci 1997;1:131-8. 47. Luke AC, Kinney SA, D’Hemecourt PA, Baum J, Owen M, Micheli LJ. Determinants of injuries in young dancers. Med Probl Perform Art 2002;17:105-12. 48. Miller C, Moa G. Injury characteristics and outcomes at a performing arts school clinic. Med Probl Perform Art 1998;13:120-4. 49. Rovere GD, Webb LX, Gristina AG, Vogel JM. Musculoskeletal injuries in theatrical dance students. Am J Sports Med 1983;11: 195-8. 50. Wiesler ER, Hunter DM, Martin DF, Curl WW, Hoen H. Ankle flexibility and injury patterns in dancers. Am J Sports Med 1996; 24:754-7. 51. Fortin JD, Roberts D. Competitive figure skating injuries. Pain Physician 2003;6:313-8. 52. Liederbach M, Compagno JM. Psychological aspects of fatiguerelated injuries in dancers. J Dance Med Sci 2001;5:116-20. 53. Garner DM, Garfinkel PE. The eating attitudes test: an index of the symptoms of anorexia nervosa. Psychol Med 1979;9: 273-9. 54. Altman DG, Lyman GH. Methodological challenges in the evaluation of prognostic factors in breast cancer. Breast Cancer Res Treat 1998;52:289-303. 55. Ramel EM, Moritz U, Jarnlo GB. Validation of a pain questionnaire (SEFIP) for dancers with a specially created test battery. Med Probl Perform Art 1999;14:196-203. 56. Hamilton WG, Geppert MJ, Thompson FM. Pain in the posterior aspect of the ankle in dancers: differential diagnosis and operative treatment. J Bone Joint Surg Am 1996;78:1491-500. 57. Labs K, Leutloff D, Perka C. Posterior ankle impingement syndrome in dancers—a short-term follow-up after operative treatment. Foot Ankle Surg 2002;8:33-9. 58. Spicer DD, Howse AJ. Posterior block of the ankle: the results of surgical treatment in dancers. Foot Ankle Surg 1999;5:187-90. 59. Stoller SM, Hekmat F, Kleiger B. A comparative study of the frequency of anterior impingement exostoses of the ankle in dancers and nondancers. Foot Ankle 1984;4:201-3. 60. Otis CL, Drinkwater B, Johnson M, Loucks A, Wilmore J. American College of Sports Medicine position stand: the female athlete triad. Med Sci Sports Exerc 1997;29:i-ix. 61. Bronner S, Ojofeitimi S, Mayers L. Comprehensive surveillance of dance injuries: a proposal for uniform reporting guidelines for professional companies. J Dance Med Sci 2006;10:69-80. 62. Liederbach M, Richardson M. The importance of standardized injury reporting in dance. J Dance Med Sci 2007;11:45-8.

Arch Phys Med Rehabil Vol 89, September 2008

Study, Design, and Country

Case Definitions

University-level and professional female ballet and modern dancers age 18 –37 years from 2 advanced performing university groups and 3 professional dance companies. 64% Response rate (39/61). (n⫽39)

Injury that kept dancer from dancing for more than 2 or 3 weeks in the past year reported as major injury. Any “recurrent physical nuisance” that interferes with, but does not stop the dancer from dancing reported as minor injury.

Point prevalence of minor injury: 74%. 1-year period prevalence of major injury: 23%.

NA

Hamilton et al23 Cross-sectional USA

Professional ballet dancers (14 F/15 M) of soloist or principal rank age 22– 41 years from 2 professional companies. Disabled dancers (n⫽15) were excluded. 64% Response rate (29/45). (n⫽29) Professional ballet and modern dancers (80 F/61 M), age 18 years or more, from 7 professional dance companies in 1987. 75% Response rate (141/188). (n⫽141)

No explicit definition of injury. Orthopedist categorized injuries into major, minor, overuse, and stress fracture.

Lifetime prevalence of major injury in females vs males: 43% vs 60%; of minor injury: 57% vs 40%; of overuse syndrome: 79% vs 53%; of stress fracture: 29% vs 20%.

Older age, joining the dance company at an older age, a greater number of years dancing, and personality characteristics suggestive of the “overachiever” were associated with a higher prevalence of injury.

Injury sustained that had affected dancing “in any way” was counted. Chronic injury defined as giving the dancer “continuing problems.”

6-month period prevalence of injury: 42%. Lifetime prevalence of injury: 84% (30% reported 1 injury, 54% reported 2 or more injuries). Point prevalence of chronic injury: 48%.

Of injuries in the past 6 months, 32% occurred during performances, 28% occurred during rehearsals, 16% occurred during dance classes, 7% had a slow/gradual onset, and 17% were attributed to various situations.

In professionals, the lifetime prevalence of ankle, knee, and foot injury was 80%, 57%, and 51%, respectively. In preprofessionals, the lifetime prevalence of ankle, knee, and foot injury was 43%, 46%, and 26%. In college dancers, the lifetime prevalence of ankle, knee, and foot injury was 36%, 35%, and 40%. In young students, the lifetime prevalence of ankle, knee, and foot injury was 27%, 23%, and 18%. Lifetime prevalence of stress fracture: 32%.

The prevalence of injuries tended to increase with level and years of training: professional dancers and those with 9 or more years of dance experience had the highest prevalence of injury.

Bowling21 Cross-sectional United Kingdom

McNeal et al34 Cross-sectional USA/Canada

Professional (49 F/50 M; age ⱖ17y), preprofessional (157 F/14 M; age range, 13–21y), college (58 F; no age range given), and young student (22 F; age ⬍13y) ballet dancers from 5 ballet schools and preprofessional companies, 3 colleges, and 4 professional companies. 80%–100% Response rates reported. (n⫽350)

No explicit definition of injury.

Kadel et al24 Cross-sectional Sweden

Professional female ballet dancers of all ranks from 2 companies. 55% Response rate (54/98). (n⫽54)

To et al31 Cross-sectional China

University-level female ballet, Chinese traditional, modern, and musical theater dance students age 17–33 years from a collegiate school of dancing. 95% Response rate reported. (n⫽98)

Stress fractures defined as physician-diagnosed based on history and physical examination, and confirmed by positive bone scan or x-ray imaging. Injury defined as an incident that resulted in time off dancing or required physiotherapy or medical consultation. No explicit definition of “chronic orthopedic problems.”

Prevalence

6-month period prevalence of ⬎5 episodes of musculoskeletal injury in eumenorrheic vs oligomenorrheic vs amenorrheic dancers: 7% vs 33% vs 38%. Point prevalence of chronic orthopedic problems in eumenorrheic vs oligomenorrheic vs amenorrheic dancers: 51% vs 73% vs 85%.

Associated Factors

Dancing more than 5 hours per day and having a history of amenorrhea lasting longer than 6 months were associated with a higher prevalence of stress fracture. Dancers with menstrual dysfunction were more likely than dancers with normal menses to have ⬎5 episodes of musculoskeletal injury in the past 6 months, have chronic orthopedic problems, and require medical consultation. Prevalence of injury was higher among ballet dancers than among modern or theatrical dancers.

Other Findings A total of 46 injuries were reported, 12 major and 34 minor injuries (some dancers had ⬎1 injury). The most common injury locations were ankle (26%), back (24%), knee (20%), and foot (15%). Mean disability duration was 10⫾10.9 months.

Most common injury locations: back/neck (29% of chronic injured vs 26% of 6-mo injured); ankle (20% vs 19%), knee (17% vs 12%), and foot/ toes (6% vs 16%). Soft tissue injuries were most common: 60% of 6-month injured and 48% of chronic injured. In professionals, 28% reported no lost time due to ankle injury; 40% reported no lost time because of knee injuries, and 37% reported no lost time because of foot injuries.

Most common stress fracture locations: 63% were metatarsal, 22% were tibial, and 7% were spinal. NA

MUSCULOSKELETAL INJURY AND PAIN IN DANCERS, Hincapié

Subjects, Setting, and Response Rate

Chmelar et al32 Cross-sectional USA

1829.e1

Arch Phys Med Rehabil Vol 89, September 2008

Supplemental Table 1: Studies of Prevalence and Associated Factors of Musculoskeletal Injury and Pain in Dancers

Supplemental Table 1: Studies of Prevalence and Associated Factors of Musculoskeletal Injury and Pain in Dancers (Continued ) Study, Design, and Country

Other Findings

Professional theatrical dancers (37 F/21 M) and actors (65 F/146 M), from 20 West End theater productions in February 1996. Disabled performers were excluded. 71% Response rate (269/379, includes both dancers and actors). (n⫽269; 58 dancers)

Subjects, Setting, and Response Rate

Injury defined as any theater-related injury resulting in physical damage to the person, even those not resulting in missed performances.

Case Definitions

NA (no dancer-specific prevalence estimate given).

A history of previous injury, greater physical demands of choreography, and performing on a raked stage were more likely to be reported by dancers with prevalent injury than those without. Prevalence OR for injury in dancers vs actors was 2.4 (95% CI, 1.3–4.5).

Ramel and Moritz26 Cross-sectional Sweden

Professional ballet dancers (38 F/26 M) age 17– 47 years from a professional company. Included 8 principals, 8 soloists, and 48 corps ranks. 84% Response rate (64/76). (n⫽64)

Musculoskeletal pain defined according to the Nordic Musculoskeletal Questionnaire as pain, aching, and/or discomfort in 9 body regions. Incapacitating pain defined as musculoskeletal pain that prevented the dancer from doing their daily work.

12-month period prevalence of musculoskeletal pain: 94%. 12-month period prevalence of incapacitating pain: 69%.

Ramel and Moritz25 and Ramel et al27 Cross-sectional with 6-year follow-up Sweden

Professional ballet dancers (75 F/53 M) age 17– 47 years from 3 major dance companies. Follow-up: 51 dancers from original study (34 F/17 M). 87% Response rate in original study (128/147); 60% response rate in follow-up (51/85 dancers employed on both occasions). (n⫽128 and n⫽51)

12-month period prevalence of musculoskeletal pain: 95%. 12-month period prevalence of incapacitating pain in dancers completing both studies: 67% in original study, 61% in follow-up. 12-month period prevalence of recurrent pain in follow-up: 90%.

Askling et al28 Cross-sectional Sweden

Preprofessional student ballet dancers (79 F/22 M) age 17–25 years at different stages of a 4-year program. 99% Response rate (98/99). (n⫽98)

Musculoskeletal pain defined according to the Nordic Musculoskeletal Questionnaire as pain, aching, and/or discomfort in 9 body regions. Incapacitating pain defined as musculoskeletal pain that prevented the dancer from doing their daily work. Recurrent pain defined as pain reported in the same body region on both study occasions. Hamstring injury defined as injury to the rear thigh. Acute injury defined as a sudden sharp pain or “pop” that the dancer relates to a specific situation. Overuse injury defined as a problem with insidious onset that continues to bother the dancer for ⱖ2 weeks.

Soloists and principals (25% of study population) accounted for 45% of the prevalence of musculoskeletal pain, with foot/ankle and low back pain most incapacitating. Female dancers reported more upper- and lower-body pain. Factors associated with musculoskeletal pain preventing regular work were increased muscular tension prior to performances and dissatisfaction with work. No significant sex differences in pain intensity or location. Follow-up: despite increased workload and older age, dancers were not more prone to incapacitating pain.

Among the dancers, a total of 65 injuries were reported. Most common injury locations: lower extremity (52%), neck and back (34%), ankle (19%), and knee (17%). Of injuries in dancers, 19% resulted in at least 1 missed performance. Most common pain locations: low back (75%), feet/ankles (61%), and neck (61%). Foot/ ankle problems were more incapacitating than low back problems (47% vs 31%).

Dubravcic-Simunjak et al29 Cross-sectional Croatia

Elite junior figure skaters (168 F/165 M) and ice dancers (68 F/68 M) age 13–20 years at Junior World Figure Skating Championships and Croatia Cup from 1998 –2002. 82% Response rate (469/572). (n⫽469; 136 ice dancers)

No explicit definition of injury.

Prevalence

Lifetime prevalence of hamstring injury: 51% (34% acute, 17% overuse).

Of acute injuries, 88% occurred during slow, controlled stretching.

Lifetime prevalence of acute injury in female vs male ice dancers: 10% vs 15%. Lifetime prevalence of overuse injury in female vs male ice dancers: 3% vs 10%.

NA

Most common pain locations: low back (70%), ankles/feet (63%), and neck (54%). Most dancers had trouble in 2 or 3 body regions. Pain during the past 7 days was also most common in the ankles/ feet (30%), low back (27%), and neck (20%).

Of 33 dancers who reported acute injury, 4 experienced more than 1 injury. Recovery time ranged from 2 weeks to 80 months (median, 8mo). Of those who reported acute injury, 30% continued their ordinary training and dancing in spite of injury. NA

MUSCULOSKELETAL INJURY AND PAIN IN DANCERS, Hincapié

1829.e2

Arch Phys Med Rehabil Vol 89, September 2008

Associated Factors

Evans et al22 Cross-sectional England

1829.e3

Study, Design, and Country

Subjects, Setting, and Response Rate

Case Definitions

McCormack et al33 Cross-sectional England

Professional ballet dancers (43 F/28 M) and preprofessional ballet dance students (older: 35 F/29 M; younger: 47 F/38 M) from 1 professional company and 1 preprofessional/youth ballet school were compared with a control group of nondancing adults (16 F/15 M) and teenagers (21 F/15 M) from the professional ballet company institution and a local high school, respectively. No response rates given. (n⫽220 dancers vs n⫽67 controls)

Hypermobility defined as a Beighton score of ⱖ4/9. BJHS defined by the revised Brighton criteria. Arthralgia defined as pain in 1–3 joints or back pain. No explicit definition of “multiple joint dislocations” and “soft tissue injuries.”

NA (study focused on hypermobility and BJHS).

Prevalence

Multiple dislocations and soft tissue injuries were reported in more dancers than controls and in more dancers with than without BJHS. Arthralgia was more common in males than females, with the most common sites the ankle, cervical spine, and lumbar spine.

Hiller et al30 Cross-sectional Australia

Uninjured ankles were compared with injured ankles in 21 intermediate or higher university-level female dancers (mean age, 22⫾7y) from 3 dance schools. Ten dancers had no ankle injury, 6 had unilateral functional ankle instability, and 5 had bilateral instability. (n⫽16 injured ankles vs n⫽26 uninjured ankles)

Functional ankle instability defined as chronic ankle instability resulting from a history of ankle inversion injury that required protected weight bearing and/or immobilization. Chronic ankle instability defined as perceiving that the ankle is chronically weaker, more painful, and/or less functional than the other ankle or than before first injury.

NA (study focused on factors associated with functional ankle instability).

Sensorimotor control was altered in functionally unstable ankles. The ability of trained dancers with functional ankle instability to regain their balance after an inversion perturbation and to stand on demipointe was impaired compared with healthy controls.

Abbreviations: BJHS, benign joint hypermobility syndrome; CI, confidence interval; F, females; M, males; NA, not available; OR, odds ratio.

Associated Factors

Other Findings Point prevalence of hypermobility in preprofessional female vs male dance students: 94% vs 83%; in professional dancers: 95% vs 82%. Point prevalence of BJHS: in preprofessional female vs male dance students: 46% vs 35%; in professional dancers: 26% vs 36%. Prevalence OR for hypermobility: preprofessional dancers vs teenage controls, 11.3 (95% CI, 4.1–31.2); professional dancers vs adult controls, 11.1 (95% CI, 3.8 –31.8). Prevalence OR for BJHS: preprofessional dancers vs teenage controls, 3.9 (95% CI, 1.3–11.3); professional dancers vs adult controls, 1.7 (95% CI, 0.6 – 4.7). NA

MUSCULOSKELETAL INJURY AND PAIN IN DANCERS, Hincapié

Arch Phys Med Rehabil Vol 89, September 2008

Supplemental Table 1: Studies of Prevalence and Associated Factors of Musculoskeletal Injury and Pain in Dancers (Continued )

Supplemental Table 2: Studies of Incidence of and Risk Factors for Musculoskeletal Injury and Pain in Dancers Study, Design, and Country

Only injuries severe enough to warrant attention by a physician were counted.

Klemp and Learmonth36 10-year retrospective single cohort South Africa

Professional ballet dancers age 19 – 47 years in a professional company during a 10-year period. (n⫽47)

All dance students (162 F/56 M) seen by staff of a schoolaffiliated sports medicine clinic for injury from September 1981 through May 1982. No exclusion criteria given. All workers’ compensation injuries sustained among the dancers (30 F/17 M) while dancing were reported. No exclusion criteria given.

Case Definitions

Wiesler et al50 9-month prospective single cohort USA

Preprofessional ballet and modern dance students from a school of performing arts in 1992–1993. (n⫽148)

87% Sample (119 F/29 M; ballet n⫽101, modern n⫽47) gave data on any lower-extremity injury that was seen at the school’s health services department. No exclusion criteria given.

Injury defined as any acute or chronic problem that warranted medical attention.

Bronner and Brownstein45 7-week retrospective single cohort USA

Professional theatrical/ballet dancers of all ranks, age 19 – 38 years and averaging 44 –54 hours of dancing per week, from a Broadway show. (n⫽30)

All injuries reported by dancers (19 F/11 M), observed and documented by the stage manager, or seen by the medical staff of the show were counted. No exclusion criteria given.

Injury defined as time lost from performing. Major injury defined as causing an absence from an entire performance. Minor injury defined as causing an absence from part of a performance. Complaint defined as a reported injury not resulting in any time lost from performance.

Hamilton et al46 4-year prospective single cohort USA

Preprofessional female ballet dance students from a school of ballet. (n⫽40)

No explicit inclusion and exclusion criteria reported. 56% Sample of intermediate/advanced female dance students (mean age, 14.9⫾1.0y) provided data on factors associated with attrition/ dropping out from ballet school.

No explicit definition of injury.

Miller and Moa48 10-month retrospective multiple cohort USA

Preprofessional ballet dance (n⫽42), music (n⫽77), theater (n⫽43), visual arts (n⫽41), and creative writing (n⫽7) students at a performing arts school in 1995–1996. (n⫽210; 42 dancers)

All students presenting to an oncampus arts medicine clinic and referred by the school nurse to a physiatrist from September 1995 to June 1996. Excludes those seeking care off-campus and those not referred for physiatrist consultation.

Injury defined as musculoskeletal injury severe enough to warrant referral to a physician.

All injuries reported to workers’ compensation were counted. Mild injury was mild pain, no loss of function, and ⬍48 hours of missed work. Moderate injury was moderate pain, some loss of function, and 48 hours to 2 weeks of missed work. Severe injury was severe pain, marked loss of function, and ⬎2 weeks of missed work.

Incidence Of the dancers, 85% consulted the sports medicine clinic for injury during the 9-month period. Of 352 total injuries, 88% were dance-related. The most common injury locations were ankle (22%), spine (18%), and foot (15%). Among 47 dancers, there were 168 injuries reported during a 10-year period. Incidence density rate: .65 injuries per dancer-year. Soft tissue injuries (ie, injuries of the muscles, tendons, and ligaments) of the ankle (23%) and knee (13%) were most common. Most injuries were mild in severity. Of the participating students, 64% consulted the school’s health services department for injury during the 9month period. Of the nonparticipating students, 64% (n⫽22) also had injuries recorded. Of all injuries, 62% were ankle or foot injuries.

Overall injury incidence of 40%. Of a total of 12 injuries, 9 major injuries resulted in 82 missed and 35 partial performances, and 3 minor injuries resulted in 22 partial performances. The most common injury locations, ankle/foot (50%) and low back (34%), accounted for over 90% of missed and partial performances. Of the dance students, 55% dropped out of ballet school during the 4-year study period.

Of the ballet dance students, 43% were injured during the 10-month school year. The most common injury locations were foot or ankle, leg, hip, knee, and back.

Risk Factors NA

Males sustained a greater number of interphalangeal joint hyperextension injuries in the hand than females.

Dancers with previous injuries were more likely to be injured than those without previous injury (63% vs 35%). Modern dancers were twice as likely as ballet dancers to have an incident injury. Characteristics not associated with incident injury: sex, body mass index, years of training, ankle range of motion. NA

Factors associated with attrition: musculoskeletal injuries were more common in students dropping out of ballet school vs those continuing school (major injuries in year 1, 50% vs 13%; knee injuries in year 2, 50% vs 7%; tendinitis in years 3 and 4, 40% vs 0%). Dance students had the highest incidence of injury (43%) compared with music (17%), theater (19%), visual arts (5%), and creative writing (0%) students.

1829.e4

Inclusion and Exclusion Criteria

Preprofessional theatrical dance students from a school of performing arts in 1981–1982. (n⫽218)

MUSCULOSKELETAL INJURY AND PAIN IN DANCERS, Hincapié

Arch Phys Med Rehabil Vol 89, September 2008

Source Population

Rovere et al49 9-month retrospective single cohort USA

Study, Design, and Country

Inclusion and Exclusion Criteria

Case Definitions

Incidence

Risk Factors

Professional ballet dancers from a ballet company with more than 130 performances in 1989 –1990. (n⫽60)

No explicit inclusion and exclusion criteria reported. Includes a complete case series of subtalar subluxations in a professional ballet company from August 1989 to July 1990.

Of the dancers, 42% (15 F/10 M) experienced subtalar subluxations over the 1-year period. This represented 10.5% of all reported injuries during the season (n⫽238) and 58% of ankle injuries (n⫽43).

All dancers who experienced a subtalar subluxation injury had reported previous ankle sprains.

Solomon et al,41 Solomon et al,42 and Solomon et al43 Series of five 1-year prospective single cohorts USA Liederbach and Compagno52 2-year retrospective multiple cohort USA

Professional ballet dancers of all ranks from a ballet company with more than 125 performances per year and a variety of choreographic styles from 1993–1998. (n range, 59 – 68, depending on year of study)

No explicit inclusion and exclusion criteria reported. All injuries reported to the company’s financial officer or ombudsperson.

Subtalar subluxation defined as an ankle injury involving a strange sensation of forward displacement of the foot, sharp pains in the anterior part of ankle and the midtarsal joints, and tenderness in the hindfoot, all resulting in being unable to dance and only partially walk on the injured foot. Injury defined as any dance-related medical problem reported by a dancer on a standardized form provided for that purpose, whether or not treatment or expense was involved.

Annual incidence of injury ranged between 77% and 94%. Over the 5 years, the most common injury locations were ankle (21%), foot/toes (17%), hip/thigh (13%), and low back (12%). Of injuries, 48% were sprains/ strains or tendinopathies.

University-level dancers from a conservatory, professional ballet dancers from a ballet company, and dancers seen at a hospital-based outpatient orthopedic clinic for dance injuries in 1998 –2000. (n⫽644)

All injuries reported to an onsite physical therapist in the university conservatory (192 F/ 90 M) and ballet company (69 F/54 M) settings, and all injuries presenting to the outpatient clinic (186 F/53 M), from September 1998 through September 2000 were counted. No exclusion criteria given.

Injury defined as causing a dancer to miss out part or all of a class, rehearsal, or performance.

NA (no numerator given). One hundred sixty injury reports were collected among the 282 university-level dancers, and 101 among the 123 professional ballet dancers. Some dancers had multiple injury reports filed during the 2-year period.

Nilsson et al38 Series of five 1-year single cohorts Sweden

Professional ballet dancers age 17– 40 years of corps, soloist, and principal ranks, from a national ballet company during 5 seasons from 1988 – 1993. (n⫽98; n per year, 72– 80)

An average of 78 dancers per year (range, 72–80; 46 F [43–48]/32 M [29–34]; mean age, 28.3y), training a total of 48 hours per week, gave information about all injuries for which in-house orthopedic consultation was sought. Excludes injuries not presenting for in-house health care.

No explicit definition of injury. Injury classified as either traumatic (ie, patient can define a specific sudden onset of pain with a defined trauma) or overuse (no explicit definition given).

Overall incidence density rate: .62 injuries per 1000 dance-hours (.56 in females vs .70 in males). Among 98 dancers, 390 injuries were tallied (208 injuries among 50 females and 182 injuries among 48 males). Of injuries, 43% were traumatic, while 57% were overuse. The most common injury locations were ankle or foot (54%), lower back or gluteal (18%), and knee (11%).

Byhring and Bo39 5-month prospective single cohort Norway

Professional ballet dancers of corps and soloist ranks from a national ballet company in 1998. (n⫽41)

80% (41/51) sample (27 F/14 M; mean age, 26.7y; range, 19–40y) provided data on musculoskeletal injuries among professional ballet dancers with a contract with the Norwegian National Ballet. Excludes those without a contract and pregnant dancers.

Injury that occurs as a result of dance participation, reduces the level of training, and requires a need for advice or treatment was counted. Injury severity graded by the time of absence from training/performing (minor, 1–7d; moderate, 8–21d; serious, ⱖ21d). Acute injury defined as a history of a single, sudden, violent trauma. Overuse injury defined as repeated microtrauma with cumulative effect on the body tissues.

Of the ballet dancers, 76% experienced 1 or more injuries during the 5-month study period. Incidence of injury was 3.2 injuries per dancer. Most injuries were mild to moderate in severity. The majority of injuries were located in the foot or ankle, followed by hip and back.

Dancers were at highest risk for injury when they resumed a heavy work schedule after a period of relative inactivity (eg, after holidays). Upper limb injuries were more common in males than in females, and foot injuries were more common in females than in males. Characteristics not associated with injury incidence: age and rank. In both the university and ballet company settings, a mood state of fatigue, several psychologic characteristics commonly associated with eating disorders (ie, drive for thinness, bulimic tendencies, perfectionism, and body dissatisfaction), dieting regularly to lose weight, and dancing more than 5 hours a day, 5 days a week were associated with injury incidence. Of the traumatic knee injuries, 79% were in men, while overuse injuries were more common in women. Upper limb injuries were also more common in males. Of ankle sprains, 75% occurred in dancers age ⱕ26 years. Stress fractures occurred more in younger dancers than the average company age (mean age, 20.7y vs 28.3y). Of the injuries, 59% occurred at the beginning of the season. Characteristics not associated with injury incidence: stress and tension, and the feeling of having influence on working conditions.

MUSCULOSKELETAL INJURY AND PAIN IN DANCERS, Hincapié

Source Population

Ménétrey and Fritschy40 1-year retrospective single cohort Switzerland

1829.e5

Arch Phys Med Rehabil Vol 89, September 2008

Supplemental Table 2: Studies of Incidence of and Risk Factors for Musculoskeletal Injury and Pain in Dancers (Continued )

Supplemental Table 2: Studies of Incidence of and Risk Factors for Musculoskeletal Injury and Pain in Dancers (Continued ) Source Population

Inclusion and Exclusion Criteria

Case Definitions

Incidence

Luke et al47 9-month prospective single cohort USA

Preprofessional ballet dance students age 14 –18 years at a performing arts high school in 2000-2001. (n⫽39)

71% sample (34 F/5 M; mean age, 15.8⫾1.0y) gave data on all selfperceived or self-reported injuries through biweekly surveys and physical therapist– assessed or reported injuries presenting for health services. Excludes injuries in students absent on a survey day.

Injury defined as both self-reported and therapist-assessed dance problems. New injury was one the dancer had not experienced, based on past medical history, location, diagnosis, and biweekly survey reports. Recurrent injury was one previously reported on biweekly surveys or during the preparticipation screening.

NA (underpowered for risk analysis).

Bronner et al44 5-year single cohort with before-after intervention USA

Professional modern dancers age 19 – 40 years from 2 dance companies in 1996 – 2001. 331,080 dance-hours at risk studied. (n⫽42)

Injury defined as any musculoskeletal complaint resulting in company financial outlay. Injury severity graded by the time of absence from dance (minor, ⬍1wk; moderate, 1–4wk; severe, ⬎4wk). Overuse injury defined as resulting from repetitive microtrauma. Traumatic injury defined as resulting from a specific macro-traumatic event.

Fortin and Roberts51 1 Competition, prospective, multiple cohort USA Noh and Morris37 10-month prospective multiple cohort Korea

208 Athletes participating in a national figure skating competition (90 singles, 60 pairs, 58 ice dance skaters). (n⫽58 ice dancers)

All 42 dancers (21 F/21 M) of the 2 companies provided information on injuries involving new workers’ compensation cases and injuries costing less than $500 (ie, self-insured cases) during years 1 and 2 (ie, before intervention) and years 3–5 (ie, after intervention). Excludes any dancer performing less than 30 days annually. All junior-level and senior-level athletes, including ice dance skaters (29 F/29 M; age range, 16–27y), participating in competition gave data on precompetition and competition injuries. No explicit inclusion/exclusion criteria reported. Twenty-seven professional dancers, 19 university dancers, and 59 high school dancers (101 F/4 M; mean age, 20.5⫾5.5y) provided information on psychosocial factors associated with injury.

One hundred twelve self-reported injuries and 71 therapist-assessed injuries were identified. Of dancers, 90% recorded a self-reported injury, and 77% presented to health services for medical attention. Incidence/1000 dance-hours was 4.7 (95% CI, 3.8–5.6) for self-reported injuries and 2.9 (95% CI, 2.2–3.6) for therapist-assessed injuries. Of the self-reported injuries, 56% were new injuries, and 44% were recurrent. Yearly incidence of workers’ compensation cases was 79% to 81% in years 1 through 3 and dropped to 24% and 17% in years 4 and 5. Injury incidences/1000 dance-hours were .51, .48, .57, .29, and .18 for years 1 through 5, respectively. Most injuries were overuse or minor sprains and strains, and the most common locations were foot and ankle, followed by low back and pelvis. Of the ice dance skaters, 17% experienced an injury during the competition. Most injuries were overuse injuries and involved the lower extremity.

NA

Dancers who reported low levels of psychosocial coping on “freedom from worry” and “confidence and achievement motivation”—2 subscales of the Athletic Coping Skills Inventory28—had a higher frequency of injury. These 2 factors explained 21% of the variance in frequency of injury.

Professional ballet dancers, university-level ballet dancers, and high-schoollevel ballet dancers from Korea. (n⫽105)

A significant injury was one that precluded training or impaired performance and presented for evaluation and treatment at the time of competition.

No explicit definition of injury. An injury scale addressed the frequency and duration of injury during the 10-month study period.

Risk Factors

Characteristic not associated with injury incidence: sex.

NA

MUSCULOSKELETAL INJURY AND PAIN IN DANCERS, Hincapié

Abbreviations: F, females; M, males; NA, not available.

1829.e6

Arch Phys Med Rehabil Vol 89, September 2008

Study, Design, and Country