Questionnaire development, dissemination and variable definitions

The Safe Dance IV questionnaire was developed based on previous studies from 1989¹⁷ to 1999,¹⁸ the Fit to Dance 2 survey²⁰ from the UK, a literature review and advice from key stakeholders, including current and recently retired professional dancers, dance physiotherapists and peak bodies. The questionnaire included components on dancer demographics, early dance training and current dance working environments, occurrence of injury in the past 12 months, injury sites and types, possible contributing factors, injury management and healthcare access and return dance practices. The questionnaire predominantly contained closed questions with limited options, though there were some open-ended questions allowing for expansion or personal opinion to be documented.

The questionnaire was available in both electronic and hard copy format. It was distributed to all known large, medium and small dance companies within Australia, as well as via social media (Facebook and Twitter), dance advocacy and support groups and independent dance networks. Data were collected between February 2016 and February 2017 using REDCap electronic data capture tools.²¹

The study inclusion criteria were any professional dancer of any dance style working in Australia. A professional dancer was defined as a person aged over 18 years, who was paid to work, rehearse or perform as a dancer in Australia for a combined period of 3 months or more over the past 12 months, and who personally identified primarily as a professional dancer. A dance-related injury was defined as a physical problem deriving from stress or other causes to do with performance, rehearsal, training, touring or other circumstances of dance life, which affects ability to participate fully in normal dance training, performance or physical activity.²⁰ A strain was an injury to a tendon or muscle and a sprain was defined as a stretch or tear to a ligament.²² Full-time dancers were defined as those who were employed or contracted to work, rehearse or perform as a dancer for at least 48 weeks out of the past 12 months. Dancers meeting the overall study inclusion criteria but were employed for less than 48 weeks of the past 12 months were considered part-time dancers. When answering the injury-related parts of the questionnaire, dancers were instructed to focus on the one injury sustained in the past 12 months that was most significant to them.

Statistical analysis

All data analysis was conducted in IBM SPSS Statistics for Windows (V.22). Statistical significance was set a priori p<0.05.

χ² or two-tailed Fisher’s exact tests were used to test for statistical differences between dance employment status (full time or part time) and demographic characteristics, dance experience, injury characteristics and injury contributors. Independent t-tests were used to determine statistical differences between dance employment status and normally distributed continuous variables, including age, body mass index and years of dance employment. Mann-Whitney U tests were used to determine differences between dance employment status and continuous ordinal data that were not normally distributed, including number of dance hours, number of injuries sustained, number of days taken to seek clinical care and days to return to dance.

Injury occurrence and characteristics

A total of 73% of all dancers included in this study reported experiencing a significant dance-related injury in the past 12 months, indicating a high period prevalence of injuries in both types of professional dancers in Australia. This finding aligns with previous research using similar methodology in company-based professional dancers from a range of styles including ballet,^{23 24} modern,²³ musical theatre²⁵ and Irish dance²⁶ and studies combining company-based and freelance modern dancers.²⁷ These injury period prevalence findings are also similar to other national dance injury investigations from the UK.²⁰

Part-time Australian dancers in this study were employed for an average of 19 weeks over the past 12 months. However 63.2% of them reported experiencing a dance-related injury over this time, compared with 79.8% of full-time dancers, suggesting that many injuries may not be due to a high dance load,²⁸ and thus may be preventable with appropriate training and dance-knowledgeable healthcare personnel. There is growing evidence in sport that undertraining may increase injury risk,^28–30 and this is a substantial consideration for part-time dancers who have unpredictable and variable dance employment contracts. Another aspect to this may be fatigue, as despite significantly different workloads, fatigue was the most commonly cited contributing factor to injury by both groups of dancers. Full-time dancers reported an intense dance training, performance and touring load, which could result in fatigue or overtraining syndrome if these were not offset by adequate rest periods.⁵ On the other hand, part-time dancers commonly reported multiple roles within the dance industry, and therefore a need to juggle multiple jobs. Other Australian-based research has also found that the lack of financial and job security in the sector requires artists to maintain other forms of employment.³¹ The ad hoc nature of part-time work also makes periodisation difficult.³² Finally, other research has proposed that athletes and dancers may experience injuries due to psychological factors as well as physical causes.^{33 34} This includes stress and a lack of social support,³⁵ coping skills³⁴ and other occupational stressors such as low work control,³⁶ which may all be relevant issues for professional dancers. Further research is required to better understand the epidemiology of injury in freelance dance artists, however it is essential that future research and interventions also consider these non-physical contributors to injury.

The ankle and the knee were among the three most commonly injured regions in both part-time and full-time dancers, consistently with other literature on dance injuries.¹⁶ However, differences were observed in prevalence of hip injuries, which represented the greatest proportion of significant injuries in part-time dancers (20%) but only 5.6% of injuries in full-time dancers, despite ballet being their most common dance style. This finding could be explained by the successful implementation of a hip strength and stability programme by major Australian dance companies, which has resulted in a reduction in time loss injuries and workers’ compensation claims relating to hip pain.³⁷ It is most likely that other differences in region of injury observed in this study are explained by dance genre, for example, the higher proportions of shoulder injuries in part-time dancers who most commonly practise contemporary dance and ankle injuries in full-time dancers who commonly practise ballet.

Treatment and management of dance-related injuries

Early diagnosis through prompt and quality healthcare is essential for injury recovery.³⁶ However, this can come at a price and more part-time dancers were required to pay for their own injury treatment compared with full-time dancers (63.9% compared with 23.9%, p=0.005). Yet despite this cost and inconvenience healthcare appears to be prioritised, as it only took approximately 1 day longer for injured part-time dancers to see a clinician (2 days compared with 1 day). On the other hand, there was still a large range in time taken for all part-time dancers to see someone for their injury (1–60 days, IQR=13) and previous research has highlighted some psychological and practical barriers to seeking care for dance-related injuries.^{38 39} Therefore, an in-depth qualitative rather than quantitative approach may be best placed to further investigate why it took so long for some dancers to seek care, and to ensure that future clinical services are accessible for all Australian dancers.

A greater proportion of injured full-time dancers in this study were unable to dance in any capacity following their injury, 21.4% compared with 5.7% of part timers (p=0.014). This could be explained by a variety of factors, besides the theory that injuries sustained by full-time dancers are more severe, and thus require full rest from dancing. Clinicians accessed by part-time dancers were less likely to be specialised in dance or associated with dance employment, therefore part timers may be dancing on injuries where dance-specialised clinicians would have provided more tailored advice not to. On the other hand, there are more practical barriers for the part-time dancer to stop dancing, such as being unable to audition for a follow-up job, not having access to paid sick leave and smaller companies not resourcing multiple casts or understudies. Therefore, a complete rest from dance may result in shuffling of other dancers’ roles or cancelled performances, which an injured dancer may be reluctant to impose.⁴⁰

Study strengths and limitations

This is the first national study in Australia to incorporate freelance dancers, which is a growing and understudied group. It is also the first known study to compare injuries in part-time dancers with full-time dancers, and presents a first step towards understanding the injury experience of a broader scope of professional dance practise. There are also limitations, including the retrospective, self-report study design. However previous research in professional athletes found an adequate level of accuracy of retrospective self-report for the number, site and type of sports injuries sustained in the preceding 12 months.⁴¹ Due to this decision to focus on a significant injury of the past 12 months, data on previous injuries and/or employment conditions were not incorporated into this study. All possible attempts were made to distribute this questionnaire as widely as possible, however it may not be a representative sample of Australian dancers. It is likely that the most independent and/or disconnected members of the dance community did not participate in this project. Additionally, there were small numbers of injured dancers, particularly part-time dancers. Therefore some injury characteristics, even when statistically non-significant, may be real with a large enough sample size to detect them. This sample size also precluded additional sub-analysis by gender or dance style, both of which influence injury rates and characteristics. Finally, although researchers attempted to collect detailed dance exposure data, the recorded responses were not complete, and therefore we were unable to report injury rates in the best practice format of injuries per 1000 dance exposure hours.^{42 43} Dancers were asked a series of questions regarding their number of hours spent in class, rehearsal and performance during a performance and non-performance weeks, however these questions were not always fully completed. In particular a typical week of dance activity was difficult for part-time and freelance dancers to report on, and this needs to be considered in the design of future studies in this group for the standardisation of reporting dance injury epidemiology research.

Clinical implications

In order to prevent and manage injuries in professional dance, it is vital for clinicians to understand the dance specific stressors that lead to different injuries.⁸ Part-time dancers often access general community clinicians, who may need additional education and support to practically advise on appropriate modification and/or return to dance, while also balancing the psychosocial implications of injury and practical demands of maintaining and thriving in a dance career. Other countries have established specialised clinics, such as the London NHS Dance Injury Clinic in the UK,⁴⁴ specifically designed for part-time and freelance dancers. This study provides evidence that such as service would meet a need in Australia and potential funding models should be further investigated.