Correlation between risk factors and musculoskeletal ...

Occupational Medicine 2011;61:90–95 Advance Access publication on 26 January 2011 doi:10.1093/occmed/kqq196

Correlation between risk factors and musculoskeletal disorders among classical musicians Department of Occupational Therapy, Tel Aviv University, PO Box 39040 Ramat Aviv, Tel Aviv, 69978 Israel. Correspondence to: Y. Kaufman-Cohen, Department of Occupational Therapy, Tel Aviv University, PO Box 39040 Ramat Aviv, Tel Aviv, 69978 Israel. Tel: 1972 3 6405446; fax: 1972 3 6409933; e-mail: [email protected] ...................................................................................................................................................................................

Background Musculoskeletal disorders in humans may originate from biomechanical, environmental, psychosocial and personal risk factors encountered in the working environment. These disorders in musicians are designated as playing related musculoskeletal disorders (PRMD). ...................................................................................................................................................................................

Aims

To investigate the correlation between biomechanical, environmental, psychosocial and personal risk factors and potential incidence of PRMD arising in professional classical musicians.

...................................................................................................................................................................................

Methods

Fifty-nine orchestral classical musicians were observed: They also filled out questionnaires providing information regarding musculoskeletal symptoms, psychosocial factors and demography.

...................................................................................................................................................................................

Results

Clinical observation using the Rapid Upper Limb Assessment (RULA) showed significantly higher scores in string musicians compared with woodwind and brass players (right RULA score F 5 4.77, P , 0.05; left RULA score, F 5 3.90, P , 0.05). A multivariate regression model showed statistical significance for five of the six dependent variables regarding prevalence of painful joints, severity of musculoskeletal symptoms and functional limitation. The regression analysis revealed that biomechanical risk factors, perceived physical environment risk factors, instrument weight and average playing hours per week, were the main predictors of PRMD.

...................................................................................................................................................................................

Conclusions The biomechanical risk factors that predict PRMD are mainly associated with the upper limbs. A high association between PRMD and clinical observation emphasizes the need for further investigation of these risk factors and to study possible implementations in order to define ‘prevention strategies’ for musical routines and patterns, as used by classical musicians. ...................................................................................................................................................................................

Key words

Biomechanical risk factors; classical musicians; musculoskeletal disorders; orchestras; playing related.

...................................................................................................................................................................................

Introduction Most professional classical musicians will suffer at some time during their career from a musculoskeletal disorder (MSD). As many as 12% of them have been reported to give up their profession permanently [1]. This phenomenon has been designated as playing related musculoskeletal disorder (PRMD) with a prevalence among musicians varying between 60 and 80% [2–5]. PRMD consists of non-specific muscle and tendon complaints that are attributed to inflammatory disorders of tendons and tendon sheaths and compression or entrapment of peripheral nerves in several areas [2,5–9]. The variation in prevalance is related to the type of instrument played [2,10]. Although specific research on PRMD risk factors has been undertaken [5,8,11–13], these studies did not exam-

ine the cumulative significance of all predictable risk factors for PRMD. Most research has concentrated on the prevalence of PRMD with reference to the biomechanical risk factors [14], and many studies are based solely on self-reported symptoms [12,15]. This study hypothesizes that the correlation between biomechanical, environmental, psychosocial and personal risk factors [observed by an occupational therapist (OT)] will predict PRMD among classical musicians.

Methods A cross-sectional study of classical musicians was performed. One of the authors (Y.K.-C.), attended rehearsals of four orchestras and invited every string and wind player to participate in the research. Cohen’s

The Author 2011. Published by Oxford University Press on behalf of the Society of Occupational Medicine. All rights reserved. For Permissions, please email: [email protected]

Downloaded from http://occmed.oxfordjournals.org/ at UNESP-Universidade Estadual Paulista JÃºlio de Mesquita Filho on January 27, 2016

Y. Kaufman-Cohen and N. Z. Ratzon

Y. KAUFMAN-COHEN AND N. Z. RATZON: MUSCULOSKELETAL DISORDERS AMONG CLASSICAL MUSICIANS 91

sessment’ (JC) and ‘quantitative job requirements’ with ‘perceived workload’ (JR and WL) were chosen for this study. PPE includes the perceived sound intensity, ambient temperature, local humidity, ventilation and illumination in the rehearsal and concert halls. JC, JR and WL are considered to be psychosocial risk factors. As per NIOSH’s suggestion, the risk factor derived from the relationship between the JR, WL and JC was named WL-JR/JC. The Rapid Upper Limb Assessment (RULA) [22] is a survey developed for use in workplaces where workrelated upper extremity disorders are reported. RULA assesses biomechanical and postural loading including the positioning of the hands/fingers on the instrument, angulations of the joints, degree of movement when the instrument is played and weight bearing on the whole body with particular attention to the neck, trunk and upper limbs. The RULA observations were undertaken by an OT who is a specialist hand therapist. The RULA was calculated for both hands and was included alternatively in three different ways in the regression analysis equation (both hands, the hand that obtained the higher score and the difference between the two hand scores—representing the asymmetric measure). Asymmetry of body posture (which is a well-known concept in ergonomics for biomechanical risk assessments) was selected for the regression analysis as both upper extremities of musicians were active and provided information regarding the impact of asymmetric upper extremity work [23]. The personal information questionnaire comprised biodemographic information including body mass index (BMI), medical history, household chores, hobbies, sport activities and assessment of playing characteristics (see Table 1).

Table 1. Descriptive statistics regarding the occupational history and observation results of orchestral classical musicians Variable

X

SD

Range

Number of years in the profession (n 5 59) Number of weekly playing hours in the orchestra (n 5 59) Number of daily playing hours besides the orchestra (n 5 59) Total of playing hours per day (n 5 59) Weight of musical instrument (n 5 59) String instruments (n 5 39) Wind instruments (n 5 20) Number of minutes for warm up prior to playing (n 5 27) Number of continuous playing hours without interval (n 5 59) Period of interval in minutes (n 5 42) RULA of upper right limb (n 5 59) String instrument Wind instrument All musicians RULA of upper left limb (n 5 59) String instrument Wind instrument All musicians

23.1 18.6 2 4.9

11.9 6 1.2 2.5

15–65 6–32 0–6 2–18

3.4 1.9 10.5 1.3 15.1

5.4 1.0 3.6 5.0 5.7

1–30 1–4 3–15 1–3 5–30

5.6 3.5 4.9

1.1 0.7 1.4

3–7 3–7 3–7

5.4 3.6 4.8

1.2 0.8 1.4

3–7 3–7 3–7


statistical power analysis [16] was implemented to calculate the multiple regressions for 10–15 independent variables with r2 of 0.25, with 80% power and a of 0.05 for the test. The determined sample size was 60 subjects (10 independent variables were finally included in the multiple regression analysis). The string and wind (includes woodwind and brass) players in this research were recruited from three major Israeli classical orchestras. The Standardized Nordic Questionnaire (SNQ) [17] considers musculoskeletal pain in various body regions over the preceding 12 months and has been shown to be both reliable and valid [18]. Since musicians can experience both biomechanical and postural overloading on the upper extremities, an addition to the original questionnaire was made using the same format of the SNQ. This reviewed the presence of pain in the muscles of the arm, elbow, forearm, palm and each of the fingers. This additional questionnaire was scored on its own and was found to be valid in another population (separate from the original SNQ) [19]. The additional questionnaire was validated against the Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire used with other populations and was found to have a correlation of r 5 0.53; P , 0.001 [19]. The DASH outcome measure [20] is a 30-item selfreport questionnaire designed to measure biomechanical function and symptoms in patients with any or several MSDs of the upper extremity. The DASH contains an optional item intended to measure performing artists such as musicians. The National Institute for Occupational Safety and Health (NIOSH) Generic Job Stress Questionnaire [21] was also administered. Four stressor factors (represented by 37 items from the NIOSH questionnaire) including: ‘perceived physical environment’ (PPE), ‘job control as-

92 OCCUPATIONAL MEDICINE

Results Four orchestras were invited to participate in the study but in one ,50% of the musicians consented to participate and it was excluded from the survey. The three orchestras included in the study employed 93 string and wind musicians on a regular basis, of whom 61 (66%) agreed to take part in the survey. Those who declined to take part were of similar age to those of the investigated population and in similar proportions of string and wind players. Two musicians (2%) were excluded from the study due to orthopaedic problems caused by traumatic injuries. Of the 59 instrumentalists, 39 (66%) were string players, 20 (34%) were wind players, 29 (49%) were men and 30 (51%) were women. Ages ranged between 26 and 66 years and the mean age was 42.9 years (SD 6 11.43). Eighty-three percentage of all the musicians in the study reported symptoms in at least one body region in the previous 12 months. Seventy-three percentages suffered PRMD in more than one joint. Fifty-eight per cent of those surveyed reported upper extremity pain during the past year. Sixty-eight per cent of these were string players and 32% wind players. No significant difference was found between those two groups with respect to age, gender, years of education, BMI, medical history and sporting activities. There was no statistically significant difference between string and wind musicians regarding the prevalence rate of pain in the upper extremities. The group surveyed suffered from a variety of PRMDs, with a particularly high prevalence of shoulder pain (55%), lower back pain (49%), upper back pain (42%) and neck pain (39%).

The RULA score was significantly higher among string players in comparison to wind players as shown in Table 1 (right RULA score F 5 4.77, P , 0.05; left RULA score, F 5 3.90, P , 0.05). According to this survey, an average of 5 h/day (SD 2.45, range 2–18) was devoted to playing the instrument, with an average of only 10 min ‘warm-up’ prior to playing (SD 3.63, range 3–15). Only half of the musicians dedicated time to warm-up exercises. Correlation analysis between PRMD and independent variables was significant for the following variables: average of playing hours in an orchestral framework (r 5 0.30, P , 0.05), warm-up prior to playing (r 5 20.55, P , 0.01), RULA difference (r 5 0.38, P , 0.01); weight of musical instrument (r 5 0.48, P , 0.01), psychosocial risk factors (r 5 0.35, P , 0.01), the perceived physical environment (r 5 0.39, P , 0.01), number of cigarettes smoked per day (r 5 0.86, P , 0.01), household chores (r 5 0.24, P , 0.05), hobbies (r 5 20.21, P , 0.05) and sporting activities (r 5 20.32, P , 0.05). The study hypothesis was only partially verified. The linear regression model revealed that biomechanical factors (RULA), perceived physical environment, individual playing characteristics and personal risk factors all predicted MSDs in the study population. Psychosocial risk factors (WL-JR/ JC) did not predict MSDs in the study population. Five of six regression analyses had significant R2 values, with an explained variance ranging between 7% and 27%. Four of five regressions predicted PRMD in the upper extremities. The greater the difference between the RULA scores for both limbs (RULA difference), the more musculoskeletal complaints and limitations of function were reported by the musicians. The regression analysis revealed that these variables included in the equation had significant b size that ranged between 25% and 42% (see Table 2). In addition to the described regression analysis above, two further analyses were completed (the RULA score for both hands and the RULA score of the hand that obtained the higher score). Therefore, the RULA was included alternatively in three different ways in the equation of the regression analysis (both hands; hand that obtained the higher score; the difference score resulting from retraction of the RULA score for the left hand from the RULA score for the right hand—representing the asymmetric measure). Results of these regression models were virtually unchanged for the three different alternatives. In addition, analysis of variance was performed to examine the gender effect on the dependent variables. No differences were found between males and females with respect to PRMD.

Discussion In our study, biomechanical and perceived physical environment risk factors were found to be the two strongest


Analysis of variance statistical analysis was used to compare the differences between the orchestras. All musicians provided informed consent and ethical clearance were received from the Tel-Aviv University board for this study. A structured clinical observation was performed during rehearsal time by one author in the following week. The biomechanical and postural loadings during instrument playing were documented on the RULA scale. All the questionnaires were renumbered prior to being filled in, allowing the participants to remain anonymous. Univariate analysis was performed and linear regression analysis (Stepwise regression model) was performed only with independent variables that showed a significant correlation with PRMD. Six regression analyses (four dependent variables of the SNQ from which three variables were of the additional SNQ and two of the DASH) were completed with 10 independent variables. These included biomechanical (RULA difference), environmental (perceived physical environment), psychosocial (WL-JR/JC), individual playing characteristics and personal risk factors (bio-demographic information).


Table 2. Regression analysis for the five significant dependent variables (n 5 59) Independent variable

R2

b

B

95% CI Lower bound

0.5 0.67 0.06

2.18 2.27 0.26

0.71 0.05

2.27 0.25

0.9

1.06

0.11 0.26 21.48 0.36 21.48 0.02

1.07 1.18 20.28 1.24 20.36 0.14

0.11

0.39

0.49

4.61

0.36

1.08

*P , 0.05; **P , 0.01; ***P , 0.001.

statistical predictors for PRMD in orchestral musicians. The RULA difference variable was the strongest predictor for PRMD in the upper extremity. High scoring on the RULA provided evidence of extreme body postures, excessive muscle activation, repetitive movements and static and dynamic loading being exerted on different anatomical regions. These risk factors combined with long playing hours heightened the risk of musculoskeletal pain. The literature presents these biomechanical risk factors as being causal of MSDs, in different vocational groups [1,14,22,24,25] and perceived physical and environmental risk factors existing in the rehearsal and concert halls may cause stress in musicians [1,13]. Environmental risk factors have both physiological and emotional influences, which might have an indirect impact on the musculoskeletal system [13,26,27]. The weight of the instrument may generate ergonomic issues related to both the static and the dynamic loading exerted on the musculoskeletal system. Static loading relates to the long periods of continuous muscular contraction required to hold the instrument and/or the maintenance of the upper limb in a particular position for a sustained period. Dynamic loading of the muscles, joints and other supporting anatomical struc-

tures results from the motion involved in playing the instrument. These efforts all have a cumulative effect on the musculoskeletal system [25,28]. Gender turned out to be the only individual variable that predicted PRMD; a finding that contrasts with the published literature, which indicates that female musicians report more MSDs than male musicians [12,13,28]. On the other hand, there is evidence that females can better perform tasks that require a higher level of muscle endurance such as increased hand grip strength with a flexed elbow [29,30]. Contrary to expectations, the psychosocial burden (WLJR/JC) was not found to predict musculoskeletal pain. In the univariate analysis, psychosocial risk factors correlated significantly with PRMD. In fact, the demands upon orchestral players, in terms of emotional investment and levels of performance required, are so intense that they may often have to ignore bodily discomfort and pain until after their playing is over [14]. The fact that anxiety measurements were not taken into account in this study might explain why the psychosocial risk factors did not predict PRMD. The prevalence of PRMD found in this study is comparable with the prevalence of 60 and 80% of


SNQ—number of symptomatic upper limb joints, in the past year RULA difference 0.13 0.38** 1.34 RULA difference 0.22 0.42** 1.47 Weight of musical 0.31** 0.16 instrument RULA Difference 0.27 0.42*** 1.49 Weight of musical 0.30** 0.15 instrument Average of weekly 0.25* 0.09 orchestra hours SNQ—functional limitations due to symptomatic upper limbs, in the past year RULA difference 0.07 0.30* 0.59 RULA difference 0.18 0.37** 0.72 Gender 20.35** 20.88 RULA difference 0.26 0.41** 0.80 Gender 20.36** 20.92 Weight of musical 0.30** 0.08 instrument SNQ—functional limitations due to symptomatic body regions, in the past year Perceived physical 0.14 0.39** 0.25 environment SNQ—severity of symptoms in the upper limb 0.08 0.31* 2.55 Warm up prior to playing DASH—physical function and symptoms in the upper limb Weight of musical 0.07 0.30* 0.60 instrument

Upper bound

94 OCCUPATIONAL MEDICINE

increased the opportunity of the participants to inform about their upper extremity symptoms. The additional questionnaire addressed the upper limbs (although used separately, in addition to the original SNQ and validated with the DASH) and thus might have affected the validity of the findings. In conclusion, both objective (RULA) and subjective perceived physical environmental risk factors (as perceived by the musicians) and biomechanical risk factors predicted PRMD, mainly with the upper extremities. This study emphasizes the need for a professional to investigate the risk factors that musicians are exposed to. Further studies are warranted, in order to define prevention strategies for musical routines and patterns, as used by orchestral classical musicians.

Key points • Playing related musculoskeletal disorders are a significant health problem associated mainly with the upper extremities within a sample of orchestral string and wind musicians from three orchestras. • The biomechanical risk factors (assessed by the Rapid Upper Limb Assessment tool for estimating extreme body positions, repetitive movements and load bearing) and the perceived physical environment were the two strongest statistical predictors for playing related musculoskeletal disorder among orchestral classical musicians. • There is a need for orchestral clinicians to investigate the risk factors and to implement preventive measures for musical routines and patterns as used by orchestral classical musicians.

Conflicts of interest None declared.

References 1. Parry CB. Prevention of musician’s hand problems. Hand Clin 2003;19:317–324. 2. Fry HJ. Incidence of overuse syndrome in the symphony orchestra. Med Probl Perform Art 1986;1:51–55. 3. Hoppman RA. Instrumental musician’s hazards. Occup Med (Lond) 2001;16:619–631. 4. Zaza C. Playing-related musculoskeletal disorders in musicians: a systematic review of incidence and prevalence. CMAJ 1998;21:1019–1025. 5. Bragge P, Bialocerkowski A, McMeeken J. A systematic review of prevalence and risk factors associated with playing-related musculoskeletal disorders in pianists. Occup Med (Lond) 2006;56:28–38.


PRMD among musicians reported in the published literature [2–4,23,28]. When considering the prevalence of pain in relation to the type of instrument played, 68% of those reporting musculoskeletal pain in the upper extremities were string players and 32% wind players that replicate almost exactly the International Conference of Symphony and Opera Musicians survey which reported a high frequency of musculoskeletal problems in string players (60%), compared with 32% of all wind players [29]. String players in this study generally suffered twice as much as the wind players from PRMD in the upper extremity. This resembles other epidemiologic findings published elsewhere [2–5]. The shoulders were the most commonly reported painful joints. Sixty-one per cent of the string players reported shoulder pain compared with 45% of the wind players. Brandfonbrener [14] and Hoppmann [3] claim, that although classical musicians often complain about primary or secondary shoulder pain, they often report pain in the arms, wrists and hands—the areas where the greatest muscular exertion occurs while performing. The dominancy of the exertion of the upper extremity was also expressed in the regression analysis. It was noted that only one dependent variable in the regression model was related to the entire body, whereas the remaining four variables dealt with PRMD solely in the upper extremities. There are some limitations to our study. Firstly, there may be selection bias due to participation rate of the study (66%). Another limitation is the exclusion of the variable ‘vibration’, a significant biomechanical risk factor, in the evaluation of influential biomechanical risk factors as a whole. To the best of our knowledge, no method has been found to suitably assess this risk factor. The anthropometric measurement used in this research was BMI. It appears that additional anthropometric data are needed that relates to the musicians, the physical measurements of the musical instrument, the height and distance of the music sheet stand and the type of chair used by each type of musician. In addition, we referred to the weight of each instrument but without presenting any specific model integrating variables such as in the case of the cello and double bass, which rest on the floor and therefore eliminated the effects of gravity. It must be noted that this is a cross-sectional study designed to predict a statistical concept rather than being a literal prediction. The cross-sectional study design precludes the drawing of conclusions about the causal relationship between risk factors and PRMD. Data regarding the prevalence of PRMD should be treated with caution since the musicians in this study might have reflected a ‘healthy worker’ effect, as it may be that some musicians had left the orchestras because of pain. The biomechanical risk factors, which predict PRMD are mainly associated with the upper limbs. This finding is probably dependent upon the additional SNQ tool, which


18. Palmer K, Smith G, Kellingray S, Cooper C. Repeatability and validity of an upper limb and neck discomfort questionnaire: the utility of the standardized Nordic questionnaire. Occup Med 1999;49:171–175. 19. Ratzon NZ, Mizrachi N. The presence of musculoskeletal disorders among amateur bowlers. Work 2008;30:369–375. 20. Hudak PL, Amadio PC, Bombardie C. The Upper Extremity Collaborative Group (UECG). Development of an upper extremity outcome measure: the DASH (disabilities of arm, shoulder and hand). Am J Ind Med 1996;29:602–608. 21. Hurrell JJ, Mclaney MA. Exposure to job stress—A new psychometric instrument. Scand J Work Environ Health 1988;14(Suppl.):27–28. 22. McAtamney L, Corlett EN. RULA: a survey method for the investigation of work-related upper limb disorders. Appl Ergon 1993;24:91–9. 23. Waters T, Putz Anderson V, Garg A, Fine L. Revised NIOSH equation for the design and evaluation of manual lifting tasks. Ergonomics 1993;36:749–776. 24. Vender MI, Kasdan ML, Truppa KL. Upper extremity disorders: a literature review to determine work-relatedness. J Hand Surg 1995;20:534–541. 25. Hoppmann RA. Instrumental musician’s hazards. Occup Med State 2001;16:619–631. 26. Hansen PA, Reed K. Common musculoskeletal problems in the performing artist. Phys Med Rehab Clin N Am 2006;17: 789–801. 27. Devereux JJ, Vlachonikolis IG, Buckle PW. Epidemiological study to investigate potential interaction between physical and psychosocial factors at work that may increase the risk of symptoms of musculoskeletal disorder of the neck and upper limb. Occup Environ Med 2002;59:269–277. 28. Norris RN. Applied ergonomics. In: Tubiana R, Amadio PC, eds. Medical Problems of the Instrumentalist Musician. London: Blackwell Science, 2000; 595–613. 29. Fishbein M, Middlestadt SE, Ottati V, Straus S, Ellis A. Medical problems among ICSOM musicians: overview of a National Survey. Med Probl Perform Art. 1988;3:1–8. 30. Strazdins L, Bammer G. Women, work and musculoskeletal health. Soc Sci Med 2004;58:997–1005.

Writing for Occupational Medicine Occupational Medicine seeks authors to write interesting or amusing filler articles for its white spaces. We welcome contributions on any topic but preferably those related to occupational medicine or medical matters. Or you may have an interesting story to tell about why you became an occupational physician. All contributions must be less than 500 words. If you have something to contribute please contact us at [email protected]


` M, Pe´rez-Soriano P, Camacho CI, 6. Cebria´ i Iranzo A Belloch SL, Cortell-Tormo JM. Playing-related musculoskeletal disorders in woodwind, brass and percussion players: a review. J Hum Sport Exerc 2010;5:94–100. 7. Winspur I. Advances in objective assessment of hand function and outcome assessment of the musician’s hand. Hand Clin 2003;19:483–493. 8. Bejjani FJ, Kaye GM, Benham BS. Musculoskeletal and neuromuscular conditions of instrumental musicians. Arch Phys Med Rehabil 1996;77:406–413. 9. Chan RF, Chow C, Lee GP et al. Self-perceived exertion level and objective evaluation of neuromuscular fatigue in a training session of orchestral violin players. Appl Ergon 2000;31:335–341. 10. Lederman RJ. Neuromuscular and musculoskeletal problems in instrumental musicians. Muscle Nerve 2003;27: 549–561. 11. Malchiare J, Roquelaure NC, Piette A, Vergrach S, Chiron H. Musculoskeletal complaints, functional capacity, personality and psychosocial factors. Int Arch Occup Environ Health 2001;74:549–557. 12. Zaza C, Farwell VT. Musician’s playing-related musculoskeletal disorders: an examination of risk factors. Am J Ind Med 1997;32:292–300. 13. Davis J, Mangion S. Predictors of pain and other musculoskeletal symptoms among orchestral instrumental musicians: elucidating specific effects. Med Probl Perform Art 2002;17:155–168. 14. Brandfonbrener AG. Musculoskeletal problems of instrumental musicians. Hand Clin 2003;19:231–239. 15. Bragge P, Bialocerkowski AE, McMeeken JM. Understanding playing related musculoskeletal disorders (PRMDs) in elite pianists: a grounded theory study. Med Probl Perform Art 2006;21:71–79. 16. Cohen J. Statistical Power Analysis for Behavioral Sciences. 2nd edn. Hillsdale, NJ: Lawrence Erlbaum Associates, 1983. 17. Kuorinka I, Koskinen P. Occupational rheumatic disease and upper limb strain in manual jobs in light mechanical industry. Scand J Work Environ Healt 1979;5:39–47.