This study focused on identifying the factors associated with increased risk of PRMD onset within a 12-month period among a study population of music students at different educational stages (i.e. from Pre-college students to university-level students). The present study is unique in being longitudinal in design and its associated factor data was obtained from a population of music students from different levels of educations and different music universities in Europe.
PRMD incidences within 6 and 12 months among music students were 28.8% and 49%, respectively. This is similar to the figure of 59% reported by Ballenberger et al.40 for the 33 music students free of baseline musculoskeletal health complaints (MHC) enrolled in their prospective longitudinal analysis. Although Ballenberger et al.’s study involved a smaller sample of 33 music students from a music university in Germany and 30 non-music controls, it is the only study that can be compared directly with the present study due to the absence of prospective longitudinal studies in the literature25-27. Although descriptive results pertaining to the singles incidences according to the academic level could suggest that the transition to higher musical training (i.e. both from Pre-college to Bachelor studies and from Bachelor 3&4-year group and Masters levels) might have played a role in the development of PRMDs (see Figure 4), both bivariate and multivariable analyses did not report any significant association. While the results of the cross-sectional baseline study revealed a significant relation between PRMDs’ prevalence and being a first- or a second-year Masters student14, the results of the present study offered contradictory evidence, showing that educational level was not influential in a 49.0% incidence of PRMDs within a 12-month period. Nevertheless, the loss-to-follow-up and diminishing sample size along with some suggestion of heteroscedasticity amongst the follow-ups should be considered within the interpretation of these results. It is plausible that under better circumstances, slight but potentially relevant differences in PRMD incidence amongst the levels of training might be resulted statistically significant.
In order to address an identified gap among current evidence, the main aim of this study was to explore the factors associated with increased risk of PRMD onset among music students. Similarly to what has emerged in the previous article regarding baseline analysis14, bivariate analysis showed that coming from West Europe was significantly associated with PRMD onset in the 6-month analysis. However, this finding was neither endorsed within the 12-month analysis (refer to Table 4) nor in the multinomial regression analysis (see Table 5). As this is the first longitudinal multicentre study amongst music students from different nationalities, its design makes it difficult to compare the results to those in earlier studies on MSK complaints among musicians coming from different countries. Future studies should explore this aspect more in detail, especially focusing on different cultures, which may reflect students’ behaviours and attitudes toward health in relation to their predisposition for PRMD onset.
The distinction between PRMDs and non-PRMDs described in the previous article regarding baseline analysis14 assumed even greater significance in this last longitudinal study, with a new interpretation that is different in its nature compared to antecedents within the literature view of evidence. For the initial cohort of students without PRMDs (n=442), students with generic MSK complaints have been included because it was assumed that this could represent an important risk factor for the onset of PRMDs40,41. At the same time, an attempt was made to offer a more focal consideration on MSK conditions and symptoms that affect the playing of music, allowing the identification of the specific factors associated with the onset of PRMDs. Interestingly, PRMD onset was associated with the absence of a MSK complaint not interfering with musicians’ playing ability (i.e. non-PRMD) at baseline in the bivariate analysis (6 and 12-month analysis). In the multinomial regression analysis (both 6 and 12-month analysis), the AOR of 0.415 for the baseline MSK complaints indicates that, keeping all the other variables at a fixed value, students with a MSK complaint at baseline assessment have a 58.5% (1-0.415=0.585) lower likelihood to have a PRMD onset (p<0.01; see Table 5). The latter means that the probability of developing a PRMD increases in music students who have not yet developed symptoms. On the other hand, self-reporting MSK symptoms was found to be potentially protective for the onset of a complaint interfering with their playing ability (i.e. PRMD). Although the latter would seem a counterintuitive and unexpected finding, it could be speculated that students with a MSK complaint at baseline had probably found effective strategies (e.g. change of playing technique or behaviours toward health) to accommodate the effects of such complaints on their ability to play their instrument at their best. Conversely, those who have not yet developed any disorder are not familiar with any comping strategy, setting into motion maladaptive behaviours with their symptom negatively influencing their playing activity.
Accordingly, sub-grouping musicians with all MSK complaints (i.e. PRMDs and non-PRMDs) might have profound implications for research and potentially for clinical practice. For instance, considering them as a homogeneous group, regardless of their interference with musicians’ playing ability could be problematic and leading to inaccurate treatments, as minor, mild and clinically irrelevant disorders are also included. Given the association between PRMD onset and the absence of symptoms, it is likely that the prevalence of PRMDs among this population will be reduced when specific treatment efforts to prevent the onset of PRMDs are made. As such, after verifying the reproducibility of this study’ observed findings among other populations or replicating the same approach with similar samples, it should be more likely to identify individuals with the same conditions and similar biopsychosocial characteristics to better inform treatment selection and mitigate the risk for developing such conditions. According to Zaza and Farewell41, having suffered from previous PRMDs was found to be positively associated with recurrent symptoms. In addition, musicians who experienced PRMDs are more prone to change their playing-related behaviours as a result of PRMDs41. Before this can happen, it would be worthwhile to assess and monitor pre-PRMD practising behaviours as soon as possible, in order to help them in the process of changing their attitudes and habits according to the factors that are more strongly correlated with the onset of PRMD symptoms. For instance, the findings of the present study might help to classify modifiable factors that can be adjusted in order to provide an improved conceptual framework for future studies that will more effectively investigate whether reduced injury minimising the onset of PRMDs is possible.
Interestingly, a change (i.e. either increase or reduction) in the level of physical activity was associated with the onset of PRMDs in both 6-month (c2(2)=12.03, p<0.01) and 12-month (c2(2)=7.92, p<0.05) bivariate analysis, as well as in both 6-month and in the 12-month multinomial regression analysis (AOR > 1). This finding was inconsistent with earlier cross-sectional studies, which found no associations between physical activity and the presence of MSK complaints6,13,19,42-47. Nevertheless, since results were related to a change in the physical activity participatory level, rather than to an increase or decrease in physical activity intensity, they should be interpreted cautiously. For example, this could be interpreted as an adaptive behaviour in which musicians decrease their physical activity due to the fear of worsening their health conditions or because their symptoms hinder its execution. Since musicians already perform an activity with their body by playing the instrument, too high and potentially incorrect physical activity (e.g. with wrong intensity) may lead to overload, thus decreasing coping strategies leading to a progressive or sudden decrease in activity. Instead, musicians should counterbalance and manage loads well with the right intensity by performing physical activity that complements their activity with the musical instrument. However, due to the limited recall period (i.e. the past 7 days) of this self-reported measure, it is difficult to determine exactly the cause and effect in this relationship. Future prospective studies with an adequate experimental design sensitivity will be able to provide new results that might be able to understand the extent of the association and the actual cause-effect relationship between these two variables. In addition, further studies should explore the knowledge among music students and professionals toward physical activity, as well as ways for improving strength and conditioning to perform at full potential. In relation to the latter point, unfortunately, average knowledge of physical activity is poor amongst university-level music students19, who tend not to score well on standardised tests of physical fitness and healthy lifestyles48, especially in relation to health responsibility and stress management49.
Moreover, the findings showed that PRMD onset was significantly associated with higher increase in the socially-prescribed perfectionism (SPP) score in the 6-month bivariate analysis (z=-3.12, p<0.01; see Table 4) and in the 6-month multinomial regression analysis (AOR > 1; see Table 5). Similarly, PRMD onset was associated with higher increase of other-oriented perfectionism (OOP) in the 6-month analysis bivariate analysis (z=-2.49, p<0.05; see Table 4). Interestingly, PRMD onset was also associated with lower rates of self-oriented perfectionism (SOP) (z=2.38, p<0.05; see Table 4), albeit only within the bivariate analysis at 6-months. According to Hewitt and Flett50,51, socially-prescribed perfectionism and other-oriented perfectionism are both focused on social expectations and demands, which are somewhat external (e.g. parental expectations). Specifically, the former represents the perception (veridical or not) that other people expect perfection from the individual, whereas the latter encompasses the demand that other people meet very high standards52. In addition, perfectionistic inclinations are characteristics of elite performers that often increase during the progress to higher levels of performance, as happens among athletes53. Given the nature of their activity, which is focused on the public or school expectations depending whether they are students or professionals, the association between PRMD onset with both SOP and OOP suggests that perfectionism could influence the onset of physical symptoms. Amongst these competitive environments, music students should therefore acquire effective strategies to cope with the high expectations (especially from others) that might result in some sort of maladaptive perfectionism49, leading to increased levels of psychological distress54 and degenerating into disruptive physical symptoms. Nonetheless, current research is scarce concerning the association between PRMDs and perfectionism. The evidence on perfectionism resides only within a restricted number of studies that have considered perfectionism levels and sometimes, contrasting results have emerged54,55.
Furthermore, similar to what emerged in the results pertaining to the baseline14, PRMD onset was significantly correlated with an increase in the levels of fatigue (i.e. CFQ 11 score) in both 6-month (z=-3.84, p<0.001; see Table 4) and 12-month (z=-2.71, p<0.01; see Table 4) bivariate analyses and both in the 6-month and 12-month multinomial regression analyses (AOR > 1; see Table 5). As fatigue might result from physical, cognitive and emotional exertion56,57, it might be assumed that many musicians experience MSK symptoms attributable to a biomechanical dimension, possible due to overuse of tissues involved in the act of playing and therefore related to their musical activities through nociceptive (e.g. repeated and constant load) or neuropathic (e.g. repetitive mechanical load resulting in damage to the peripheral nerve) mechanisms3. Biomechanical approaches to the assessment of movement tasks and fatigue should be combined with appropriate patterns of MSK complaints to quantify the strain generated in specific tissues during playing activity. Future research should focus on effective approaches to identify successful strategies to address painful conditions by developing prediction models that might be used to build safe and effective training guidelines for musicians58. A proper strategy for PRMD prevention and fatigue management could be a potential contribution within a healthier educational context and might reduce the impact of PRMDs among music students aspiring to become professional musicians59. Indeed, based on self-reported PRMDs’ rates from the current literature, it seems that the prevalence of MSK complaints amongst professionals and music students is relatively unchanged over recent decades27. This is despite the prevention strategies that had been established in Europe60-65 and the UK66,67. It could be speculated that PRMD rates might be related to insufficient health promotion and inefficient prevention awareness during music students’ training. This indicates that better results could be obtained by addressing health awareness taking into consideration the findings of the present study and the latest research findings from prospective longitudinal studies to make them effective and functional resources for music students’ training. In order to optimise the effectiveness of prevention and clinical approaches, a close collaboration between music institutions and healthcare professionals is fundamental. Especially in primary and secondary prevention, proactive behaviours, such as providing music students with information about their condition and facilitating management options68, are key prerogatives for an effective preventive and clinical practice.
Limitations
There are limitations to be aware of when considering the study’s design. As outlined previously28, the invitation for participants to complete the questionnaire was sent by the school registries, without the possibility of reinforcing the invitation by sending a reminder in another form (e.g. via a telephone interview). Reminder e-mails were sent to the school registries, along with flyers to be distributed at the conservatories, in order to persuade students participating in the research to complete the questionnaires at the baseline and at the follow-ups. Despite extensive attempts to keep students actively engaged in the study, some students dropped out during the first and second follow-ups. Hence, missing data might have seriously compromised the interpretation and the meaning of results. Beyond any doubts, the use of panel data analysis could have contributed to a better understanding of the cause-effect relationship between PRMD onset and all the variable considered, while providing potentially superior estimates. Indeed, the initial plan was to use all the variables included in the model of the cross-sectional baseline analysis14 and explore whether the longitudinal analyses involving a panel model could have provided superior estimates as compared to cross-sectional models of association. However, a panel analysis could not be conducted as it would have required a larger sample size within the second follow-up in order to establish adequate experimental design sensitivity, which due to the inevitable drop-outs, could not have been assured, or attained. Indeed, the participants included in the second follow-up were only 118 and this reduction might have decreased the power of the tests, especially because the variables considered as potential contributors/factors were many (i.e. 21). Therefore, considering the reduced sample in the second follow-up with so many variables and the main aim of the study (i.e. the evaluation of the factors associated with the onset of PRMDs), a logistic regression model for predicting the onset of PRMDs was preferred, as outlined in the description of the statistical analysis. Overall, this approach to modelling of the data offered a best compromise for interrogating as large a sample size as was possible effectively (n=442), taking into account both the onset of PRMDs at 6 months and that at 12 months. Although the factors considered could be interpreted only as ‘associated factors’ rather than ‘causal factors’ (also taking into account the limited time of one-year follow-up), this analysis could be nevertheless considered a step forward in the literature, especially considering its longitudinal approach.
Moreover, due to the lack of a direct contact with students, relevant information from non-respondents or drop-out students had not been accessible, which could have been used to assess for the intrusion of biases within the study’s results. For instance, the authors cannot exclude a potential sampling bias as the information concerning the number of students enrolled in each school participating in the study is not available because it consists of confidential data, without a formal permission to publish. In addition, another limitation consists of the impossibility to control every information pertaining to the participants, as well as the institutional level of behaviours or attitudes toward prevention. For instance, engaging in health-prevention programmes could represent a potential confounder that might have affected the results. Nevertheless, despite the limitations and the exploratory nature of the data, the present study provided an original contribution for the interpretation of PRMDs and their associated burden among music students of different nationalities and at different levels of training.