Study design and participants
Participants in this cross-sectional study were professional and conservatory student double bassists and bass guitarists. The bassists were recruited in 2009 and 2013/14 from one Dutch professional orchestra, from three conservatories in the Netherlands and via the website of the International Society of Bassists (www.isbworldoffice.com; Accessed January 2014). Bassists were notified and recruited through their teachers or colleagues and were given further information by a researcher (AT). The group of bassists recruited via the website received information via a standard digital introduction text. Bassists (18 years and older) who were able to fill out the questionnaire in English or Dutch were included if they had graduated from, or were a student, at a conservatory.
All participants provided informed consent. Because of the type of study, involving a questionnaire and healthy volunteers, the Medical Ethics Committee decided that no approval was needed.
All potential participants received a web-based questionnaire, via a URL link. They received the same instructive e-mail and questionnaire, in Dutch or English, depending on their preference. The explanatory information provided to the participants at the start of the study informed them about the need for medical research among musicians, but not about the specific goal of the study.
In the absence of an existing questionnaire, we created a questionnaire suitable for measuring music-related issues among bassists (Online resource 1). Self-reported functioning, height and weight, physical and mental health status, pain location and pain intensity were assessed using questions from existing questionnaires. The combined questionnaire required approximately 20–25 min to complete.
Mono- and multi-instrumentalism
The bassists were asked whether they played bass guitar, double bass, both bass instruments and/or another instrument for at least 5 h a week each.
Mono/multi-instrumentalism was dichotomized into a score of ‘0’ (mono-instrumentalism) if the bassist played only one type of bass instrument without playing another instrument and ‘1’ (multi-instrumentalism) if they played more than one type of instrument. Bassists were assigned to the ‘multi-instrumentalism’ category if they reported playing for more than 20% of their playing time on their least frequently used bass instrument. They were also assigned to the multi-instrumentalism group if they indicated that they played another instrument for at least 5 h a week. The dichotomization in the case of another instrument was arbitrarily based on exposure to at least 1 h of occupational stress from the instrument nearly every workday. On average, a professional musician plays 1300 h of music a year, i.e. approximately 25 h a week (Paarup et al. 2011); a playing time of at least 5 h a week is, therefore, equivalent to at least one-fifth of the playing time being spent on one other instrument besides their main instrument. Studies (Abréu-Ramos and Micheo 2007; Benjjani et al. 1984; Hochberg and Lederman 1995; Wu 2007) have reported an association between the amount of playing time and the prevalence of MSC, finding that playing an instrument for at least 1 h a day or more had a significant impact.
The questions regarding MSC were divided into two time-related categories; ‘complaints occurring longer than 3 months ago’ and ‘complaints in the last 3 months’. Respondents ranked each item on a four-point scale ranging from ‘always’, ‘often’, ‘rarely’ to ‘never’. The intensity of pain during the last week was also measured using a Numeric Rating Scale (Hartrick et al. 2003) ranging from ‘no pain’ (score 0) to ‘worst pain’ (score 10). The location of MSC was assessed for the following parts of the upper body half (left or right): neck, back, shoulder, upper arm, elbow, forearm, wrist and/or fingers (see Woldendorp et al. 2015 for the exact definition of the body parts). The analysis was based on the data regarding ‘complaints in the last 3 months’. The pain intensity scores during the last week were used to characterize the population, but not for the analyses, as we assumed that the data from the last week would be too vulnerable to bias due to fluctuations over time.
Multiple bio-psycho-social factors have been reported to contribute to chronic pain in musicians (Bragge et al. 2006; de Souza et al. 2012; Pascarelli and Hsu 2001; Wu 2007), and were added to our analysis as potential confounders (for an overview of the potential confounders we studied, see Table 1 and Online Resources 2 and 3).
Data on playing characteristics was assessed using items 1.5, 2.7–2.10, 2.12, 2.13, 2.15–2.17, 5.1 and 5.2 of the International Society of Bassists ‘Body and Bass’ Survey (ISBS) (Gilbert 2008). The ISBS is a short, descriptive, non-validated questionnaire which is used to collect reliable information specifically relating to bassists (Gilbert 2008). It includes 42 items divided over 5 dimensions, regarding technical playing aspects, physical symptoms, mental/emotional symptoms, general information and two open questions allowing the opportunity to suggest anything that could diminish complaints.
The prevalent playing positions of the study participants were later researched on the Internet. General health status was assessed with one question of the Short Form 36 Health Survey [item 1 (Stewart and Ware 1992)]; referred to as ‘subjective health score’. Psychological distress was assessed with the Brief Symptom Inventory (BSI) (Derogatis and Melisaratos 1983). The BSI is the shortened version of the Symptoms Checklist-90, questioning physical and psychological symptoms across nine dimensions: somatization; obsession-compulsion; interpersonal sensitivity; depression; anxiety; hostility; phobic anxiety; paranoid ideation and psychoticism, plus a global score (Global Severity Index). The BSI contains 53 items. Participants rate each item on a 5-point scale ranging from 0 (not at all) to 4 (extreme).
Bassists can play their instrument either right- or left-handed. The movement patterns of the hand which sounds the strings (above the resonance box) via the fingers, a plectrum or bow are different from those of the hand (at the neck of the bass) responsible for the melody. For this reason, the terms ‘left’ and ‘right’ were substituted in the data analysis by ‘neck side’ and ‘box side’. The complaint scores related to the shoulder and forearm locations (due to elevated positioning of the arm in playing the double bass) were clustered as right and left ‘shoulder area’, respectively. Because all of the muscles inserted at the wrist originate from the forearm, complaint scores from the wrist and forearm (at the right side due to the flexed position of the wrist in playing the bass guitar) were clustered as right and left ‘wrist area’, respectively. The MSC scores were dichotomized into ‘no complaints’ (answering categories ‘no complaints’ or ‘rarely’ for the body region) and ‘complaints’ (‘often’ or ‘always’). Since the shoulder and wrist areas consisted of several separately scored body regions, the highest scores for frequency of complaints and pain intensity were taken as the score for that area.
The health-related items of smoking, alcohol use, drug abuse and body mass index (BMI) were dichotomized into an ‘objective health score’ of ‘healthy’ versus ‘unhealthy’. ‘Unhealthy’ was scored when at least one of the following was present: smoking more than 21 cigarettes or consuming more than 21 units of alcohol a week, using hard drugs (yes) and/or a BMI score lower than 18 or higher than 25. The data from the question about ‘playing another instrument for at least 5 h a week’ were dichotomized into a score of ‘0’ if ‘no’ and ‘1’ if ‘yes’.
Demographic and playing characteristics are presented as means and standard deviations (for interval/ratio data), medians (for ordinal data) and percentages (for nominal and dichotomized data). The interval/ratio data were tested for normal distribution (with the Shapiro–Wilk test because half of the subpopulations included <50 bassists).
As the first step in the analyses, we determined the differences in frequencies of MSC scores (during the last 3 months in the various joint areas of the upper body) between the multi- and mono-instrumentalists. Analyses were performed using the Chi-square test or Fisher’s exact test for dichotomous data. Fisher’s exact test was used instead of the Chi-square test if there was insufficient data in one or more cells. Body regions showing an association with MSC (or a tendency toward one) with p < 0.20 were selected for further analyses.
Secondly, relevant non-MSC variates were selected for the final step. Analyses were performed using the t test for normally distributed data, the Mann–Whitney U test for non-normally distributed data and the Chi-square test or Fisher’s exact test for dichotomous data, with the potential confounder as one variable and the MSC in the selected body areas as the dependent variable. The third and final step consisted of a backward stepwise logistic regression to ascertain the effects of multi/mono-instrumentalism and covariates on the likelihood that participants had MSC in the body areas found in step 1. Multi- or mono-instrumentalism, and the variables showing an association with a p value <0.20 in step 2, were entered as the independent variable, while MSC in the selected body area was entered as the dependent variable. Variables such as ‘playing both bass instruments’ and ‘playing another instrument’ are directly related to ‘multi-instrumentalism’, and ‘problems carrying equipment’ is a consequence rather than a covariate. Therefore, these variates were excluded from the final analysis.
Level of significance in the final models was set at p ≤ 0.05, two-tailed. All data were analysed using SPSS, version 20.