Use of psychotherapy and psychotropic medication among Norwegian musicians compared to the general workforce

Abstract

Previous research has reported higher prevalence rates of anxiety and depression among musicians, compared to the general workforce. We compared the use of psychotherapy and psychotropic medication with other major occupational groups, and expected to find higher use among musicians. Musicians from the Norwegian Musicians’ Union (n = 1,607) answered an online questionnaire about demographic characteristics, mental health, use of healthcare services and use of psychotropic medication. They were compared to a sample of the Norwegian workforce (n = 2,550) from the Norwegian survey of level of living. Based on chi-square and logistic regression analysis, adjusted for age, sex, education, and cohabitation status, we found that musicians reported higher use of psychotherapy and psychotropic medication. Use of psychotherapy was reported most frequently among vocalists, while rock musicians reported the most widespread use of psychotropic medication. Overall, musicians had three-fold higher odds of use of psychotherapy and 50% higher odds of use of psychotropic medication compared to the general workforce. This is consistent with previous findings indicating high rates of sleep-difficulties and psychological distress among musicians. The results underline the importance of investigating both the content and quality of services provided.

Keywords

healthcare services mental health musicians psychotherapy

Musicians face a number of physical, social, and psychological challenges during their careers (Cooper & Wills, 1989; Kenny & Ackermann, 2009; Vaag, Giæver, & Bjerkeset, 2013) and studies indicate that mental health problems are reported frequently among musicians compared to other professions (Fishbein, Middlestadt, Ottati, Straus, & Ellis, 1988; Kyaga et al., 2013; Raeburn, 1987; Raeburn, Hipple, Delaney, & Chesky, 2003; Tynes, Eiken, Grimsrud, Sterud, & Aasnæss, 2008; Vaag, Bjørngaard, & Bjerkeset, 2016). A Swedish 40-year prospective population-based study indicated that musicians were more likely to suffer from bipolar disorder and anxiety than were the matched controls (Kyaga et al., 2013). It has also been reported that rock and pop stars from Europe and North America have an almost two-fold increased mortality rate, 3 to 25 years post fame, compared to a demographically matched sample (Bellis et al., 2007).

To date, the largest survey on mental health problems and use of healthcare services among musicians is a study of 2,212 orchestral musicians in the USA (Fishbein et al., 1988). The survey was given in conjunction with the International Conference of Symphony and Opera Musicians (ICSOM). Their results showed that symptoms of psychological problems, such as acute anxiety (13%) and depression (17%), were reported frequently. Among musicians who reported symptoms of stage fright (24%), 40% had tried prescribed medication, 25% psychological counseling, 17% aerobic exercise, 13% hypnotics and 11% had consulted their general practitioner. A recent qualitative review of 96 autobiographical books, written in English by recognized rock artists and published between 1974 and 2010, found that 62% of the artists described substance addiction and 40% had participated in drug rehabilitation (Oksanen, 2013).

Findings from Germany indicate that mental distress is higher among opera and orchestra musicians, compared to the general population (Voltmer et al., 2012), while two recent studies from Brazil and Australia have also reported high rates of symptoms of depression and anxiety (Barbar, de Souza Crippa, & de Lima Osório, 2014; Kenny, Driscoll, & Ackermann, 2014). Among the 230 professional and amateur musicians included in the study from Brazil, 28% reported medication use, but only 3% reported use of antidepressants and/or anxiolytics (Barbar et al., 2014). In Australia, 377 full-time symphony musicians were asked to answer what kind of measures they had undertaken in order to relieve music performance anxiety. Among them, 6% reported having consulted a psychologist, 4% a psychiatrist, 5% reported use of anxiolytics, 4% antidepressants and as many as 31% reported having used beta-blockers (Kenny et al., 2014).

The recent findings underline the importance of the growing interest in the mental health and use of healthcare services among musicians, but there is lack of large-scale studies directly comparing the relative difference in healthcare use between musicians with other occupational groups.

In our survey of 1,667 Norwegian musicians we previously reported considerably higher levels of sleep difficulties and insomnia symptoms (Vaag, Saksvik-Lehouillier, Bjørngaard, & Bjerkeset, 2015), as well as anxiety and depression symptoms (Vaag et al., 2015) among musicians, compared to the general population and workforce. In this study, we aim to investigate whether these differences are reflected in the self-reported use of psychotherapy and psychotropic medication, using the same cross-sectional sample of musicians, compared to the general workforce.

Based on current knowledge, we have the following hypotheses:

H1: Musicians will have a higher self-reported use of psychotherapy (having consulted a psychologist and/or psychiatrist during the last 12 months) than the general workforce.

H2: Musicians will have a higher self-reported use of psychotropic medication than the general workforce.

In addition, we want to investigate the use of psychotherapy and medication among different groups of musicians (i.e., with regard to type of employment, instrument and genre), differences between musicians and different occupational groups and tendencies in use of mental healthcare grouped by different levels of psychological distress.

Method

Participants and setting

Sample of musicians

Between 1 February and 1 April 2013, a total of 4,168 members of the Norwegian Musicians’ Union were invited to participate in an online survey about their psychosocial work environment, mental health and use of healthcare services. Due to the expected high prevalence of teachers working as part-time musicians, members listed as both musicians and/or teachers were invited. Three reminders were sent, with two-week intervals. A total number of 2,121 members (51%) responded. Among them, there were 1,016 (48%) women and 1,105 men, with a mean age of 44.5 years (SD = 10.7). Of these, 1,607 (76% of the eligible) members confirmed that they had been working as musicians in the last 12 months. In our data analysis, the sample of performing musicians was compared to a workforce sample from the health-related version of the Norwegian survey of level of living 2012 (www.ssb.no/en).

Workforce sample

The sample comprised 9,771 subjects (16 years+) who were randomly drawn from the Norwegian population register and invited to participate in the Norwegian survey of level of living in 2012. The initial data collection was done using computer-assisted telephone interviews, after which the participants were invited to fill in an initial postal or web-based questionnaire. A total of 5,660 participated in the interviews (58%), of which 4,015 people (71% of the eligible) also completed the questionnaire between October 2012 and January 2013. Of these, 2,610 (65%) were listed as employed with an International Standard Classification of Occupations (ISCO)-08-code representing major occupational groups. Sixty respondents had not responded on the measure of psychological distress (HSCL-25) and were removed from analysis. Among the remaining workforce sample of 2,550 people, 88 respondents had minor missing data on psychological distress. In total, 121 data points were missing in the study variables (0.2%). The expectation-maximization technique in SPSS (IBM Corporation, 2010) was performed to impute values on these missing data points. This sample of 2,550 workers was used as reference group in our analysis. A documentation report describing the Norwegian survey of level of living in more detail is available (Amdam & Vrålstad, 2012).

Measures

Use of psychotherapy

The Norwegian healthcare system is financed by taxation and is based on the principles of universal access, decentralization, and free choice of provider (Ringard, Sagan, Saunes, & Lindahl, 2013). In addition, there are private healthcare providers offering mental healthcare services. These are not financially supported by the government healthcare system, but may be financed by insurance companies. In this study we do not differentiate between public and private healthcare. Nevertheless, our results provide data on the use of healthcare services among musicians within the principal context of universal access.

Use of healthcare services was measured by asking each respondent if they had consulted a psychologist or psychiatrist within the last 12 months (“During the last 12 months, have you visited…” “…a psychologist,” “…a psychiatrist?” yes/no). A dichotomous variable was created and respondents that had been visiting one or both of the professional providers were coded as 1.

Use of psychotropic medication

Use of psychotropic medication was measured by asking each respondent: “During the last 4 weeks, how often have you used the following:” to answer if they had been using each of the following psychotropic medications: sedatives, antidepressants, hypnotics and/or Attention Deficit Hyperactivity Disorder (ADHD) medication. The response alternatives were “not used during the last 4 weeks,” “less than weekly use,” “weekly, but not daily use,” and “daily use.” Self-reported use of one or more of these medications was dummy-coded as 1.

Anxiety, depression, and psychological distress

Symptoms of anxiety and depression were measured using a 25-item version of The Hopkins Symptom Checklist 25 (Derogatis, Lipman, Rickels, Uhlenhuth, & Covi, 1974; Strand, Dalgard, Tambs, & Rognerud, 2003), which is derived from the HSCL-90 (Derogatis et al., 1974). HSCL-25 is a self-administered and widely used instrument measuring symptoms of anxiety and depression across clinical and research settings. All 25 questions measure psychological distress, but the items can also be divided into two subscales: anxiety (10 items) and depression (15 items). Each item (symptom) is measured on a Likert scale from 1 to 4, where 1 indicates no symptoms, 2 slight presence, 3 substantial presence, and 4 severe presence of the symptom. An index variable based on the mean score of HSCL-25 symptoms was created (α = .93). Mean score above 1.75 has been recommended as a cut-off defining prevalence of psychological distress (Strand et al., 2003). In order to compare the use of psychotherapy and psychotropic medication in regards to different levels of psychological distress, we categorized the mean score of HSCL-25 symptoms into lowest degree of symptoms (under 1.20), low degree of symptoms (1.20–1.55), moderate degree of symptoms (1.55–1.75), and high degree of symptoms (1.75 and above). The lowest group represented the ones who scored under the median of 1.20, while 1.55 and 1.75 were chosen as cut-offs for moderate and high degree of symptoms according to previous suggestions of grouping cases of psychological distress into “probable psychiatric problem” (1.55) and “probable need for psychiatric treatment” (1.75) (Nettelbladt, Hansson, Stefansson, Borgquist, & Nordström, 1993).

Statistics

Data were analyzed using STATA, version 12.0 (StataCorp, 2011). Descriptive statistics and chi-square tests were used to compare the use of healthcare and psychotropic medication between the study groups. Two separate logistic regression analyses were conducted using psychotherapy and use of psychotropic medication as dependent variables. We compared all musicians, and different subgroups of musicians (based on employment, setting, instrument, and genre), with the workforce sample. Analyses were done on the raw data as well as adjusted for sex, age, education, and cohabitation status. An additional analysis was conducted investigating differences between musicians and other major occupational groups (according to the ISCO-08 classification system). In addition to calculating odds ratios (OR), we also estimated prevalence differences (PD) with 95% confidence intervals (CI) by calculating marginal effects using mfx in STATA. The prevalence differences were estimated holding other variables at mean levels.

Results

The sample of musicians was mainly within the ages of 25 to 66 and slightly younger than the general workforce (Table 1). There were also a higher proportion of men among the musicians (57% vs. 51%), and the sample of musicians had a significantly higher education level compared to the general workforce. A slightly higher proportion of the general workforce (91.5%) reported using their regular GP compared to musicians (88.3%).

Table 1.

Descriptive statistics: Demographic characteristics, use of psychotherapy and psychotropic medication among Norwegian musicians (n = 1,607) and workforce (n = 2,550).

	Workforce (n = 2,550)	Musicians (n = 1,607)	ES¹
Age mean (CI)	45.0 (44.5–45.5)	43.1 (42.6–43.6)
Women (%)	1289 (50.5)	691 (43.0)	.07
Education (%)			.49
Unknown/no education	30 (1.2)	7 (.4)
Primary	339 (13.3)	9 (.6)
High school	1026 (40.2)	117 (7.3)
University 3 years	825 (32.4)	722 (44.9)
University 5 years+	330 (12.9)	752 (46.8)
Cohabitation status (%)			.07
Married	1378 (54.0)	858 (53.4)
Cohabitant	464 (18.2)	373 (23.2)
Single	708 (27.8)	376 (23.4)
Use of Regular GP (%)	2389 (91.5)	1419 (88.3)	.05
Psychotherapy use (%)	103 (4.0)	175 (10.9)	.13
Psychologist	91 (3.6)	149 (9.3)	.12
Psychiatrist	22 (.9)	39 (2.4)	.06
Use of psychotropic medication (%)	215 (8.6)	162 (10.1)
Sedatives	43 (1.7)	50 (3.1)	.05
Antidepressants	78 (3.1)	49 (3.0)
Hypnotics	159 (6.3)	101 (6.3)
ADHD medication	5 (.2)	7 (.4)
HSCL-25 Distress (%)	211 (8.3)	281 (17.5)	.14
Psychotherapy	43 (20.5)	79 (28.4)	.09
Psychotropic med.	59 (28.6)	67 (23.8)
HSCL-25 Depression (%)	255 (10.0)	323 (20.1)	.14
Psychotherapy	49 (19.3)	85 (26.7)	.09
Psychotropic med.	69 (27.6)	75 (23.2)
HSCL-25 Anxiety (%)	182 (7.1)	236 (14.7)	.12
Psychotherapy	32 (17.6)	65 (27.9)	.12
Psychotropic medication	56 (31.3)	62 (26.3)

Chi-square tests with Cramer’s V listed as effect size measure (ES) when p < .05.

Among musicians, there was a higher self-reported use of a psychologist, a psychiatrist, and psychotherapy in total (Table 1). The differences between musicians and workforce, with regard to use of psychotherapy, were also consistent when looking at respondents scoring above the 1.75 cut-off for symptoms of psychological distress, anxiety, and depression (HSCL-25). In contrast, except for the use of sedatives, few differences in use of psychotropic medication were seen between the two groups in descriptive data.

Use of psychotherapy

Logistic regression analysis, adjusting for age, sex, education, and cohabitation status showed that musicians had an almost three times higher odds of use of psychotherapy (aOR 2.86 (2.11–3.88), aPD 5.8 (3.2–7.7)) than the sample of the general workforce. Vocalists (aOR 3.86 (2.60–5.72), aPD 11.0 (6.2–15.8)), freelance musicians (aOR 3.34 (2.42–4.77), aPD 8.6 (5.4–11.8)) and lead performers (aOR 3.10 (2.08–4.62), aPD 7.0 (4.0–10.1)) reported the highest use during the last 12 months, but the tendency of musicians reporting higher use of psychotherapy was seen almost across all different groups of musicians (Table 2).

Table 2.

Logistic regression analysis of use of psychotherapy (psychiatrist and/or psychologist) during the last 12 months in Norwegian musicians (n = 1,602) compared to the general workforce (n = 2,547). Estimated odds ratio (OR) and prevalence differences (PD) with 95% CI.

	Use (%)	Crude		Adjusted¹
		OR (CI)	PD (CI)	OR (CI)	PD (CI)
Workforce (n = 2,547)	103 (4.0)	1 (Reference)		1 (Reference)
All Musicians (n = 1,602)	175 (10.9)	2.91 (2.26–3.74)	6.9 (5.2–8.6)	2.86 (2.11–3.88)	5.8 (3.9–7.7)
Employed (n = 304)	24 (7.9)	2.03 (1.28–3.23)	5.2 (1.0–9.5)	2.02 (1.23–3.32)	4.5 (.5–8.5)
Freelance (n = 726)	92 (12.7)	3.44 (2.57–4.62)	10.1 (7.0–13.2)	3.34 (2.42–4.77)	8.6 (5.4–11.8)
Combined (n = 572)	59 (10.3)	2.73 (1.95–3.81)	7.9 (4.5–11.2)	2.64 (1.80–3.86)	6.5 (3.2–9.9)
Solo/lead performer (n = 396)	50 (12.6)	3.43 (2.40–4.90)	10.9 (6.6–15.2)	3.10 (2.08–4.62)	8.4 (4.3–12.5)
Ensemble/band (n = 733)	76 (10.4)	2.74 (2.02–3.74)	7.7 (4.8–10.6)	2.86 (2.01–4.09)	7.0 (4.0–10.1)
Orchestra (n = 295)	28 (9.5)	2.49 (1.61–3.85)	7.3 (2.7–11.9)	2.38 (1.48–3.83)	5.9 (1.6–10.2)
Other (n = 178)	21 (11.8)	3.17 (1.93–5.21)	10.4 (4.0–16.8)	3.03 (1.79–5.14)	8.5 (2.7–14.4)
Vocal (n = 311)	53 (17.0)	4.87 (3.42–6.95)	16.0 (10.7–21.2)	3.86 (2.60–5.72)	11.0 (6.2–15.8)
Violin (n = 259)	30 (11.6)	3.11 (2.02–4.77)	9.8 (4.6–14.9)	2.75 (1.72–4.40)	7.3 (2.6–12.0)
String (n = 253)	18 (7.1)	1.82 (1.08–3.05)	4.2 (–.3–8.6)	2.21 (1.26–3.87)	5.2 (.4–10.1)
Woodwind (n = 202)	15 (7.4)	1.90 (1.09–3.34)	4.6 (–.4–9.6)	1.79 (.99–3.24)	3.5 (–.9–8.0)
Brass (n = 215)	15 (7.0)	1.78 (1.02–3.12)	4.0 (–.8–8.8)	1.80 (.99–3.26)	3.6 (–.9–8.1)
Key instrument (n = 227)	24 (10.6)	2.81 (1.76–4.47)	8.6 (3.2–13.9)	2.90 (1.75–4.81)	7.9 (2.6–13.2)
Drums (n = 102)	11 (10.8)	2.87 (1.49–5.53)	9.1 (1.0–17.1)	3.21 (1.61–6.43)	9.3 (1.1–17.5)
Other (n = 33)	9 (27.3)	8.90 (4.03–19.63)	29.4 (11.8–47.1)	9.45 (4.07–21.93)	28.0 (9.8–46.3)
Pop (n = 106)	11 (10.5)	2.78 (1.44–5.35)	8.9 (.9–16.8)	2.55 (1.28–5.07)	6.8 (–.2–13.8)
Rock (n = 83)	6 (7.2)	1.85 (.79–4.34)	4.5 (–3.4–12.4)	2.13 (1.92–5.74)	5.1 (–2.8–13.0)
Jazz (n = 180)	11 (10.5)	2.97 (1.79–4.91)	9.5 (3.3–15.7)	3.32 (1.92–5.74)	9.6 (3.1–16.0)
Classical (n = 764)	82 (10.8)	2.87 (2.12–3.88)	8.1 (5.2–11.0)	2.64 (1.85–3.78)	6.3 (3.3–9.2)
Other (n = 473)	56 (11.8)	3.19 (2.26–4.48)	9.8 (5.9–13.7)	3.27 (2.23–4.78)	8.8 (4.9–12.7)

Adjusted for age, sex, education and cohabitation status.

The higher degree of use of psychotherapy among musicians was also evident when comparing musicians to all other ISCO-classified groups of occupations such as managers, academic professionals, technicians, and clerical support workers, even after controlling for age, sex, education, and cohabitation status (Table 3).

Table 3.

Logistic regression analysis of use of psychotherapy (psychiatrist and/or psychologist) among Norwegian musicians (n = 1,607), compared to different occupational groups (total n = 2,547). Estimated odds ratio (OR) and prevalence differences (PD) with 95% confidence intervals.

	No. distress	Raw		Adjusted¹
	Use (%)	OR (CI)	PD (CI)	OR (CI)	PD (CI)
Musicians compared to^a	175 (10.9)	1 (Reference)	Reference	1 (Reference)	Reference
1 Managers (n = 244)	6 (2.5)	4.86 (2.13–11.10)	10.7 (3.8–17.6)	3.77 (1.62–8.76)	7.6 (1.8–13.4)
2 Professionals (n = 813)	42 (5.2)	2.25 (1.59–3.19)	4.9 (2.6–7.2)	2.50 (1.73–3.59)	4.9 (2.7–7.1)
3 Technicians (n = 450)	15 (3.3)	3.56 (2.08–6.09)	8.2 (4.1–12.2)	3.35 (1.90–5.91)	6.8 (3.0–10.5)
4 Clerical support (n = 151)	5 (3.3)	3.58 (1.45–8.85)	8.2 (1.3–15.2)	3.89 (1.50–10.06)	7.8 (1.1–14.4)
5 Service/sales (n = 441)	24 (5.5)	2.12 (1.36–3.29)	4.5 (1.6–7.3)	2.67 (1.51–4.74)	5.3 (1.8–8.9)
6/7 Farm/Crafts² (n = 232)	4 (1.7)	6.99 (2.57–19.02)	14.0 (4.7–23.3)	4.68 (1.62–13.54)	9.1 (1.3–17.0)
8/9/0 Other³ (n = 219)	7 (3.2)	3.71 (1.72–8.01)	8.5 (2.6–14.4)	2.97 (1.28–6.89)	6.0 (.6–11.4)

Each occupational profession, according to the ISCO (numeric codes), is used as reference group and compared to the sample of musicians.

Adjusted for age, sex, education and cohabitation status.

Agricultural workers, fishermen and crafts.

Military, cleaners, machine operators and other.

Figure 1 illustrates how self-reported use of psychotherapy differed between musicians and workforce, grouped by gradually increasing symptom levels of psychological distress. There was a clear tendency of musicians having consulted psychotherapy to a higher degree than the workforce, seen across different grouped levels of symptoms of distress reported at the time of the study.

Figure 1.

Use of psychotherapy (Predictive margins with 95% CI) among musicians (n = 1,602) and workforce (n = 2,547) grouped by level of psychological distress (HSCL-25) and adjusted for age, sex, education and cohabitation status.

Use of psychotropic medication

Overall, adjusted analysis showed higher use of psychotropic medication (aOR 1.55 (1.19–2.02), aPD 3.5 (1.3–5.7)) among musicians (Table 4). This was especially evident among rock musicians (aOR 4.45 (2.46–8.05), aPD 20.0 (8.4–31.7)) and string instrumentalists (aOR 2.60 (1.69–4.00), aPD 10.3 (4.2–16.4)). Key instrument players (aOR 2.10 (1.36–3.24), aPD 7.4 (1.9–12.9)) and musicians combining freelance work and employment (aOR 1.70 (1.21–2.41), aPD 4.8 (1.2–8.4) also reported higher use.

Table 4.

Logistic regression analysis of more than weekly use of psychotropic medication (sedatives, antidepressants, hypnotics and ADHD medication) among Norwegian musicians (n = 1607) compared to the general workforce (n = 2,498). Estimated odds ratio (OR) and prevalence differences (PD) with 95% confidence intervals.

	Use (%)	Raw		Age/sex/ed. adjusted
		OR (CI)	PD (CI)	OR (CI)	PD (CI)
Workforce (n = 2,498)	215 (8.6)	1 (Reference)		1 (Reference)
All Musicians (n = 1,607)	162 (10.1)	1.19 (.96–1.47)	1.5 (–.4–3.3)	1.55 (1.19–2.02)	3.5 (1.3–5.7)
Employed (n = 306)	35 (11.4)	1.37 (.94–2.00)	2.9 (–1.0–6.8)	1.61 (1.07–2.42)	4.3 (.0–8.7)
Freelance (n = 727)	67 (9.2)	1.08 (.81–1.44)	.6 (–1.8–3.1)	1.43 (1.03–1.97)	3.0 (.0–6.0)
Combined (n = 574)	60 (10.5)	1.24 (.92–1.68)	1.9 (–.9–4.7)	1.70 (1.21–2.41)	4.8 (1.2–8.4)
Solo/lead performer (n = 397)	39 (9.8)	1.16 (.81–1.66)	1.3 (–2.0–4.5)	1.41 (.95–2.09)	3.0 (–.8–6.7)
Ensemble/band (n = 734)	61 (8.3)	.96 (.72–1.30)	−.3 (–2.7–2.1)	1.34 (.96–1.87)	2.4 (–.6–5.4)
Orchestra (n = 298)	29 (9.7)	1.14 (.76–1.72)	1.2 (–2.5–4.9)	1.50 (.97–2.34)	3.6 (–.9–8.1)
Other (n = 178)	33 (18.5)	2.42 (1.61–3.62)	10.1 (4.1–16.0)	2.79 (1.80–4.30)	11.5 (4.9–18.1)
Vocal (n = 311)	34 (10.9)	1.30 (.89–1.91)	2.4 (–1.4–6.1)	1.46 (.97–2.21)	3.3 (–.8–7.4)
Violin (n = 260)	24 (9.2)	1.08 (.69–1.68)	.6 (–3.2–4.5)	1.33 (.83–2.14)	2.4 (–2.0–6.8)
String (n = 253)	33 (13.0)	1.59 (1.08–2.36)	4.5 (.1–8.9)	2.60 (1.69–4.00)	10.3 (4.2–16.4)
Woodwind (n = 202)	17 (8.4)	.98 (.58–1.63)	−.2 (–4.4–4.0)	1.16 (.67–1.99)	1.2 (–3.4–5.7)
Brass (n = 217)	13 (6.0)	.68 (.38–1.21)	−2.8 (–6.3–.7)	.94 (.51–1.72)	−.5 (–4.8–3.9)
Key inst. (n = 227)	32 (14.1)	1.74 (1.17–2.60)	5.6 (.8–10.4)	2.10 (1.36–3.24)	7.4 (1.9–12.9)
Drums (n = 103)	5 (4.9)	.54 (.22–1.35)	−4.0 (–8.5–.5)	.89 (.35–2.26)	−.5 (–7.2–5.6)
Other (n = 34)	4 (11.8)	1.42 (.49–4.06)	3.3 (–8.0–14.6)	2.06 (.70–6.03)	7.3 (–6.7–21.4)
Pop (n = 106)	10 (9.4)	1.11 (.57–2.15)	.9 (–5.1–6.8)	1.81 (.91–3.62)	5.7 (–2.4–13.9)
Rock (n = 83)	16 (19.3)	2.54 (1.44–4.45)	11.0 (2.1–19.9)	4.45 (2.46–8.05)	20.0 (8.4–31.7)
Jazz (n = 180)	15 (8.3)	.97 (.56–1.67)	−.3 (–4.7–4.1)	1.40 (.79–2.46)	2.9 (–2.7–8.5)
Classical (n = 764)	78 (10.2)	1.21 (.92–1.59)	1.6 (–.8–4.1)	1.46 (1.05–2.01)	3.2 (.2–6.1)
Other (n = 474)	43 (9.1)	1.06 (.75–1.49)	.5 (–2.5–3.4)	1.31 (.90–1.90)	2.3 (–1.1–5.6)

Adjusted for age, sex, education and cohabitation status.

Differences between musicians and other professions were most evident when compared to managers (aOR 2.80 (1.50–5.22), aPD 8.7 (2.8–14.6)) and technicians (aOR 2.15 (1.35–3.40), aPD 6.3 (2.2–10.3)), while there were smaller differences compared to academic professionals, clerical workers, and service and sales workers (Table 5).

Table 5.

Logistic regression analysis of use of psychotropic medication (sedatives, antidepressants, hypnotics and ADHD medication) among Norwegian musicians (n = 1,607), compared to different occupational groups (n = 2,498). Estimated odds ratio (OR) and prevalence differences (PD) with 95% confidence intervals.

	No. distress	Raw		Adjusted¹
	Use (%)	OR (CI)	PD (CI)	OR (CI)	PD (CI)
Musicians compared to ^a	162 (10.1)	1 (Reference)	Reference	1 (Reference)	Reference
1 Managers (n = 240)	12 (5.0)	2.13 (1.17–3.89)	6.6 (.9–12.3)	2.80 (1.50–5.22)	8.7 (2.8–14.6)
2 Professionals (n = 796)	74 (9.3)	1.09 (.82–1.46)	.7 (–1.6–3.1)	1.37 (1.01–1.86)	2.5 (.0–4.9)
3 Technicians (n = 447)	26 (5.8)	1.82 (1.18–2.79)	5.1 (1.2–9.0)	2.15 (1.35–3.40)	6.3 (2.2–10.3)
4 Clerical support (n = 147)	19 (12.9)	.76 (.45–1.26)	−2.2 (–6.2–1.7)	1.05 (.59–1.85)	.4 (–4.0–4.7)
5 Service/sales (n = 430)	55 (12.8)	.76 (.55–1.06)	−2.1 (–4.7–.4)	1.00 (.65–1.53)	.0 (–3.3–3.2)
6/7 Farm/crafts² (n = 227)	13 (5.7)	1.85 (1.03–3.31)	5.3 (–.1–10.6)	1.98 (1.04–3.79)	5.6 (–.1–11.2)
8/9/0 Other³ (n = 211)	16 (7.6)	1.37 (.80–2.33)	2.6 (–2.0–7.2)	1.77 (.97–3.22)	4.6 (–.5–9.6)

Each occupational profession, according to the ISCO (numeric codes), is used as reference group and compared to the sample of musicians.

Adjusted for age, sex, education and cohabitation status.

Agricultural workers, fishermen and crafts.

Military, cleaners, machine operators and other.

Figure 2 shows the self-reported use of psychotropic medication among musicians and workforce, grouped by different symptom levels of psychological distress (HSCL-25). The confidence intervals were largely overlapping, and no clear difference was seen between the two groups on use of medication.

Figure 2.

Use of psychotropic medication (predictive margins with 95% CI) among Norwegian musicians (n = 1,607) and workforce (n = 2,498) grouped by level of psychological distress (HSCL-25) and adjusted for age, sex, education, and cohabitation status.

Discussion

In this study, musicians reported a three-fold increased odds of use of psychotherapy and 50% higher odds of use of psychotropic medication, compared to the general workforce. These findings are in line with our a priori hypotheses (H1 and H2), and consistent with our previous findings indicating increased rates of psychological distress in musicians (Vaag et al., 2015). Musicians reported higher use of psychotherapy than all other major occupational groups, and higher use of psychotropic medication than managers, technicians, academic professionals, and farmers/craft workers. Of note, our results also indicate considerable differences in use of psychotherapy and psychotropic medication across subgroups of musicians.

Strengths and limitations

To our best knowledge, this is the first large-scale study directly comparing the use of psychotherapy and psychotropic medication in musicians to a sample of the general workforce. The large sample size also made it possible to study differences regarding type of employment, instrumental group and music genre, as well as comparing musicians to different occupations. The data collections were conducted with close proximity in time (Workforce sample: October 2012–February 2013/Musicians sample: February–April 2013) and the same self-report instruments were used in both samples.

Major limitations in this study are the cross-sectional design and the lack of information about the incidence, magnitude, and length of healthcare and medication use. Further, information on the content of the psychotherapeutic services provided would have been useful. Our outcome measures are based on retrospective self-report of use of psychotherapy (last 12 months) and use of psychotropic medication (last month), and the results are therefore susceptible for recall bias. Nevertheless, there is little reason to believe this operates differentially in musicians and the workforce. In other words, (absolute) prevalence estimates might be affected by recall bias in both groups, yet we expect the relative differences between the study groups to be largely unaffected.

Use of psychotherapy and psychotropic medication in musicians

We have previously reported that the prevalence of anxiety and depression symptoms was two times higher in musicians (Anxiety: 14.7%/Depression: 20.1%) compared to the general workforce (Anxiety: 7.1%/Depression: 10.0%) (Vaag et al., 2015). This is reflected in the increased use of psychotherapy, and the pattern seems consistent across different groups of musicians. Overall, the pattern in use of psychotropic medication was less pronounced than for the use of psychotherapy, except from rock musicians, string and key instrumentalists. These subgroups reported 2–4 fold increased odds of use of psychotropic medication compared to the workforce.

Unfortunately, it is difficult to compare our results with previous research on musicians’ use of psychotropic medication, especially due to the fact that previous research has mainly focused on medication use in conjunction with music performance anxiety (MPA; Kenny et al., 2014). In our study, respondents were not specifically asked if they used psychotherapy and/or medication in order to relieve MPA. However, we did ask if the musicians had ever used alcohol or medication to deal with MPA; 15% answered “yes” (not shown in results) and 7% confirmed that they had been using such strategies during the last 12 months. Among classical musicians, the proportions were 19% and 9%. Although methods and research settings differ, the numbers are substantially lower than the reports of use of beta-blockers among Australian orchestral musicians (Kenny et al., 2014) and might indicate a lower degree of medication use for MPA in Norway. Nevertheless, our results generally show an increased use of psychotropic medication compared to the general workforce.

Our results also show that musicians tend to seek help from a psychiatrist and/or psychologist in higher numbers than the workforce at lower degrees of distress. Since the respondents were asked if they had consulted a psychiatrist and/or psychologist during the last 12 months, while the measure of psychological distress was based on symptoms experienced in the last week, it is important to interpret these findings with great caution. But the results could suggest differences in healthcare seeking tendencies.

Research from the late 1980s and early 1990s showed that musicians reported lower symptom load than other patients seeking outpatient psychotherapeutic healthcare (Cohen & Kupersmith, 1986; Miller & Kupersmith, 1990). These observations could indicate that musicians have a lower threshold of seeking psychotherapeutic help than other occupational groups, but it may also suggest that musicians respond better to psychotherapy, or that musicians consult psychotherapy based on problems other than just anxiety and depression-related issues (such as MPA without other anxiety and depression symptoms).

Our sample of musicians reported a high educational level, which in turn has been linked to more widespread use of psychotherapeutic healthcare (Jokela, Batty, Vahtera, Elovainio, & Kivimäki, 2013; Roness, Mykletun, & Dahl, 2005; Svensson, Nygård, Sørensen, & Sandanger, 2009). Norway is a large but sparsely populated country, and musicians living in different regions might have different access to healthcare. Both educational institutions and music-related jobs are mostly found in urbanized areas, where healthcare services are more easily accessible. Nevertheless, research comparing the general population in rural (Lofoten, Northern Norway) and urbanized (Oslo) areas in Norway has indicated that there are not substantial differences in mental healthcare seeking across regions (Svensson et al., 2009). Due to a lack of exact overlap between the membership regions of the Norwegian Musicians’ Union and the regions used in the Norwegian level of living surveys, we were unable to control for this in our analysis. A sub-analysis of three comparable regions in our dataset, however, showed a higher prevalence of psychotherapy use in the most densely populated region, for both musicians and workforce (not shown in results). In the Oslo region (Oslo and Akershus Counties) 7% of the workforce sample (n = 655), and 12.0% of musicians (n = 628) in a mostly overlapping region (Oslo, Akershus and Østfold Counties) reported use of a psychologist and/or psychiatrist. In the middle part of Norway (Sør- and Nord-Trøndelag Counties), the corresponding numbers were 3% (n = 228) and 9% (n = 205), while they were 1% (n = 227) and 8% (n = 168) for Northern Norway (Nordland, Troms and Finnmark Counties). These results indicate a reduction in healthcare use in less populated areas and reduction is seen in both groups, although to a lesser extent among musicians. One possible reason may be that musicians live or travel (being more mobile due to concerts and touring) near urbanized areas to a higher extent than other workers, and therefore have the potential of increased access to healthcare.

Besides different demographic and settlement patterns, it is also possible that psychotherapy is well suited for many musicians. Personality research has indicated that musicians tend to score high on both neuroticism and openness to experience (Cooper & Wills, 1989; Dyce & O’Connor, 1994; Gillespie & Myors, 2000; Shuter-Dyson, 2000). These studies are based on small samples of musicians and limited use of reference samples. But if there is a tendency of higher level of neuroticism among musicians, this could be a contributing factor to the high levels of symptoms of sleep and distress reported in this group (Griffith et al., 2010; Gurtman, McNicol, & McGillivray, 2013; Kotov, Gamez, Schmidt, & Watson, 2010; Malouff, Thorsteinsson, & Schutte, 2005; van de Laar, Verbeek, Pevernagie, Aldenkamp, & Overeem, 2010). Further, high levels of openness to experience could be related to an increased interest in the exploratory and imaginative features of psychotherapy as proposed by some scholars (Widiger & Mullins-Sweatt, 2010; Widiger & Presnall, 2013). To date, research has not been conducted investigating these hypotheses.

There is also research suggesting that musicians are better at decoding emotions than others (Thompson, Schellenberg, & Husain, 2004), and a recent Swedish study investigating musical activity and emotional competence among 8,000 Swedish twins found a significant association between musical engagement and emotional competence (Theorell, Lennartsson, Mosing, & Ullén, 2014). Emotional competence is thought to be associated with willingness to seek help when in need (Ciarrochi & Deane, 2001). Openness to experience and emotional competence could thus, at least partly, explain the openness to psychotherapy in our study, but further studies need to examine this hypothesis.

Conclusion and implications for further research

A substantially higher proportion of musicians seek psychotherapy and use of psychotropic medication compared to the general workforce. Vocalists, rock musicians, string players, and keyboard instrumentalists might represent subgroups of special interest in terms of psychotherapy and psychotropic drug use. These groups of musicians were also the ones reporting the highest degrees of psychological distress in our previously reported study (Vaag et al., 2015), and the overall results also correspond our previous findings regarding the increased amount of psychological distress (Vaag et al., 2015) and sleep difficulties (Vaag et al., 2015) among musicians, compared to the general workforce.

We have only described the amount of mental health services, yet very little is known about the quality and content of the services musicians receive. Future research should follow a sample of musicians over time, and include measurements of incidence of mental health symptoms and patterns of healthcare use, controlling for a larger amount of confounding variables than was possible in this study. New knowledge should contribute to development of more specific and tailored preventive measures and psychological treatments (i.e., for MPA) for musicians.

Footnotes

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: the Norwegian Extra Foundation for Health and Rehabilitation through EXTRA funds.

References

Amdam

Vrålstad

(2012). Levekårsundersøkelsen om helse, omsorg og sosial kontakt 2012. Oslo: Statistics Norway. Retrieved from http://www.ssb.no/helse/artikler-og-publikasjoner/_attachment/159601?_ts=143bdd4f9c0

Barbar

A. E. M.

de Souza Crippa

J. A.

de Lima Osório

(2014). Performance anxiety in Brazilian musicians: Prevalence and association with psychopathology indicators. Journal of Affective Disorders, 152–154(2014), 381–386. doi:10.1016/j.jad.2013.09.041

Bellis

Hennell

Lushey

Hughes

Tocque

Ashton

J. R.

(2007). Elvis to Eminem: Quantifying the price of fame through early mortality of European and North American rock and pop stars. Journal of Epidemiology and Community Health, 61(10), 896–901. doi:10.1136/jech.2007.059915

Ciarrochi

J. V

Deane

F. P.

(2001). Emotional competence and willingness to seek help from professional and nonprofessional sources. British Journal of Guidance and Counselling, 29(2), 233–246.

Cohen

B. J.

Kupersmith

J. R. F.

(1986). A study of SCL-90 scores of 87 performing artists seeking psychotherapy. Medical Problems of Performing Artists, 1(4), 140–142.

Cooper

C. L.

Wills

G. I. D.

(1989). Popular musicians under pressure. Psychology of Music, 17(1), 22–36.

Derogatis

L. R.

Lipman

R. S.

Rickels

Uhlenhuth

E. H.

Covi

(1974). The Hopkins Symptom Checklist (HSCL): A self-report symptom inventory. Behavioral Science, 19(1), 1–15.

Dyce

J. A.

O’Connor

B. P.

(1994). The personalities of popular musicians. Psychology of Music, 22(2), 168–173.

Fishbein

Middlestadt

S. E.

Ottati

Straus

Ellis

(1988). Medical problems among ICSOM musicians: Overview of a national survey. Medical Problems of Performing Artists, 3(1), 1–8.

10.

Gillespie

Myors

(2000). Personality of rock musicians. Psychology of Music, 28(2), 154–165.

11.

Griffith

J. W.

Zinbarg

R. E.

Craske

M. G.

Mineka

Rose

R. D.

Waters

A. M.

Sutton

J. M.

(2010). Neuroticism as a common dimension in the internalizing disorders. Psychological Medicine, 40(7), 1125–36. doi:10.1017/S0033291709991449

12.

Gurtman

C. G.

McNicol

McGillivray

J. A.

(2013). The role of neuroticism in insomnia. Clinical Psychologist, 18(3), 116–124. doi:10.1111/cp.12029

13.

IBM Corporation. (2010). IBM SPSS Statistics 19 [Computer software]. Armonk, NY: IBM Corporation.

14.

Jokela

Batty

Vahtera

Elovainio

Kivimäki

(2013). Socioeconomic inequalities in common mental disorders and psychotherapy treatment in the UK between 1991 and 2009. British Journal of Psychiatry, 202, 115–120. doi:10.1192/bjp.bp.111.098863

15.

Kenny

D. T.

Ackermann

(2009). Optimizing physical and psychological health in performing musicians. In Hallam

Cross

Thaut

(Eds.), The Oxford handbook of music psychology (pp. 390–400). Oxford: Oxford University Press.

16.

Kenny

Driscoll

Ackermann

(2014). Psychological well-being in professional orchestral musicians in Australia: A descriptive population study. Psychology of Music, 42(2), 210–232. doi:10.1177/0305735612463950

17.

Kotov

Gamez

Schmidt

Watson

(2010). Linking “big” personality traits to anxiety, depressive, and substance use disorders: A meta-analysis. Psychological Bulletin, 136(5), 768–821. doi:10.1037/a0020327

18.

Kyaga

Landén

Boman

Hultman

C. M.

Långström

Lichtenstein

(2013). Mental illness, suicide and creativity: Forty-year prospective total population study. Journal of Psychiatric Research, 47(1), 83–90. doi:10.1016/j.jpsychires.2012.09.010

19.

Malouff

Thorsteinsson

Schutte

(2005). The relationship between the five-factor model of personality and symptoms of clinical disorders: A meta-analysis. Journal of Psychopathology and Behavioral Assessment, 27(2), 101–114. doi:10.1007/s10862–005–5384-y

20.

Miller

D. K.

Kupersmith

J. R.

. (1990). “Louisville-PACH”: Psychiatric problems of performing artists. Medical Problems of Performing Artists, 5(1), 19–22.

21.

Nettelbladt

Hansson

Stefansson

C.-G.

Borgquist

Nordström

(1993). Test characteristics of the Hopkins Symptom Check List-25 (HSCL-25) in Sweden, using the Present State Examination (PSE-9) as a caseness criterion. Social Psychiatry and Psychiatric Epidemiology, 28(3), 130–133. doi:10.1007/BF00801743

22.

Oksanen

(2013). Addiction and rehabilitation in autobiographical books by rock artists, 1974–2010. Drug and Alcohol Review, 32(1), 53–59. doi:10.1111/j.1465–3362.2012.00474.x

23.

Raeburn

S. D.

(1987). Occupational stress and coping in a sample of professional rock musicians. Medical Problems of Performing Artists, 2(3), 77–82.

24.

Raeburn

S. D.

Hipple

Delaney

Chesky

(2003). Surveying popular musicians’ health status using convenience samples. Medical Problems of Performing Artists, 18(3), 113–119.

25.

Ringard

Sagan

Saunes

I. S.

Lindahl

A. K.

(2013). Norway: Health system review. Health Systems in Transition, 15(8), 1–162.

26.

Roness

Mykletun

Dahl

A. A.

(2005). Help-seeking behaviour in patients with anxiety disorder and depression. Acta Psychiatrica Scandinavica, 111, 51–58. doi:10.1111/j.1600–0447.2004.00433.x

27.

Shuter-Dyson

(2000). Profiling music students: Personality and religiosity. Psychology of Music, 28(2), 190–196. doi:10.1177/0305735600282008

28.

StataCorp. (2011). Stata statistical software: Release 12 [Computer software]. College Station, TX: StataCorp LP.

29.

Strand

B. H.

Dalgard

O. S.

Tambs

Rognerud

(2003). Measuring the mental health status of the Norwegian population: A comparison of the instruments SCL-25, SCL-10, SCL-5 and MHI-5 (SF-36). Nordic Journal of Psychiatry, 57(2), 113–8. doi:10.1080/08039480310000932

30.

Svensson

Nygård

J. F.

Sørensen

Sandanger

(2009). Changes in formal help seeking for psychological distress: The OsLof study. Nordic Journal of Psychiatry, 63(3), 260–266. doi:10.1080/08039480902880105

31.

Theorell

T. P.

Lennartsson

A.-K.

Mosing

M. a

Ullén

(2014). Musical activity and emotional competence —a twin study. Frontiers in Psychology, 5(Article 774): 1–8. doi:10.3389/fpsyg.2014.00774

32.

Thompson

W. F.

Schellenberg

E. G.

Husain

(2004). Decoding speech prosody: Do music lessons help? Emotion, 4(1), 46–64.

33.

Tynes

Eiken

Grimsrud

T. K.

Sterud

Aasnæss

(2008). Arbeidsmiljø og helse—slik norske yrkesaktive opplever det. Oslo: STAMI–The National Institute of Occupational Health. Retrieved from http://www.stami.no/arbeidsmiljo-og-helse-hovedrapport

34.

Vaag

Bjørngaard

J. H.

Bjerkeset

(2016). Symptoms of anxiety and depression among Norwegian musicians compared to the general workforce. Psychology of Music, 44(2), 234–248. doi:10.1177/0305735614564910

35.

Vaag

Giæver

Bjerkeset

(2013). Specific demands and resources in the career of the Norwegian freelance musician. Arts & Health, 6(3), 205–222. doi:10.1080/17533015.2013.863789

36.

Vaag

Saksvik-Lehouillier

Bjørngaard

J. H.

Bjerkeset

(2015). Insomnia in Norwegian musicians compared to the general population. Behavioral Sleep Medicine. doi:10.1080/15402002.2015.1007991

37.

van de Laar

Verbeek

Pevernagie

Aldenkamp

Overeem

(2010). The role of personality traits in insomnia. Sleep Medicine Reviews, 14(1), 61–68. doi:10.1016/j.smrv.2009.07.007

38.

Voltmer

Zander

Fischer

J. E.

Kudielka

B. M.

Richter

Spahn

(2012). Physical and mental health of different types of orchestra musicians compared to other professions. Medical Problems of Performing Artists, 27(1), 9–14.

39.

Widiger

T. A.

Mullins-Sweatt

S. N.

(2010). Clinical utility of a dimensional model of personality disorder. Professional Psychology: Research and Practice, 41(6), 488–494. doi:10.1037/a0021694

40.

Widiger

T. A.

Presnall

J. R.

(2013). Clinical application of the five-factor model. Journal of Personality, 81(6), 515–527. doi:10.1111/jopy.12004