Individuals with fibromyalgia report greater pain sensitivity than healthy adults while listening to their favorite music: the contribution of negative affect

Abstract

Objective

We investigated the impact of favorite music on pain processing among individuals with fibromyalgia. We also examined differences in pain processing between individuals with fibromyalgia and healthy controls (HC) while listening to favorite music and explored whether psychosocial factors contributed to these differences.

Methods

Individuals with fibromyalgia and HC completed baseline psychosocial questionnaires and then underwent quantitative sensory testing (QST) during 3 randomized music conditions (meditative music, favorite music, white noise). Among individuals with fibromyalgia, Friedman tests were used to investigate differences in QST across conditions. Analyses of Covariance were used to examine group (HC vs fibromyalgia) differences in QST during favorite music. Correlations were conducted to explore associations of baseline psychosocial factors with QST during favorite music. Mediation analyses were conducted to explore whether psychosocial factors contributed to greater pain sensitivity among individuals with fibromyalgia compared to HC during favorite music.

Results

Individuals with fibromyalgia were less sensitive to pressure pain while listening to their favorite music compared to white noise. Compared to HC, individuals with fibromyalgia reported higher baseline negative affect and lower pain thresholds and tolerances during favorite music. Negative affect partially mediated the relationship between pain status (HC vs fibromyalgia) and pain sensitivity during favorite music.

Conclusions

Individuals with fibromyalgia were less pain sensitive while listening to favorite music than white noise, although they were more sensitive than HC. Greater negative affect endorsed by individuals with fibromyalgia contributed to their greater pain sensitivity. Future studies should explore the impact of favorite music on clinical pain.

Clinical trails registration

This study was registered with ClinicalTrials.gov (NCT04087564) and began on 6/13/2019.

Introduction

The idea of using music to help manage pain, as an inexpensive, flexible, and non-opioid alternative adjunctive therapy, is both compelling and promising, with several initial clinical studies showing reduced pain severity during and after listening to music.^1–4 Research investigating the ability of music to alter pain processing in an experimental setting has suggested that individuals report higher pressure pain thresholds and tolerances (eg, less pain sensitivity) while listening to music.^5–7 Despite its promise, the effects of music on pain are inconsistent across studies, potentially due to varying musical genres played (predominantly classical or meditative music) and interindividual variability in participants’ musical tastes compared to studies that allow individuals to self-select their own music to listen to. A recent study of healthy adults showed that participants had higher pain thresholds while listening to music that they liked compared to music that they disliked.⁸ Similarly, we demonstrated that healthy adults reported higher pain thresholds and tolerances while listening to their self-selected favorite music, which ranged widely in genre, compared to when they listened to meditative music or white noise.⁹ However, it remains unclear whether the beneficial effects of listening to one’s favorite music generalize to clinical populations with chronic pain, such as individuals with fibromyalgia.

Fibromyalgia is characterized by persistent widespread pain, mood disorders, fatigue, and cognitive difficulties.¹⁰ Individuals with fibromyalgia report greater pain sensitivity on several modalities of standardized pain testing (eg, quantitative sensory testing) compared to healthy, pain-free controls.^11–13 However, few investigations using participant-chosen favorite music to alter pain processing have been conducted, and whether the presence of favorite music might diminish differences in pain sensitivity that are often observed between individuals with fibromyalgia and healthy adults remains unclear.

Additionally, individuals with fibromyalgia tend to report higher levels of emotional distress, including negative affect, depression, and pain catastrophizing, compared to healthy adults.¹⁴ Psychological factors can play a key role in modulating pain, particularly among individuals with fibromyalgia, with the emotional distress and trait negative affect that is often experienced to a greater degree in these individuals being associated with higher levels of clinical pain.^15,16 Yet less work has investigated how participants’ baseline psychological characteristics relate to experimental pain sensitivity. One study among healthy adults showed that higher trait negative affect was associated with greater ischemic pain sensitivity, but unrelated to pressure and heat pain sensitivity.¹⁷ Understanding the relationship between participants’ baseline psychological characteristics and experimental pain sensitivity in the context of favorite music may provide further insight into factors that contribute to observed differences in pain processing between individuals with fibromyalgia and healthy controls.

The present study investigated the effects of favorite music on pain processing among individuals with fibromyalgia, with a previously studied group of healthy, pain-free controls as a reference group. We first explored the impact of favorite music, compared to meditative music and white noise, on pain processing among individuals with fibromyalgia. Based on our prior findings in healthy, pain-free adults, we hypothesized that individuals with fibromyalgia would have significantly lower pain sensitivity while listening to favorite music compared to both meditative music and white noise. We then aimed to gain deeper insight into differences in pain processing between individuals with fibromyalgia and healthy controls by comparing pain sensitivity between these groups while listening to their favorite music. We hypothesized that even in the presence of participants’ favorite music, individuals with fibromyalgia would show greater pain sensitivity compared to healthy controls. We next investigated how associations between baseline psychosocial factors and pain processing while listening to favorite music differed between individuals with fibromyalgia and healthy controls. Finally, exploratory analyses looked at whether baseline psychosocial factors contributed to differences in pain processing observed between individuals with fibromyalgia and healthy controls.

Methods

Individuals with fibromyalgia and healthy, pain-free adults were recruited for this in-person experimental quantitative sensory testing (QST) lab-based study. Participants were recruited through flyers, the online Partners Healthcare Rally website, and by contacting participants from previous studies via email and phone calls. Healthy adults were recruited between July 2019 and February 2020 and individuals with fibromyalgia were recruited between July 2019 and May 2023. Eligibility criteria for all participants included ≥18 years of age, fluency in English, and no diagnosis of hearing loss. Healthy adults were excluded if they had a diagnosis of chronic pain or neuropathy, or a history of chronic opioid use (active prescription >30 days). Individuals with fibromyalgia self-reported whether they had a diagnosis of fibromyalgia and completed the fibromyalgianess scale,¹⁰ which includes aspects of the fibromyalgia diagnosis (eg, widespread pain, symptom severity, duration of symptoms) and was designed to identify individuals with fibromyalgia. Trained research assistants then scored potential participants’ responses to determine whether they met the criteria for fibromyalgia created by Wolfe et al. (2011).

Eligible participants were scheduled for a 1.5–2 hour in-person study visit and were asked to provide a list of their 7 favorite songs prior to their visit. Upon arrival, participants provided informed consent, completed baseline pain and psychosocial questionnaires, and underwent several rounds of QST (Figure 1). At the end of the study, participants received a $40 Amazon gift card. The Partners Human Research Committee/Institutional Review Board approved all study procedures. This study was registered with ClinicalTrials.gov (NCT04087564).

Figure 1.

Flow chart of participant recruitment.

We have previously published data from this group of healthy adults, where we demonstrated that healthy adults reported significantly higher pressure pain thresholds and tolerances at the trapezius and forearm, as well as higher heat pain threshold, while listening to their self-selected favorite music compared to both meditative music and white noise.⁹ We did not observe a significant difference in temporal summation of pain (TSP) or conditioned pain modulation (CPM) across the three music conditions among healthy adults. In the present study, we expand upon our prior work by examining differences in pain sensitivity across these three music conditions among individuals with fibromyalgia to explore whether the beneficial effects of favorite music generalize to individuals with chronic pain. All data pertaining to the sample of individuals with fibromyalgia are novel and have not been published elsewhere, and as such, the present analyses do not overlap with our prior work.

Baseline questionnaires

Pain. The Brief Pain Inventory (BPI)¹⁸ was used to assess pain intensity and interference at baseline. Individual questions assessed participants’ current pain, as well as their average, lowest, and worst pain over the past week, on a scale of 0 (no pain) to 10 (worst pain imaginable). These 4 items were averaged for a total pain intensity score (range: 0–10). Ten questions assessed the extent to which participants’ pain interfered with their daily life (eg, sleep, walking ability) over the past week on a scale of 0 (no interference) to 10 (complete interference). Scores on each of these questions were summed for a total pain interference score (range: 0–100).

Psychosocial factors. The Patient Reported Outcome Measurement Information System (PROMIS)¹⁹ short forms were used to assess participants’ anxiety, depression, and sleep disturbance over the past week. Seven items measured anxiety (eg, “I felt worried”) and 8 items measured depression (eg, “I felt hopeless”) on a scale of 1 (never) to 5 (always). Items were summed and higher scores reflected greater anxiety (range: 7–35) and depression (range: 8–40), respectively. Eight items measured sleep disturbance (eg, “my sleep was restless”) on a scale of 1 (not at all) to 5 (very much). Items were summed, and higher scores indicated greater sleep disturbance (range: 8–40). The 14-item Pain Catastrophizing Scale (PCS)²⁰ was used to assess pain-related rumination, magnification, and helplessness (eg, “there is nothing I can do to reduce the intensity of the pain”). All items were rated on a scale from 0 (not at all) to 4 (all the time) and were summed for a total score (range: 0–56). The Positive and Negative Affect Schedule (PANAS)²¹ was used to measure participants’ general positive and negative mood. Ten items assessed participants’ average positive affect (eg, “excited,” “inspired”) and 10 items assessed negative affect (eg, “irritable,” “hostile”). All items were rated on a scale from 1 (very slightly or not at all) to 5 (extremely), and appropriate items were summed to create a total positive and negative affect score (ranges: 10–50).

Quantitative sensory testing (QST)

All participants completed four sessions of QST to measure pain sensitivity. The first session was used as a baseline condition to introduce participants to the testing tools and answer any questions in the absence of music. Participants then completed three additional identical QST sessions, each performed during a different music condition (meditative music, favorite music, white noise). The ordering of music conditions was randomized.

Pressure pain threshold and tolerance. Using a digital pressure algometer with a flat, round transducer (probe area, 0.785 cm), we measured participants’ pressure pain thresholds and tolerances. Forearm pressure pain threshold and tolerance were measured at the dorsal aspect of the proximal forearm on both the left and right sides. Trapezius pressure pain threshold and tolerance were measured over the trapezius muscle at the upper back on both the left and right sides. Pressure was applied at a rate of 1 lb/second, with a maximum of 20 lbs. Participants were instructed to indicate when the pressure first became painful (threshold) and then when they wanted the pressure to stop (tolerance). We created four pressure pain-related outcomes by averaging thresholds and tolerances across sides (left and right): (1) forearm threshold, (2) forearm tolerance, (3) trapezius threshold, and (4) trapezius tolerance.

Heat pain threshold. Heat pain threshold was assessed using a 1.6 × 1.6 cm square (2.56 cm²) contact thermode (Medoc Advanced Medical Systems, Ramat Yishai, Israel) on the volar side of participants’ left forearm. The baseline temperature started at 32°C and gradually increased at a rate of 2°C/second. Participants pressed a button on a remote to indicate when the stimulation first became painful (threshold).

Temporal summation of pain (TSP). Standardized weighted pinprick applicators (128 mN, 256 mN, and 512 mN) were used to assess temporal summation of pain (TSP) by applying the weighted metal filaments to the dorsal aspect of the index finger between the first and second interphalangeal joints. After each application, participants rated their pain (0 = no pain, 10 = worst pain imaginable). The heaviest weight that induced a pain score between 1 and 3/10 on a single application was used for repeated administration to measure TSP. The filament was applied 10 times consecutively to the same spot on the index finger at a rate of 1 pinprick/second. Participants again rated their pain after the tenth application. The difference between the pain score of the tenth application and the first single application was calculated as TSP (T10-T1).

Conditioned pain modulation (CPM). To test for conditioned pain modulation (CPM), we applied a painful stimulus alone and then again in the presence of another painful stimulus. CPM refers to the inhibition, or masking, of the perception of a painful stimulus by another painful stimulus. The CPM paradigm involved both mechanical (ie, pressure pain) and thermal (ie, cold pain) stimuli. Pressure pain threshold over the left trapezius served as the test stimulus (baseline threshold) in the present study (see “Pressure Pain” above). Cold pain induced by having participants submerge their right hand up to their wrist in a cold-water bath for 5 seconds served as the conditioning stimulus. After 5 seconds of hand immersion, participants’ pressure pain threshold over the trapezius was again assessed (ie, conditioned threshold). We calculated CPM using the following equation:

$$\left(\frac{\mathit{Baseline}\mathrm{ }\mathit{Threshold}-\mathit{Conditioned}\mathrm{ }\mathit{Threshold}}{\mathit{Baseline}\mathrm{ }\mathit{Threshold}}\right)\times \mathrm{ }100$$

Music conditions

Participants were seated in a chair in a quiet, private room while listening to the music and undergoing QST procedures. All music was played from a JBL Charge 3 speaker and participants were able to adjust the volume of the speaker so that they could both hear the music and respond to verbal instructions throughout QST. Only the trained research staff and participants were in the room during testing.

For the favorite music condition, participants were instructed to “Please list seven (7) of your favorite songs. Provide the song titles and artist’s name. These [songs] do not have to be anything specific other than songs you enjoy listening to.” Participants selected a wide variety of musical genres, such as Pop, Rock/Metal, Alternative/Indie, and R&B/Soul (see Supplemental Material for the full list of songs and genres). All songs were preloaded onto a playlist prior to the study visit, and during the favorite music condition, these songs were played in a randomized order using the shuffle feature on Apple Music. Although 7 songs were requested, not all songs may have been played depending on how long each QST session took to complete (ie, ∼20 minutes).

In the meditative music condition, participants listened to one of two tracks. One track included instrumental music designed to be relaxing from the Unwind smartphone-based application (Unwind; Bose Corporation, Boston, MA, USA). The Unwind app has been used in prior work among patients presenting to the emergency department.² However, during the course of the study, the Unwind app became intermittently unavailable for use. As such, a second track entitled “Beautiful Relaxing Music for Stress Relief—Meditation Music, Sleep Music, Ambient Study Music” by Peter Helland on YouTube was used (https://www.youtube.com/watch?v=2OEL4P1Rz04) when the Unwind app became unavailable. This instrumental track is described as meditative/relaxing music and was chosen for its similarity to the Unwind app.

In the white noise condition, participants listened to a track selected by the research staff titled “White Noise” from the Lullabies for Deep Meditation, Baby Sweet Dream & White Noise Research album on Apple Music (https://music.apple.com/us/album/white-noise-loopable-with-no-fade/1180798156?i=1180798172).

Data analysis

Participants’ baseline pain, psychosocial, and demographic characteristics are presented as means, standard deviations, and percentages. Among the subsample of individuals with fibromyalgia, Friedman tests were used to explore differences in pain sensitivity across the 3 music conditions (ie, meditative music, favorite music, white noise). A post hoc power analysis indicated that a sample size of 35 individuals with fibromyalgia was sufficient to detect a difference in pain sensitivity across the 3 music conditions based on an effect size of 0.5, significance level of .05, and power of 0.80.

Next, we aimed to gain deeper insight into pain sensitivity during the favorite music condition. First, we explored if demographic characteristics that were significantly different between healthy controls and individuals with fibromyalgia (ie, gender, age) were significantly associated with pain sensitivity during the favorite music condition. We then conducted analyses of covariance (ANCOVAs) to examine differences in pain sensitivity during the favorite music condition between healthy controls and individuals with fibromyalgia, while controlling for gender. Second, we conducted correlation analyses to explore associations of baseline psychosocial factors with pain sensitivity during the favorite music condition, separately for healthy controls and individuals with fibromyalgia. Finally, mediation analyses were conducted using the PROCESS macro for SPSS to explore whether baseline psychosocial factors mediated, or partially accounted for, group (healthy controls vs individuals with fibromyalgia) differences in pain sensitivity during the favorite music condition, while controlling for gender. Mediation analyses were conducted with pain status (individuals with fibromyalgia vs healthy controls) entered as the independent variable (x variable), psychosocial factors as the mediator (m variable), and pain sensitivity as the outcome variable (y variable). The mediation models used 5000 bootstrapped resamples. All data were analyzed using SPSS (v28).

Results

Baseline participant characteristics

The sample consisted of 109 participants in total (39 individuals with fibromyalgia; see Table 1). Participants had an average age of 39 years (SD = 18.1; range: 19–86) and 76% were female. Participants racially identified as White (65%), Asian (17%), Black/African American (7%), “other” (6%), more than 1 race (2%), American Indian/Alaskan Native (1%), 2% preferred not to answer, and 11% were Hispanic/Latino.

Table 1.

Baseline demographic, pain, and psychosocial characteristics of participants.

	Whole Sample	Individuals with Fibromyalgia	Healthy Controls	P-value
	(n = 109)	(n = 39)	(n = 70)
Variables	M±SD or N(%)	M±SD or N(%)	M±SD or N(%)
Demographics
Age (yrs)	39.0 ± 18.1	46.7 ± 15.3	35.0 ± 18.3	<.001
Gender				.002
Male	25 (23%)	2 (5%)	23 (33%)
Female	83 (76%)	36 (92%)	47 (67%)
Other	1 (1%)	1 (3%)
Race				.187
White	71 (65%)	30 (77%)	41 (59%)
Asian	18 (17%)	2 (5%)	16 (23%)
Black/African American	8 (7%)	4 (10%)	4 (6%)
American Indian/Alaskan Native	1 (1%)	–	1 (1%)
More than one race	2 (2%)	–	2 (3%)
‘Other’	7 (6%)	2 (5%)	5 (7%)
Prefer not to say	2 (2%)	1 (3%)	1 (1%)
Ethnicity				.322
Not Hispanic or Latino	95 (87%)	36 (92%)	59 (84.3%)
Hispanic or Latino	12 (11%)	2 (5%)	10 (14.3%)
Prefer not to say	2 (2%)	1 (3%)	1 (1.4%)
Baseline Pain
BPI Pain intensity (range: 0–10)	2.6 ± 2.7	5.8 ± 1.3	0.8 ± 1.3	<.001
BPI pain interference (range: 0–100)	19.7 ± 25.9	48.5 ± 18.6	3.7 ± 11.5	<.001
Baseline Psychosocial factors
Negative Affect (range: 10–50)	16.2 ± 6.2	20.2 ± 7.2	14.0 ± 4.2	<.001
Positive Affect (range: 10–50)	34.4 ± 7.3	30.5 ± 7.3	36.7 ± 6.3	<.001
Depression (range: 8–40)	11.9 ± 4.9	15.3 ± 5.4	9.9 ± 3.2	<.001
Anxiety (range: 7–35)	13.2 ± 6.2	17.9 ± 6.7	10.6 ± 3.9	<.001
Sleep disturbance (range: 8–40)	19.6 ± 8.2	27.2 ± 6.7	15.3 ± 5.4	<.001
Pain catastrophizing (range: 0–56)	10.8 ± 10.8	17.7 ± 11.4	7.0 ± 8.3	<.001

Mann-Whitney U tests were used to examine group differences in continuous variables and χ² tests for categorical variables. BPI = brief pain inventory.

Compared to healthy controls, individuals with fibromyalgia were significantly older and a relatively higher proportion were female (Table 1). There was not a significant difference between the groups with respect to race/ethnicity. As expected, individuals with fibromyalgia reported higher BPI pain intensity and pain interference at baseline compared to healthy controls. Similarly, at baseline, individuals with fibromyalgia reported higher levels of trait negative affect, depression, anxiety, sleep disturbance, and pain catastrophizing, as well as less trait positive affect compared to healthy controls (Table 1).

Music and pain sensitivity among individuals with fibromyalgia

We found that among individuals with fibromyalgia, forearm pressure pain threshold and tolerance were significantly different between the three conditions, such that individuals had a significantly lower forearm pressure pain threshold and tolerance (ie, more pain sensitive) during the white noise condition compared to both the favorite music and meditative music conditions (P < 0.05; Table 2). There was also a significant difference in trapezius pressure pain tolerance across the three conditions, such that individuals had a significantly lower trapezius pressure pain tolerance during the white noise condition compared to the favorite music condition (P < .05). Notably, pressure pain was not significantly different between the favorite music and meditative music conditions among individuals with fibromyalgia (P > .05), unlike differences we had previously observed in healthy controls.⁹ We also did not observe a significant difference in trapezius pressure pain threshold, heat pain threshold, TSP, or CPM across the three conditions (P > 0.05; Table 2).

Table 2.

Pain sensitivity across music conditions among individuals with fibromyalgia.

	Meditative Music	Favorite Music	White Noise
QST Variables	M±SD	M±SD	M±SD	χ2(2)	P-value	Effect size
Forearm pressure pain threshold†◊	3.61 ± 1.81	3.71 ± 2.11	3.44 ± 2.15	8.28	.016	0.12
Forearm pressure pain tolerance†◊	6.11 ± 3.63	6.73 ± 4.49	5.55 ± 3.45	14.11	<.001	0.19
Trapezius pressure pain threshold	5.46 ± 3.55	6.06 ± 4.17	5.18 ± 3.29	3.71	.156	0.05
Trapezius pressure pain tolerance†	8.42 ± 4.87	9.04 ± 5.54	8.03 ± 5.06	8.10	.017	0.11
Heat pain threshold	39.00 ± 2.55	39.77 ± 2.71	39.58 ± 2.94	3.08	.214	0.04
Temporal summation of pain (TSP)	2.52 ± 1.95	2.39 ± 1.73	2.82 ± 1.72	4.69	.096	0.06
Conditioned pain modulation (CPM)	−12.46 ± 41.64	−30.10 ± 65.19	−12.20 ± 33.54	1.56	.459	0.02

Friedman tests were used to explore differences in pain sensitivity across the three music conditions. Pairwise comparisons were used to determine which conditions where significantly different from each other (P < .05):

^†=significant difference between the favorite music and white noise conditions;

^◊=significant difference between the meditative music and white noise conditions. Higher scores for threshold and tolerance reflect lower pain sensitivity. Lower scores on CPM reflect greater modulation of pain (lower pain sensitivity) and lower scores on TSP reflect lower pain sensitivity. CPM = conditioned pain modulation; QST = quantitative sensory tests; TSP = temporal summation of pain.

Differences in pain sensitivity during favorite music based on pain status

We were interested in exploring differences in pain processing between healthy controls and individuals with fibromyalgia in different contexts, particularly while listening to one’s favorite music. First, we explored whether participants’ age or gender were related to pain sensitivity while listening to favorite music. We found that men reported significantly higher pressure pain thresholds and tolerances than women (P < .05), although there were no gender differences in heat pain, TSP, or CPM (P > .05). We did not observe any significant relationship between age and pain sensitivity on any QST measure while listening to favorite music (P > .05). As such, ANCOVAs were conducted to examine differences in pain sensitivity during the favorite music condition between healthy controls and individuals with fibromyalgia, while controlling for only gender. We also conducted sensitivity analyses controlling for both gender and age, which showed that the main pattern of findings remained the same (see Supplemental Material). Our results revealed that in the presence of their self-selected favorite music, individuals with fibromyalgia reported significantly lower forearm and trapezius pressure pain thresholds and tolerances, as well as lower heat pain threshold, compared to healthy controls (P < .05; Table 3; Figure 2). However, we did not observe a significant difference in TSP or CPM between healthy controls and individuals with fibromyalgia (P > .05).

Table 3.

Group differences in pain sensitivity during the favorite music condition.

	Individuals with Fibromyalgia	Healthy Controls	F(df, n)	P-value	ηp2
QST Variables	M±SE	M±SE
Forearm pressure pain threshold	4.22 ± 0.57	7.62 ± 0.40	22.63 (1,102)	<.001	0.18
Forearm pressure pain tolerance	7.64 ± 0.72	10.63 ± 0.51	11.07 (1,104)	.001	0.10
Trapezius pressure pain threshold	6.90 ± 0.73	10.69 ± 0.51	17.39 (1,103)	<.001	0.14
Trapezius pressure pain tolerance	10.07 ± 0.75	13.95 ± 0.54	16.88 (1,104)	<.001	0.14
Heat pain threshold	39.89 ± 0.53	41.74 ± 0.38	7.70 (1,103)	.007	0.07
Temporal summation of pain (TSP)	2.29 ± 0.29	2.39 ± 0.21	0.07 (1,104)	.798	0.00
Conditioned pain modulation (CPM)	−30.66 ± 7.40	−13.52 ± 5.22	3.43 (1,103)	.067	0.03

ANCOVAs, with gender as a covariate, were used to explore group differences in pain sensitivity during the favorite music condition. Higher scores for threshold and tolerance reflect lower pain sensitivity. Lower scores on CPM reflect greater modulation of pain (less sensitivity) and lower scores on TSP reflect lower pain sensitivity. QST = quantitative sensory tests.

Figure 2.

Comparison of pain sensitivity on measures of (A) pressure pain and (B) heat pain between healthy controls and individuals with fibromyalgia while listening to their self-selected favorite music.

Next, we explored correlations between baseline psychosocial factors and pain sensitivity during the favorite music condition, assessing these separately for individuals with fibromyalgia and healthy controls (Table 4). Among individuals with fibromyalgia, baseline negative affect was consistently correlated with lower forearm and trapezius pressure pain thresholds and tolerances, as well as lower heat pain threshold (P < .05). No other psychosocial factor was consistently associated with any measure of pain sensitivity during the favorite music condition (P > 0.05). Among healthy controls listening to their self-selected favorite music, no baseline psychosocial factor (including negative affect) was consistently associated with any measure of pain sensitivity (Table 4).

Table 4.

Associations between baseline psychosocial factors and pain sensitivity during the favorite music condition.

	Forearm Pain Threshold	Forearm Pain Tolerance	Trapezius Pain Threshold	Trapezius Pain Tolerance	Heat Pain Threshold	Conditioned Pain Modulation	Temporal Summation of Pain
Individuals with Fibromyalgia
Positive affect	0.16	0.19	0.14	0.18	0.14	0.31	−0.21
Negative affect	−0.44 **	−0.39 *	−0.42 **	−0.33 *	−0.45 **	−0.09	0.21
Depression	−0.28	−0.31	−0.32	−0.24	−0.27	−0.11	0.13
Anxiety	−0.20	−0.18	−0.23	−0.12	−0.20	−0.11	0.20
Sleep disturbance	−0.03	0.17	0.05	0.14	−0.05	0.40 *	−0.06
Pain catastrophizing	−0.19	−0.21	−0.26	−0.22	−0.30	0.03	0.02
Healthy controls
Positive affect	0.21	0.24 *	0.21	0.19	−0.01	0.19	−0.01
Negative affect	−0.14	−0.16	−0.17	−0.08	−0.07	−0.19	0.01
Depression	−0.12	−0.10	−0.15	−0.08	−0.10	−0.10	0.14
Anxiety	−0.17	−0.19	−0.26 *	−0.18	−0.02	−0.18	0.10
Sleep disturbance	−0.16	−0.23	−0.15	−0.10	−0.19	−0.06	0.29 *
Pain catastrophizing	−0.11	−0.07	−0.11	−0.10	−0.07	−0.21	0.09

Higher scores for threshold and tolerance reflect lower pain sensitivity. Lower scores on CPM reflect greater modulation of pain (less sensitivity) and lower scores on TSP reflect lower pain sensitivity.

^* P < .05,

^** P < .01.

Since higher negative affect was uniquely and consistently associated with greater pain sensitivity during the favorite music condition among individuals with fibromyalgia, we investigated whether negative affect partially accounted for, or mediated, the group differences in pain sensitivity, while controlling for gender (Figure 3). The results of the mediation analyses revealed that there was a significant indirect effect of pain status on pain sensitivity through negative affect for all five threshold and tolerance measures (ie, 1: Forearm pressure pain threshold; 2: Forearm pressure pain tolerance; 3: Trapezius pressure pain threshold; 4: Trapezius pressure pain tolerance; and 5: Heat pain threshold) (Table 5; see Supplemental Material for full statistical details). These findings suggest that higher levels of negative affect endorsed by individuals with fibromyalgia, compared to healthy controls, contributed to greater pain sensitivity on measures of pressure and thermal pain.

Figure 3.

Mediation analysis with baseline negative affect as a mediator of the relationship between pain status (healthy controls vs individuals with fibromyalgia) and pain sensitivity during the favorite music condition.

Table 5.

Negative affect as a mediator of the association between pain status and pain sensitivity.

	Total Effect (Path c)	Direct Effect (Path c′)	Indirect Effect
Dependent Variable	beta	beta	beta [95% CI: LL, UL]
Forearm pressure pain threshold	−3.40**	−2.71**	−0.69 [−1.32, −0.15]
Forearm pressure pain tolerance	−2.99*	−1.90	−1.09 [−1.97, −0.37]
Trapezius pressure pain threshold	−3.80**	−2.66*	−1.13 [−2.24, −0.28]
Trapezius pressure pain tolerance	−3.87**	−2.97*	−0.91 [−2.05, −0.05]
Heat pain threshold	−1.84*	−1.12	−0.73 [−1.49, −0.10]

Pain status (healthy controls vs individuals with fibromyalgia) was entered as the independent variable (x variable), baseline negative affect as a mediator (m variable), and pain sensitivity during favorite music as the outcome variable (y variable). Gender was included as a covariate in all models. Path c represents the effect of pain status (x) on pain sensitivity (y); path c′ represents the effect of pain status (x) on pain sensitivity (y) while controlling for the role of baseline negative affect (m); the indirect effect represents whether there was a significant mediational effect.

^* P < .01,

^** P < .001.

CI = confidence interval; LL = lower limit; UL = upper limit.

Discussion

We investigated the impact of different types of music on pain processing among individuals with fibromyalgia and found that these participants were less sensitive to pain, reporting a significantly higher pressure pain threshold and tolerance, while listening to both their self-selected favorite music and meditative music compared to white noise. At baseline, individuals with fibromyalgia reported significantly higher levels of negative affect, depression, anxiety, pain catastrophizing, and sleep disturbance compared to healthy controls. We then explored pain processing while listening to favorite music and found that, even in the presence of their favorite music, individuals with fibromyalgia reported significantly lower pain thresholds and tolerances than healthy controls. Interestingly, baseline negative affect was associated with lower pain thresholds and tolerances while listening to favorite music among individuals with fibromyalgia, but not among healthy controls. Mediation analyses then showed that higher levels of baseline negative affect among individuals with fibromyalgia partially contributed to their lower pain thresholds and tolerances. These findings provide insight into a potential modifiable psychosocial factor that contributed to observed differences in pain processing between individuals with fibromyalgia and healthy controls.

In the present study, we demonstrated that individuals with fibromyalgia had a higher pressure pain threshold and tolerance (ie, less pain sensitive) while listening to their self-selected favorite music compared to white noise. However, no difference was observed between favorite music and meditative music, and as such, these findings only partially supported our hypothesis. Our findings suggest that music has a positive impact on pain processing among individuals with a chronic pain state. In our prior work, and the work of others, healthy adults reported higher pain thresholds and tolerances while listening to their self-selected favorite music compared to when they listened to either meditative music, white noise,⁹ or music they disliked,⁸ suggesting that favorite music may have a larger analgesic benefit. The type of music someone is listening to, and how much they like or dislike it, may importantly influence the degree to which that music modulates their processing of pain. Alternatively, favorite music may serve as a form of distraction from a painful stimulus, and indeed, research has shown that distraction alters pain processing.²² For example, while listening to one’s favorite music, an individual may focus on and sing along to the lyrics in their head, which could distract them from painful stimuli. Interestingly, one study found that knowing more lyrics of a song while undergoing experimental painful stimulation was associated with reporting higher pain tolerance.²³ While it has been posited that distraction may be a mechanism by which music reduces pain, more research is needed to test this, particularly within the context of favorite music and among individuals with chronic pain.

As with many analgesic therapies, there may also be some degree of variability in efficacy. We observed interindividual variability in the degree of the impact of music, but also variability between tests of different aspects of nociceptive processing. Specifically, as in our prior work among healthy adults,⁹ we saw a difference in pressure pain threshold and tolerance, but did not observe a difference in TSP or CPM during favorite music or meditative music compared to white noise among individuals with fibromyalgia. Some considerations regarding why this might be the case include that TSP and CPM are both more complex tests, which measure aspects of central modulation of nociceptive processing. It is possible that the degree of additional modulation by music is small enough that it does not rise above the background of variability in these tests’ modulation of pain.

Individuals with fibromyalgia reported greater pain sensitivity even while listening to their self-selected favorite music compared to healthy controls. Similar to prior studies,^24,25 some healthy controls had a pressure threshold or tolerance of 20 lbs, particularly at the trapezius, which was the maximum pressure applied. Additionally, individuals with fibromyalgia reported higher levels of baseline emotional distress compared to healthy controls, including negative affect, depression, and anxiety. Interestingly, baseline negative affect, but not other baseline psychosocial factors, was significantly related to greater pain sensitivity while listening to favorite music among individuals with fibromyalgia, but not among healthy controls. The measure of negative affect used in the present study (ie, PANAS) captures a broad range of negative emotional experiences, which may be particularly relevant to individuals with fibromyalgia. For example, the PANAS includes some anger-related items (eg, hostile, irritable), and researchers have suggested that anger is uniquely important as a pain-amplifying factor among individuals with fibromyalgia.²⁶ Studies have also suggested stronger pain augmentation during the experience of negative affect among individuals with fibromyalgia, compared to healthy controls and individuals with other chronic pain conditions,^27,28 indicating that individuals with fibromyalgia who are prone to high levels of negative affect may be particularly vulnerable to increased pain. In line with prior work, our mediation analyses showed that negative affect partially contributed to group differences in pain sensitivity, such that individuals with fibromyalgia reported greater negative affect at baseline, and in turn, they had lower pain thresholds and tolerances than healthy controls while listening to their favorite music. These findings suggest that this greater breadth of negative emotional experiences may contribute to greater pain sensitivity in individuals with fibromyalgia, even in the presence of their favorite music, which may positively modulate mood or serve as a distraction from negative cognitions.

Our findings highlight the importance of recognizing higher levels of negative affect among individuals with fibromyalgia and how it may contribute to their pain processing. Behavioral interventions, such as cognitive behavioral therapy (CBT), have shown to reduce negative affect among individuals with fibromyalgia and also may be particularly helpful in modulating pain processing.^29,30 Studies also suggest that CBT may reduce clinical pain, as well as comorbid symptoms of depression and anxiety.^29,31 While CBT is a promising treatment option, it may not be accessible to everyone, due to barriers related to an individual’s resources, time, and economic factors. In contrast, music is highly accessible and typically inexpensive. One study in college students found that listening to participant-chosen music reduced negative affect compared to a silent control, while also increasing state mindfulness.³² Interestingly, this study also showed that self-selected music decreased negative affect to a variable degree, being more effective among those who scored higher in state mindfulness, consistent with the idea that the effects of music on negative affect may vary based on individual characteristics. Thus, while initial evidence suggests that music may be a beneficial treatment option for targeting both pain and negative affect, this modulatory effect on negative affect may vary among individuals, and its potential use among different individuals with fibromyalgia requires more study.

Limitations and future directions

There are important study limitations to consider when interpreting the present findings. The relatively modest sample size of individuals with fibromyalgia, a product of the COVID-19 pandemic on recruitment, may have contributed to a higher likelihood of Type II error when comparing the impact of favorite music on pain processing relative to meditative music. Future studies would benefit from recruiting a larger and more demographically diverse sample to better understand the impact of favorite music compared to other music or sounds conditions. While we investigated the impact of self-selected favorite music, we did not direct participants to choose songs that made them feel a certain way (eg, relaxing music, energizing music, happy music, etc), thus limiting any conclusions about the relative efficacy of different types of music per se. Even within the meditative music condition, there was some variation in the exact sound track that participants heard. When the Unwind app became intermittently unavailable, we substituted it with similar sounding meditative music available through another commonly available source (YouTube). While we did not observe a significant difference in pain sensitivity on QST measures between participants who received the app-delivered meditative music versus the meditative track from YouTube (data not shown), we were likely underpowered to make this comparison, and thus, we cannot conclude what type of meditative music may be most effective.

Another limitation is that we did not collect data regarding participants’ medication use, which might influence pain sensitivity. Ultimately, the comparison was between conditions within the same participant, theoretically making the potential influence of medications less important, as it would have been present in all conditions. We also only investigated the impact of music on experimental pain, and therefore, the question of whether the positive impact of self-selected favorite music applies to clinical pain is not answered by these findings. In addition, the extent to which music may influence pain interference in daily functioning, in contrast to its relatively modest modulation of pain sensitivity on QST, has yet to be studied. We also did not assess negative affect in the moment (ie, state affect), and as such, changes in current negative affect across music conditions and the impact of current negative affect on pain sensitivity could not be specifically evaluated. Similarly, our data are correlational, and thus, directionality and causality between negative affect and pain sensitivity cannot be inferred. How favorite music modulates negative affect among individuals with chronic pain as they experience pain in the course of the day will require further study. Future studies should aim to employ longitudinal and ecological data collection, whereby individuals with chronic pain listen to music daily and report the impact of music on clinical pain severity, pain-related daily functioning, and current mood.

Conclusion

Individuals with fibromyalgia were less pain sensitive (higher pressure pain threshold and tolerance) while listening to their favorite music and meditative music compared to white noise. Additionally, even in the presence of their favorite music, individuals with fibromyalgia reported significantly higher pressure and thermal pain sensitivity than healthy controls. Interestingly, higher levels of negative affect at baseline endorsed by individuals with fibromyalgia contributed to their greater pain sensitivity while listening to their favorite music. Future studies are needed to explore whether the beneficial impact of self-selected favorite music extends to clinical pain severity, pain-related daily functioning, and current mood.

Supplementary material

Supplementary material is available at Pain Medicine online.

Funding

This work was supported by the National Institutes of Health (NIH R35 GM128691-01) to K.L.S.

Conflicts of interest: None declared.