Biopsychosocial contributors to irritability in individuals with shoulder or low back pain

ABSTRACT

Objectives

Irritability is a foundational clinical reasoning concept in rehabilitation to evaluate reactivity of the examination and treatment. While originally theorized to reflect tissue damage, a large body of evidence supports pain is a biopsychosocial experience impacted by pain sensitivity and psychological factors. Therefore, the purpose of this study was to examine biopsychosocial contributors to irritability.

Methods

40 patients with shoulder (n = 20) and low back (n = 20) pain underwent Quantitative Sensory Testing (QST) (Pressure Pain Threshold, Heat Pain Threshold, Conditioned Pain Modulation, Temporal Summation), completed pain-related psychological questionnaires, an Exercise-Induced Hypoalgesia protocol, and standardized irritability assessment based on Clinical Practice Guidelines. Participants were then categorized as irritable or not irritable based on Maitland’s criteria and by irritability level based on Clinical Practice Guidelines. An independent samples t-test examined for differences in QST and psychological factors by irritability category. A MANOVA examined for differences in QST and psychological factors by irritability level (high, moderate, low).

Results

Significantly lower heat and pressure pain thresholds at multiple locations (p < 0.05), as well as less efficient conditioned pain modulation (p = 0.02), were demonstrated in individuals categorized as irritable. Heat and pressure pain thresholds were also significantly lower in patients with high irritability compared to other levels. Significantly higher depression and anger, as well as lower self-efficacy, were reported in individuals with an irritable presentation.

Discussion/Conclusion

Biopsychosocial factors, including widespread hyperalgesia and elevated psychological factors, may contribute to an irritable presentation.

Introduction

Tissue irritability is a clinical reasoning concept in rehabilitation introduced by Geoffrey Maitland to determine the appropriate vigor for examination and treatment. Maitland proposed that individuals may be classified as ‘irritable’ or ‘not irritable’ based on the vigor of activity to produce symptoms, the severity of symptoms, and the time for symptoms to decrease [1]. Irritability is a frequently cited clinical reasoning concept in rehabilitation case reports, viewpoint articles [2–5], and recommended in Clinical Practice Guidelines to determine treatment intensity [6,7]. This deeply entrenched concept is used by rehabilitation providers in the clinical reasoning process for determining intensity of treatment [4,8].

Maitland advocated for reducing the intensity of treatment for patients considered to be irritable as a method to mitigate pain while allowing for a tolerable range of motion exercise [1]. Therefore, it is common knowledge within the physical therapy community that irritability is traditionally conceptualized as representing a peripheral mechanism, reflecting the injured tissue’s ability to accommodate physical stress. However, this characterization is not supported by the large body of evidence demonstrating a poor relationship between pain intensity and the extent of tissue damage [9–11]. This view also does not account for the current understanding of pain as a complex biopsychosocial experience affected by the interaction between the peripheral tissue and nervous system processing. As a biopsychosocial experience, pain is influenced by variability in pain sensitivity [12,13] and psychological factors [14–17]. Quantitative Sensory Testing (QST) is the systematic application of sensory stimuli and may be applied to make behavioral inferences on peripheral and central nervous system mechanisms underlying maladaptive changes in pain processing [18–21]. Pain-related psychological factors impact the transition from acute to chronic pain [22,23].

The traditional view of irritability encourages physical therapists to include interventions for shoulder pain that minimize physical stress to the tissue for individuals with high irritability, including: modalities, manual therapy, and mobility exercises [7]. As a result, physical therapists are more likely to select interventions that do not adhere to evidence-based practice for patients with high irritability [8]. This view may lead rehabilitation providers to under dose intervention intensity in an effort to minimize physical stress to the tissue leading to non-optimal management and reinforcement of avoidance behaviors.

The purpose of this research is to examine biopsychosocial contributors to inter-individual variability in irritability in patients presenting with shoulder or low back pain by determining the magnitude of difference in pain sensitivity and pain-related psychological factors between individuals categorized as irritable or not irritable and high, moderate, or low irritability.

Methods

An observational study in patients with low back or shoulder pain was conducted between June 2022-May 2023 to explore if inter-individual differences in irritability differ by psychological factors, pain sensitivity, and exercise-induced hypoalgesia. Participants were recruited from both the community and a university physical therapy clinic. This study was approved by the University of Central Florida Institutional Review Board (Study #4353) for Human Subjects Research. All participants provided written informed consent to enroll in the study.

Participants

Participants between 18 and 75 years who were currently experiencing either shoulder or low back pain with symptom intensity rated as 3/10 or higher within the past 24 hours attended one testing session. Participants were excluded from the study if they had: 1) systemic medical conditions that affected sensation, 2) low back or shoulder surgery or fracture within the past 6 months, 3) a chronic pain condition, such as fibromyalgia, 4) current medical diagnosis of myelopathy, 5) blood clotting disorders, 6) contraindication to the application of ice, 7) blood pressure greater than 140/90 mmHg.

Demographic and clinical questionnaires:

Participants reported demographic and clinical symptoms using a standard questionnaire. Disability was measured in participants with shoulder pain with the QuickDASH as it demonstrates good test–retest reliability, validity, and responsiveness in this population [24,25]. Disability was calculated in participants with low back pain with the Oswestry Disability Index (ODI) as it demonstrates good construct validity and reliability [26,27]. Pain intensity was measured with a 101-point numerical pain rating scale (NPRS) where 0=no pain and 100=worst pain imaginable [28,29]. Average pain intensity was calculated as the mean of the best, worst, and current pain intensity in the past 24 h.

Exercise-induced hypoalgesia (EIH):

Pressure pain threshold (PPT) was examined at the mid belly of the quadriceps muscle before and after an isometric quadriceps contraction [30]. PPT is a common outcome measure in EIH research [31–33] and prior studies in patients with shoulder pain have used the quadriceps as the exercising muscle [30]. A handheld dynamometer (MicroFET, Hoggan Scientific) was applied to the distal tibia while the participant performed a maximal voluntary contraction of knee extension. This was repeated three times with a 1-min rest between each set. Participants held an isometric quadriceps contraction at 30% of the average maximum voluntary isometric contraction for three minutes [34–36]. EIH was calculated with the following formula: Pre Exercise PPT – Post Exercise PPT = EIH Magnitude. A negative value for EIH indicated hypoalgesia (lessening of pain sensitivity). Participants waited quietly for 30 min after the exercise to allow pain sensitivity to normalize. EIH effects have been reported to last between 15 and 30 min and, therefore, 30 min was sufficient to allow for inhibitory effects to washout [37,38].

Pain-related psychological questionnaires

Fear of Pain Questionnaire (FPQ-9)

Individuals reported fear of pain during nine items from ‘not at all’ to ‘extreme.’ Higher scores indicated a greater fear of pain [39].

Pain Catastrophizing Scale (PCS)

Individuals reported the frequency of catastrophizing statements during a painful episode. Higher scores indicated higher pain catastrophizing levels [40–42].

Fear Avoidance Beliefs Questionnaire (FABQ)

The FABQ includes a work and physical activity subscale with higher scores indicating greater fear avoidance behaviors [23,43,44].

Tampa Scale of Kinesiophobia-11 (TSK-11)

The TSK-11 measures fear of movement, and higher TSK-11 scores indicate greater fear of movement and injury/re-injury due to pain [45].

Center of Epidemiologic Studies – Depression (CES-D)

Participants rated how often they experienced symptoms associated with depression over the past week. Higher scores indicated greater symptoms [46].

State-Trait Anger Expression Inventory

This questionnaire measures anger and how it contributes to medical conditions, such as pain [47,48].

Pain Anxiety Symptoms Scale-20 (PASS-20)

The PASS measures pain-related anxiety with higher scores indicating greater pain-related anxiety [49].

Pain Self-Efficacy Questionnaire (PSEQ)

The PSEQ measures pain self-efficacy with higher scores indicating higher self-efficacy [23].

Multi-modal pain sensitivity measures

All pain sensitivity testing was conducted by the same researcher, a physical therapist with over 7 years of experience administering these measures.

Heat pain threshold (HPT)

Participants were provided a 2 × 1 inch thermode attached to a TCS-II (QST.Lab, Strasbourg, France). The thermode increased from a baseline of 32° Celsius at 1°C/second to a maximum of 50° Celsius. Participants were instructed to indicate when the sensation first changed from warmth to pain (pain threshold) by pressing a button. Participants then rated the pain using the NPRS. This procedure was completed two times medial to the PSIS (two inches lateral to L5 spinous process), muscle belly of the deltoid (approximately two inches distal to the acromion), and muscle belly of the tibialis anterior muscle (two inches distal and lateral to the tibial tuberosity) on the same side as the participant’s shoulder or low back pain and the average analyzed. Testing on the same side as the participant’s pain is consistent with prior literature as a method to standardize the location [50,51]. A lower HPT indicates a higher pain sensitivity.

Temporal summation (TS)

Participants were provided a thermode to place on the palmar surface of the hand. A thermode attached to a TCS-II (QST.Lab, Strasbourg, France) delivered a train of 10 heat pulses to the skin with temperatures peaking at 49°C at a rate to maintain the desired inter-stimulus interval [52]. Application of noxious stimulation at greater than 3 Hz/second by thermal stimuli may repeatedly activate C-fibers producing a NMDA receptor-dependent increase of second-order neuron output at the spinal cord [53,54]. Participants rated the magnitude of their ‘second pain’ sensation following each heat pulse using a 101-point NPRS. These response ratings are believed to be primarily C-fiber mediated and are indicative of a pain facilitatory process [12]. Temporal Summation was calculated as the difference in pain ratings between the 5^th pulse and the 1^st pulse [55].

Aftersensations

Aftersensations are also primarily C-fiber mediated and associated with clinical pain intensity, as well as widespread pain [56]. After the tenth pulse, participants were asked if any pain sensation persisted in the hand. If yes, participants were asked to rate the pain in their hand for up to 60 s.

Pressure pain threshold (PPT)

A digital computerized pressure algometer (AlgoMed, Ramat Yishai, Israel) with a 1 cm diameter rubber tip was applied at a constant rate. Participants were instructed to press a button when the sensation first changed from pressure to pain (pain threshold). Participants rated the pain at threshold using the NPRS. This procedure was completed two times medial to the PSIS, muscle belly of the deltoid, and muscle belly of the anterior tibialis muscle on the same side as the participant’s shoulder or low back pain and the average analyzed. A lower PPT indicates a higher pain sensitivity.

Conditioned pain modulation (CPM)

CPM is a behavioral measure of a pain inhibitory process that is believed to be lower brainstem mediated [57]. The CPM testing protocol is based on the recommendations for testing [58]. PPT was applied to the web space of the foot opposite the patient’s shoulder or low back pain. Participants immersed the hand on the same side of the patient’s pain into water cooled by a refrigeration unit (ARCTIC Series Refrigerated Bath Circulator, ThermoFisher Scientific, Massachusetts, USA) that circulated water continuously to maintain a constant temperature of 8° Celsius (males) or 10° Celsius (females) for 60 s. PPT was re-measured on the web space of the foot. CPM Efficiency was calculated using the following formula: Pre Testing Stimulus – Post Testing Stimulus. Negative numbers indicated an efficient pain modulation, or a pain inhibitory effect.⁷⁷

A summary of the outcomes assessed in this study can be found in Table 1.

Table 1.

Summary of biopsychosocial outcomes assessed in the study.

Exercise-Induced Hypoalgesia	Psychological Factors	Quantitative Sensory Testing
Step 1) PPT applied to quadriceps Step 2) Isometric contraction of the quadriceps for 3 minutes at 30% MVC Step 3) PPT re-applied to the quadriceps	Fear of Pain Questionnaire-9 Pain Catastrophizing Scale Fear Avoidance Beliefs Questionnaire Tampa Scale of Kinesiophobia-11 Center of Epidemiologic Studies-Depression State-Trait Anger Expression Inventory Pain Anxiety Symptoms Scale-20 Pain Self-Efficacy Questionnaire	HPT PPT TS CPM

Note: PPT=Pressure Pain Threshold, HPT = Heat Pain Threshold, TS=Temporal Summation, CPM = Conditioned Pain Modulation.

Irritability assessment

Irritability assessment was performed in a standardized manner by one licensed physical therapist with approximately 10 years of combined clinical and research experience. The irritability assessment protocol was based on published criteria in shoulder pain Clinical Practice Guidelines [7,59,60]. The shoulder pain and mobility deficits clinical practice guidelines [7] and Staged Approach for Rehabilitation of Shoulder Disorders clinical reasoning model [59] presents irritability levels (high, moderate, or low) with thresholds based on pain intensity, frequency of night or rest pain, active and passive range of motion, and disability level.

Participants were asked to rate their current, best, and worst shoulder or low back pain intensity within the past 24 h using the 11-point NPRS (0=no pain, 10=worst pain imaginable). Disability level was determined based on scores on the ODI or quickDASH. Next, participants underwent range of motion testing. Individuals with shoulder pain were instructed to complete active shoulder flexion, abduction, external rotation with the arm abducted to 90 degrees, and internal rotation with the arm abducted to 90 degrees one time. Using a standard script, participants were asked to notify the assessor when they experienced pain during the range of motion but continue as far as they were willing. Participants were then moved to a supine position where the assessor completed passive shoulder flexion, abduction, external and internal rotation with the arm abducted to 90 degrees one time. Participants were asked to notify the assessor when they experienced pain during range of motion but allow passive motion as far as they were willing. Pain ratings during active and passive range of motion were recorded with the 11-point NPRS. The same assessor then determined if active range of motion was less than, similar, or more than passive range of motion.

Participants with low back pain completed active lumbar flexion, extension, and lateral flexion in standing and asked to notify the assessor when they experienced pain during range of motion but continue as far as they were willing. Passive range of motion was not examined in participants with low back pain. As a result, the assessor determined if pain occurred before end of active range of motion, at the end of active range of motion, or with passive over pressure. These assessment techniques fulfilled the criteria listed in the Clinical Practice Guidelines and based on this assessment, participant’s irritability was classified as high, moderate, or low irritability [7]. This approach has been previously studied in individuals with shoulder pain with established interrater reliability = 0.69(95% CI = 0.59–0.78) [7,60].

Participants were categorized as high, moderate, or low irritability based on the published Clinical Practice Guidelines [7]. Briefly, high irritability was characterized by: pain intensity greater than 7/10 with consistent night pain, high disability, pain occurring before end range, and active range of motion being significantly less than passive range of motion due to pain. Moderate irritability was characterized by: pain intensity = 4–6/10, intermittent night pain, moderate levels of reported disability, pain at end range, and active and passive range of motion being similar. Low irritability was characterized by: pain equal to or less than 3/10, no night pain, minimal levels of disability, pain with overpressure, and the same active and passive range of motion [7].

Participants were also categorized based on Maitland’s categories (irritable or not irritable). Irritability judgments by physical therapists treating patients with low back pain based on Maitland’s classification (irritable or not) demonstrated moderate inter-rater reliability (k = 0.44) [61]. Prior studies have categorized patients as irritable or not irritable based clinical experience and the following criteria: 1) vigor of activity to cause symptoms, 2) severity of symptoms once provoked, and 3) time for symptoms to reduce [62]. We applied these criteria and, for greater standardization, asked all participants to report a task that provoked their shoulder or low back pain. Participants were asked to provide the amount of time they could tolerate the task before the onset of pain (time to onset), time tolerated in task before stopping due to pain (time to stop), and time the pain persisted after completion of the task (pain persistence). This allowed for the assessor to classify participants as irritable or not irritable based on Maitland’s criteria of vigor of activity to provoke symptoms, symptom intensity, and persistence of symptoms [1]. Participants who reported a short time to onset with over 1 h of pain persistence were classified as irritable. As an exploratory measure of irritability, the patient’s irritability level was rated as a continuous variable from 0 to 100 with the following anchors: 0 = no irritability, 25=low, 50=moderate irritability, 75=high, 100 = maximum irritability.

Statistical analysis

Descriptive statistics were calculated for the total sample and by anatomical location of pain (shoulder or low back). Differences in age, sex, pain intensity, and pain duration between anatomical locations were examined with a one-way ANOVA for continuous variables and a chi-square analysis for categorical variables. Because participants with shoulder and low back pain did not significantly differ by clinical or demographic variables listed in Table 1, we decided to combine the groups for statistical analysis and analyze the total sample.

Separate independent samples t-tests were conducted examining for group (irritable or not irritable) by biological, pain sensitivity, and psychological factors. For EIH, a repeated measures ANOVA examined for time (pre- and post-exercise PPT) by irritability category interaction effects. Simple effects decomposition with Bonferroni correction was performed.

Due to the large number of outcomes and moderate to large associations between psychological factors, a multivariate statistical approach was implemented. Assumptions of univariate normality, linearity, and homoscedasticity were examined. Three separate multivariate analysis of variances (MANOVAs) were performed for biological, psychological, and pain sensitivity factors. Alpha level was set at p < 0.05. For EIH, a repeated measures ANOVA examined for time (pre- and post-exercise PPT) by irritability level interaction effects. Simple effects decomposition with Bonferroni correction was performed.

Results

Demographic and clinical factors

Twenty participants with low back pain and twenty participants with shoulder pain completed the study (total n = 40). 13% of the sample were patients currently receiving treatment at a physical therapy clinic. Demographic and clinical characteristics for the total sample and by anatomical location of pain can be found in Table 2.

Table 2.

Demographic and clinical factors.

	Total Sample (n = 40)	Low Back Pain (n = 20)	Shoulder pain (n = 20)	p-value
Irritability Classification % irritable	32.50%	25.00%	35.00%	0.49
Irritability Level % high % medium % low	15.00% 27.50% 57.50%	25.00% 20.00% 55.00%	5.00% 35.00% 60.00%	0.17
101-point Irritability Scale	31.46 ± 22.88	30.10 ± 24.28	32.89 ± 21.87	0.71
Age	26.00 ± 10.62	23.63 ± 6.72	28.35 ± 13.21	0.15
Sex % female	60.00%	70.0%	50.00%	0.50
Race % White % African American % Asian % American Indian % Other	71.80% 5.10% 12.80% 5.10% 5.10%	68.4% 0.0% 21.1% 10.5% 0.0%	75.00% 10.00% 5.00% 0.0% 10.00%	0.15
Ethnicity % Hispanic	35.00%	35.00%	35.00%	0.54
Average Pain Intensity NPRS	33.11 ± 13.99	31.90 ± 12.95	35.76 ± 15.11	0.58
Pain Duration weeks	103.75 ± 141.62	132.40 ± 163.34	75.10 ± 112.95	0.06
Disability		7.50 ± 5.82	25.12 ± 13.63

Note: NPRS = Numerical Pain Rating Scale (101 point scale). p-value indicates differences between individuals with shoulder and low back pain.

Irritability category (irritable or not irritable)

32.50% of the total sample (n = 13) were classified as ‘irritable’ based on Maitland’s dichotomous classification of irritable or not irritable. Of the participants classified as ‘irritable,’ 41.70% were high irritability, 41.70% were moderate irritability, and 16.60% were low irritability. Of the participants were classified as ‘not irritable,’ 3.60% were high irritability, 21.40% were moderate irritability, and 75.0% were low irritability. Mean ± standard deviation irritability level based on the 101-point continuous scale (0=no irritability and 100=worst irritability imaginable) = 31.46 ± 22.88. Individuals classified as ‘irritable’ based on Maitland’s definition had significantly higher continuous irritability scores (p < 0.01, mean ± standard deviation for irritable = 58.16 ± 16.37 and not irritable = 19.81 ± 13.74).

Biologic factors

Irritability category did not significantly differ by age (p = 0.37), sex (p = 0.73), race (p = 0.83), ethnicity (p = 0.56), and pain duration (p = 0.78). Participants categorized as irritable displayed significantly higher scores on the ODI (p = 0.01, d = 1.25) but not the quickDASH (p = 0.11, d = 0.61).

Pain sensitivity

HPT: HPT was significantly lower at the tibialis anterior (t(38) = −2.29, p = .01, d=−.79) and deltoid (t(38) = −1.69, p = 0.04, d=−.61) in individuals categorized as irritable. HPT applied to the low back did not significantly differ by irritability category (t(38) = −0.47, p = 0.32,d = −0.16).

TS: Temporal summation did not differ t(38) = 0.28, (p = .36, d = .12), as well as aftersensations (p > 0.05, d range= −0.25 to −0.94).

PPT: As demonstrated in Figure 1, PPT was significantly lower across all sites for individuals categorized as irritable (low back (t(38) = −2.21, p = .02, d=−.71); deltoid (t(38) = 0.10, p = .01, d=−.79); tibialis anterior (t(38) = −2.79, p = .01,d = −1.07).

Figure 1.

PPT by irritability category.

Note: PPT=Pressure Pain Threshold, kPa=kilopascals, *indicates statistical significance p < 0.05.

CPM: CPM was significantly less efficient in individuals categorized as irritable (t(38) = 2.33, p = .02, d = .66).

EIH: PPT significantly differed at baseline between irritability classification (p < 0.01). EIH effects did not differ by irritability level F([1,37] = 0.71, p = 0.42, partial eta² = 0.02). A main effect of time was observed F [1,37] = 6.59, p = 0.01, partial eta² = 0.12). PPT applied to the quadriceps significantly decreased after the exercise (mean difference (95% confidence interval) = −44.66 (−79.89 - −9.43).

Psychological factors

As demonstrated in Figure 2, depression (t(38) = 4.86, p < .01, d = 2.12) and anger (t(38) = 1.99, p = .01, d = .79) were significantly higher with lower self-efficacy (t(38) = −2.71, p < .01,d = −1.01) in individuals categorized as irritable. Kinesiophobia, fear of pain, pain-anxiety, and pain catastrophizing did not differ (p > .05, d ranged from .12 to .52).

Figure 2.

Psychological factors by irritability category.

Note: *indicates statistical significance p < 0.05.

Irritability level (high, moderate, or low)

17.50% were classified as high irritability (n = 7), 25.00% were classified as moderate irritability (n = 10), and 57.50% were classified as low irritability (n = 23) based on the irritability levels in the Orthopedic Clinical Practice Guidelines. Irritability levels also (high, moderate, or low) significantly differed by the continuous irritability rating (p < 0.01, mean ± standard deviation high irritability = 62.42 ± 21.12, moderate irritability = 44.70 ± 12.06, low irritability = 17.96 ± 14.67).

Biologic factors

Irritability level (low, moderate, high) did not significantly differ by age (p = 0.85), sex (p = 0.88), race (p = 0.49), ethnicity (p = 0.69), pain duration (p = 0.54), and pain intensity (p = 0.25).

Pain sensitivity

The MANOVA was significant (Pillai’s Trace F [16,52] = 0.01, partial eta² = 0.42).

HPT: While the main effects of HPT applied to the low back were not significant (p = 0.68, partial eta² = 0.02), main effects of the deltoid (p = 0.03, partial eta² = 0.19) and tibialis anterior (p = 0.04, partial eta² = 0.17) were significant. HPT was significantly higher at the tibialis anterior and deltoid in individuals with low irritability compared to moderate (p = 0.04).

TS: TS (p = 0.27, partial eta² = 0.08) did not significantly differ by irritability category.

PPT: As demonstrated in Figure 3, the main effects of PPT were significant at the low back (p = 0.02, partial eta² = 0.21) and tibialis anterior (p = 0.01, partial eta² = 0.26) but not deltoid (p = 0.06, partial eta² = 0.16). PPT was significantly higher in individuals with low irritability compared to high (p = 0.01–0.02) across these sites.

Figure 3.

Irritability level by PPT.

Note: PPT=Pressure Pain Threshold, kPa=kilopascals, *indicates statistical significance p < 0.05.

CPM: CPM (p = 0.07, partial eta² = 0.15) was not significantly different between irritability levels.

EIH: EIH effects did not significantly differ by anatomical location of pain and both groups demonstrated decreases in PPT after exercise. PPT significantly differed at baseline between irritability level (p < 0.01). Significant time (pre/post) x irritability level (low, moderate, or high) interaction effects were not observed (F [2,36] = 1.47, p = 0.24, partial eta² = 0.08). A main effect of time was observed F [1,37] = 6.59, p = 0.01, partial eta² = 0.12). PPT applied to the quadriceps significantly decreased after the exercise (mean difference (95% confidence interval) = −44.66 (−79.89 - −9.43).

Psychological factors

A significant omnibus was not observed (Pillai’s Trace [18,50] = 0.96, p = 0.51, partial eta² = 0.26). As demonstrated in Figure 4, irritability significantly differed by self-efficacy (p = 0.04, partial eta² = 0.18). Individuals with low irritability displayed significantly higher self-efficacy than moderate and high irritability. Irritability levels did not significantly differ by depression (p = 0.06, partial eta² = 0.16), fear-avoidance beliefs work (p = 0.67, partial eta² = 0.03), fear avoidance beliefs physical activity (p = 0.08, partial eta² = 0.14), kinesiophobia (p = 0.54, partial eta² = 0.04), fear of pain (p = 0.86, partial eta² = 0.01), anxiety (p = 0.41, partial eta² = 0.06), catastrophizing (p = 0.32, partial eta² = 0.07), anger (p = 0.49, partial eta² = 0.04).

Figure 4.

Psychological factors by irritability level.

Note: *indicates statistical significance p < 0.05.

Discussion

The results of this study suggest irritability differs by pain sensitivity (HPT, PPT, CPM) and pain-related psychological factors (self-efficacy, depression, anger). Biopsychosocial factors contribute to differences in irritability which may have important clinical implications.

Irritability classification

Sixty percent of the participants in our study were classified as low irritability, 30% with moderate irritability, and 10% with high irritability. In comparison, a study of 101 patients with shoulder pain observed 30.2% with low irritability, 45.5% with moderate irritability, and 24.3% with high irritability [60]. The research team hypothesizes a greater number of patients with shoulder pain were classified as high irritability in the prior study as they only included patients seeking care at a physical therapy clinic [59]. Higher pain and disability is associated with seeking care for musculoskeletal pain [63]. Consistent with literature demonstrating 26.2% of patients with low back pain are irritable, 25% of patients with low back pain in our study were categorized as irritable [62].

Biological factors

Irritability level and category did not differ by demographic and clinical factors. A paucity of evidence exists describing differences in irritability by similar biological factors. Clinical factors are easily ascertained in the clinical setting and are part of the irritability classification published in Clinical Practice Guidelines [7]. While we anticipated individuals categorized as high irritability may report significantly greater pain intensity, our results were not statistically significant. The research team hypothesizes this may be due to naturally occurring differences in the number of participants per group, as well as overall lower pain intensity in the total sample as participants were recruited from the community and university campus clinic. Future research is needed to validate these results in a larger sample.

Pain sensitivity

A novel contribution of our study is that, to our knowledge, we are the first to examine biopsychosocial factors that contribute to differences in irritability. This provides important mechanistic understanding of factors that may be contributing to an irritable presentation and future novel treatment targets. Alterations in pain processing in response to mechanical and heat stimuli are demonstrated in patients with musculoskeletal pain [20,64–66]. Changes in pain sensitivity local to the anatomical location of pain and distant to the primary site may reflect changes in central pain processing [67–69]. Individuals with low back and shoulder pain who were classified as irritable displayed remote increased pain sensitivity as behaviorally measured with PPT and HPT at the tibialis anterior.

Individuals with musculoskeletal pain may display local and widespread hyperalgesia that support central and peripheral changes in pain sensitivity [19,20,70]. Hyperalgesia (increased pain sensitivity) may reflect increased responsiveness of peripheral nociceptors or increased synaptic activity in central dorsal horn neurons [18,69,71]. Individuals with an irritable presentation displayed widespread hyperalgesia to pressure and heat stimuli and significantly less efficient CPM, suggesting alterations in central pain inhibitory capacity [72]. Less efficient CPM has been previously demonstrated in patients with musculoskeletal pain [73–76]. This study adds to the current knowledge by now demonstrating alterations in pain sensitivity are demonstrated in patients with irritable presentations.

Exercise may also modulate the perception of pain yet inconsistent effects are observed in patients with persistent pain conditions [31,33]. Our results indicate EIH did not significantly differ by irritability level as significant interaction effects were not observed. A significant main effect of time (hyperalgesia) was observed, indicating exercise produced a significant reduction in pressure pain threshold, yet this effect was not unique to a particular irritability level. Consistent with prior literature, individuals with shoulder pain and low back pain demonstrate inconsistent effects in response to exercise with some displaying hyperalgesia [30].

Psychological factors

Negative (depression, anger) and positive (self-efficacy) psychological factors differ by irritability category. Psychological factors are implicated in the transition to chronic pain as they predict disability and pain intensity after musculoskeletal injury [22,77]. Psychological factors related to pain sensitivity as central sensitization mediates the relationship between depression, anxiety, catastrophizing and pain intensity [78]. Depression is associated with poorer clinical outcomes and development of persistent pain in patients with shoulder and low back pain. Self-efficacy, a positive factor, was significantly lower in individuals categorized as irritable and high irritability. Self-efficacy is inversely related to low back pain intensity and disability [79,80]. This suggests negative and positive psychological factors are integral to an irritable presentation.

Clinical implications

Collectively, the results of this study suggest biopsychosocial factors contribute to differences in irritability levels. Given that clinical characteristics of musculoskeletal pain are often greater between patients than by medical diagnosis [81], grouping patients based on mechanistic classifications of pain (nociceptive, neuropathic, nociplastic, or mixed pain) has been endorsed [82]. Mechanistic-based classifications are clinically relevant as treatment may be selected based on the underlying pain mechanism [83]. For example, a patient with a primarily nociceptive pain presentation is characterized by pain due to activation of nociceptors and may best respond to manual therapy treatment that is guided by the patient’s pain intensity levels. However, a patient with a primarily nociplastic pain presentation is characterized by pain that is due to altered nociception without clear tissue damage and may respond best to manual therapy treatment that is applied with a less aggressive technique applied with a graded application scheme [84,85].

An additional clinical implication of the results of this study are that dosing intervention intensity based on tissue irritability may lead to suboptimal management. Although this would need to be examined in future clinical trials, the results of this study suggest that pain sensitivity and psychological factors may be more precise treatment targets rather than minimizing stress to the tissue. As a step toward more precise clinical care, future clinical trials may aim to test this hypothesis.

Individuals identified as irritable may share similar widespread elevations in pain sensitivity and pain-related psychological factors, which define a nociplastic pain presentation [86,87]. These findings help to define irritability within the context of a mechanistic approach to managing patients with musculoskeletal pain conditions with the potential to assist rehabilitation practitioners in interpreting and responding to the clinical determination of level of irritability.

Limitations

Our sample size is small (n = 40) but robust enough to present preliminary data supporting our hypothesis. We also conducted a cross sectional study so it is unknown if these factors change over time as a patient becomes less irritable with treatment. Furthermore, we acknowledge that a small sample of participants were currently receiving physical therapy care which could impact the results.

Conclusions

Local and widespread changes in pain sensitivity, as well as heightened negative psychological factors, may contribute to an irritable presentation in patients with low back and shoulder pain.

Biographies

Abigail T. Wilson is an Assistant Professor in the Division of Physical Therapy at the University of Central Florida.

William J. Hanney is an Associate Professor in the Division of Physical Therapy at the University of Central Florida.

Randi Richardson is a Clinical Assistant Professor in the Division of Physical Therapy at the University of Central Florida.

Sheila Klausner is a Clinical Assistant Professor in the Division of Physical Therapy at the University of Central Florida.

Joel E. Bialosky is a Clinical Professor in the Department of Physical Therapy at the University of Florida.

Funding Statement

There is no funding to report.

Clinical trials registry

Not Applicable for this study – This is an observational study in which participants were not assigned interventions.

Data availability statement

Data set available upon request.

Disclosure statement

No potential conflict of interest was reported by the authors.

Institutional review board approval

This study was approved by the University of Central Florida Institutional Review Board for Human Subjects Research. All participants provided written informed consent to enroll in the study.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/10669817.2023.2294679