Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2022 Feb 3;26(4):855–872. doi: 10.1002/ejp.1912

Psychosocial and psychophysical assessment in paediatric patients and young adults with chronic back pain: A cluster analysis

D D Ocay 1,2, A Loewen 3, S Premachandran 1,2, P M Ingelmo 4,5, N Saran 6, J A Ouellet 6, Catherine E Ferland 1,2,5,7,✉
PMCID: PMC9304192  PMID: 35090183

Abstract

Background

Identifying subgroups with different clinical profiles may inform tailored management and improve outcomes. The objective of this study was to identify psychosocial and psychophysical profiles of children and adolescents with chronic back pain.

Methods

One hundred and ninety‐eight patients with chronic back pain were recruited for the study. Pain assessment was mainly conducted in the form of an interview and with the use of validated pain‐related questionnaires assessing their psychosocial factors and disability. All patients underwent mechanical and thermal quantitative sensory tests assessing detection and pain thresholds, and conditioned pain modulation efficacy.

Results

Hierarchal clustering partitioned our patients into three clusters accounting for 34.73% of the total variation of the data. The adaptive cluster represented 45.5% of the patients and was characterized to display high thermal and pressure pain thresholds. The high somatic symptoms cluster, representing 19.2% of patients, was characterized to use more sensory, affective, evaluative and temporal descriptors of pain, more likely to report their pain as neuropathic of nature, report a more functional disability, report symptoms of anxiety and depression and report poor sleep quality. The pain‐sensitive cluster, representing 35.4% of the cohort, displayed deep tissue sensitivity and thermal hyperalgesia.

Conclusions

This study identified clinical profiles of children and adolescents experiencing chronic back pain based on specific psychophysical and psychosocial characteristics highlighting that chronic pain treatment should address underlying nociceptive and non‐nociceptive mechanisms.

Significance

To our current knowledge, this study is the first to conduct cluster analysis with youth experiencing chronic back pain and displays clinical profiles based on specific physical and psychosocial characteristics. This study highlights that in a clinical context, chronic pain assessment should include multiple elements contributing to pain which can be assessed in a clinical context and addressed when pathoanatomical symptoms are unidentifiable.

1. INTRODUCTION

Chronic or recurrent back pain in the paediatric population is less prevalent than adults, affecting 14–24% of children and adolescents and is usually associated with post‐trauma or known severe pathological conditions (Altaf et al., 2014; Balague et al., 1999; Davis & Williams, 2008; Haidar et al., 2011; Moreno, 2017). However, when pathoanatomical symptoms are unidentifiable, the diagnosis is labelled as non‐specific chronic back pain. Patients with chronic back pain experience functional disability, higher rates of missed school, poor sleep quality and mental health problems when compared to age‐matched pain‐free controls (Balagué et al., 1995; Huguet & Miro, 2008; O'Sullivan et al., 2011; Watson et al., 2002; Wojtowicz & Banez, 2015), and are at risk of experiencing chronic pain throughout adulthood (Brattberg, 2004; Hestbaek et al., 1976; Jeffries et al., 2007; Mikkelsson et al., 2008).

A major limitation in treatment outcomes for chronic back pain is the heterogeneity of the population. Moreover, there are limited paediatric studies, especially randomized control trials, that have documented standardized measures associated with treatment response (McGrath et al., 2008; Randall et al., 2018; Simons et al., 2018). We have previously shown that different pain processing mechanisms may be involved in adolescents with idiopathic scoliosis and chronic back pain (Teles et al., 2019). These results highlight that despite the similar diagnosis, characterizing the psychophysical profile of patients with chronic pain through quantitative sensory tests (QST) may be relevant to consider as a component to guide pain management to become tailored to address underlying etiological mechanisms.

Due to the heterogeneity within chronic pain conditions and that different chronic pain conditions may share similar characteristics (Diatchenko et al., 2006), researchers and clinicians have turned to identifying subgroups with distinct psychophysical profiles in different samples of patients with chronic pain. Subgroups of adult patients with chronic low back pain (Coronado et al., 2014; Rabey et al., 2015), temporomandibular disorder (Bair et al., 2016) and other chronic pain conditions (Baron et al., 2017) have been successfully identified. Rabey et al. (2015) investigated subgroups in a cohort of chronic low back pain based on their QST results. They identified three clusters in which those that displayed increased thermal and pressure pain sensitivity had a greater proportion of females, and higher scores for depression and poor sleep quality (Rabey et al., 2015). The main limitation of this study was including only the QST results as factors in their cluster analysis. Numerous factors influence quantitative sensory testing, such as age, sex and psychosocial factors (Blankenburg et al., 2010; Cornelissen et al., 2014; Hirschfeld et al., 2012; Rolke et al., 2006). Including these factors within the cluster analysis may give more insight into data interpretation and interventions tailored for these subgroups. Bair et al. (2016) included psychophysical and psychosocial measures in their cluster analysis. However, they used a supervised cluster approach which involves selecting specific variables in their analysis aligning with their objective to identify risk factors for chronic pain in healthy individuals who have a similar psychophysical profile as patients with temporomandibular disorders (Bair et al., 2016).

Researchers and clinicians have also turned to identify heterogeneous subgroups of paediatric chronic pain patients (Scharff et al., 2005; Schurman et al., 2008; Wager et al., 2014; Walker et al., 2012). However, these studies strictly investigated pain and psychosocial characteristics in their cluster analysis and there are limited data evaluating subgroups based on the psychophysical profile of paediatric chronic pain patients. Therefore, the objective of this study was to identify specific psychophysical and psychosocial profiles among a cohort of paediatric patients with chronic back pain. The aim was to conduct an unsupervised statistical clustering approach involving the QST results and psychosocial context of the patients. We hypothesized that subgroups of patients with chronic back pain can be clustered based on similar psychophysical and psychosocial characteristics.

2. Methods

2.1. Study approval

Ethics approval was obtained prior to the beginning of the recruitment from the Research Ethics Board of McGill University (A11‐M62‐15B). Participants received written informed consent prior to inclusion in the study and a signature was obtained by the participant or their parent/legal guardian, if the participant was under the age of 14 years old, prior to the beginning of the study. The study was conducted in accordance with the Declaration of Helsinki. All participants were de‐identified according to the institutional ethics guidelines.

2.2. Participants

Patient recruitment occurred between January 2016 and October 2017. Potential participants from the spine and orthopaedic outpatient clinics and the Chronic Pain Services of our institution were identified by a research assistant based on the presence of chronic pain reported in their electronic medical charts or by reference of the patient's physician. At their hospital visit for treatment‐seeking either for an orthopaedic condition or for pain itself, patients were approached by a research assistant to participate in the study and to confirm eligibility criteria prior to receiving signed consent. Inclusion criteria were being aged between 10 and 21 years old with chronic back pain (persistent or recurrent pain at least once a week for longer than three months) (Treede et al., 2019). Patients who did not speak English or French or had a diagnosis of developmental delay that would interfere with completing measures were excluded.

2.3. Primary outcome measures

2.3.1. Sociodemographic characteristics and medical history

Patient characteristics such as age, sex, ethnicity and pathology were collected by a research assistant.

2.3.2. Clinical characteristics

Pain assessment was mainly conducted in the form of a face‐to‐face interview and with the use of standardized pain‐related questionnaires that have been validated in clinical paediatric studies assessing pain (Claar & Walker, 2006; David et al., 2015; Palermo, 2009; Siu et al., 2012). Patients were asked about the duration and frequency of their pain. The location of pain was reported using a modified version of the adolescent paediatric pain tool (APPT) (Fernandes et al., 2014), in which a diagram of the back was divided into 10 segments to identify specific pain locations (Savedra et al., 1989; Teles et al., 2019). In addition, pain intensity experienced over the last month in each divided back segment of the diagram was reported using an 11‐point numerical rating scale (NRS 0–10, 0 = no pain, 10 = the worst pain imaginable). Moreover, the pain experience was assessed using a list of 67 descriptive words in the APPT, assessing the four dimensions of pain (sensory, affective, evaluative and temporal) (Savedra et al., 1993). The APPT has been shown to have adequate content, construct, and criterion validity and reliability in clinical and non‐clinical groups of children and adolescents between 8 and 17 years old (Jacob et al., 2014). To identify if their pain had a neuropathic component, the Douleur Neuropathique 4 (DN4) questionnaire was completed by patients. By summing all 10 questions, scores equal to or greater than 4 indicated that the pain experienced by the patient is likely neuropathic (Bouhassira et al., 2005; David et al., 2015). The DN4 questionnaire has not been validated in children and adolescents. However, despite its very low‐level evidence for satisfactory criterion validity and low‐level evidence for satisfactory construct validity and reliability, the DN4 questionnaire has been described to be the most suitable for clinical use (de Leeuw et al., 2020; Mathieson et al., 2015).

2.3.3. Anxiety and depressive symptoms

The revised child anxiety and depression scale (RCADS) questionnaire was completed by patients to assess children's self‐report of depression and anxiety corresponding to the 4th edition of the diagnostic and statistical manual of mental disorders (Chorpita et al., 2000). Based on the patient's age and grade in school, their total scores are converted into a T‐score, in which a T‐score between 65 and 69 indicates borderline clinical threshold, and a T‐score of 70 or higher indicates above clinical threshold for anxiety and depression. The RCADS has been validated in clinical and non‐clinical groups of children and adolescents in grades 3–12 and showed good internal consistency (Cronbach α = 0.78–0.88) and item set and factor definitions consistent with DSM‐IV anxiety disorders and depression (Chorpita et al., 2000, 2005).

2.3.4. Functional disability

The functional disability inventory (FDI) questionnaire was completed by patients, in which the total score is summed to detect different levels of disability (Walker & Greene, 1991). The FDI has been reported to have high internal consistency, moderate to high test‐retest reliability, moderate cross‐informant (parent‐child) reliability and good predictive validity (Claar & Walker, 2006; Walker & Greene, 1991). The FDI is based on four‐level classifications system: A score of 0 to 12 inclusively represents no/minimal disability and patients can function well, despite experiencing pain; a score from 13 to 20 inclusively represents mild disability; a score from 21 to 29 inclusively represents moderate disability; a score of 30 or higher represents severe disability.

2.3.5. Sleep quality

The Pittsburgh sleep quality index (PSQI) questionnaire was completed by patients to assess sleep quality, in which a global score of 5 or higher indicated poor sleep quality (Buysse et al., 1989). The PSQI is the most commonly used measure in clinical and research settings showing good internal consistency (Cronbach α = 0.70–0.83) and has been validated in clinical and non‐clinical groups of adolescents (Larche et al., 2021; Mollayeva et al., 2016; Raniti et al., 2018).

2.3.6. Quantitative sensory testing

Each patient underwent a specific protocol of mechanical and thermal quantitative sensory tests (QST) to obtain a comprehensive profile of somatosensory functioning. The protocol was based on an initiative of the Quebec Pain Research Network (Ferland et al., 2018a, 2018b. All tests were conducted by research assistants who were trained and evaluated by the principal investigator of the study. Mechanical and thermal procedures were performed on the left volar forearm, 2 inches from the left elbow crease as the control area and followed by the most painful anatomical region of the back indicated by the patient as the affected area. A demonstration of every test was explained and performed on the left thenar eminence of the patient. The protocol previously described (Teles et al., 2019) consisted of four tests assessing six parameters: Mechanical detection threshold, pressure pain threshold, heat pain threshold, heat tolerance threshold, temporal summation of pain and conditioned pain modulation.

Mechanical quantitative sensory testing assessment

Mechanical detection threshold (MDT), using standardized von Frey filaments (Touch‐Test™ Sensory Evaluators, USA) with forces ranging between 0.008 and 300 grams, was evaluated to assess tactile sensitivity (Blankenburg et al., 2010; Teles et al., 2019; Thibault et al., 1994). The geometric mean of six threshold values was calculated and reported in grams. Pressure pain threshold (PPT), using the JTech Algometer (JTech Medical, USA) with a 1‐cm2 probe, was evaluated to assess deep‐tissue sensitivity (Blankenburg et al., 2010; Teles et al., 2019). The pressure was applied increasing at a rate of ~1 N/s (~10 kPa/s) until the patient reported pain. The mean of three recorded values was calculated and reported in Newtons.

Thermal quantitative sensory testing assessment

Heat pain threshold (HPT) and heat pain tolerance threshold (HTT) was evaluated using a 9‐cm2 warm calibrated thermode connected to the Q‐sense apparatus (Medoc, Israel). The thermode, initially set at 32.0°C, was placed on the left volar forearm of the patient and increased at a rate of 0.3°C/second to reach the maximum value of 50.0°C as a security cut‐off. HPT (when the patient first report pain) and HTT (when the pain was intolerable) were assessed three times and the mean was calculated and reported in degree Celsius.

A conditioned pain modulation (CPM) paradigm was then performed using tonic heat on the right forearm as the test stimulus and the cold pressor task on the left arm as the conditioning stimulus as previously described protocols (Ferland et al., 2018a, 2018b; Potvin & Marchand, 2016; Teles et al., 2019). First, a thermode was applied to the forearm to reach a pre‐determined test temperature to a 5/10 pain intensity. Once the target temperature was reached, it remained constant for 120 seconds. Patients were not told that the temperature of the thermode would remain constant over time to avoid expectation effects. Using a computerized pain scale (CoPS 0–10; 0 = no pain, 10 = the worst pain imaginable), patients were asked to continuously rate their pain to identify if there is temporal summation of pain (TSP) (Ferland et al., 2018a, 2018b; Teles et al., 2019). The presence of temporal summation (i.e. endogenous facilitatory pain response) was defined as a 2/10 increase in pain intensity using the CoPS at the end of the test in comparison to the pain intensity 60 s after the beginning of the test. A change in pain intensity of 2/10 on a NRS was determined as a minimum clinically significant difference (Farrar et al., 2001). Once the tonic heat test was completed, patients performed a cold pressor task (CPT) involving the immersion of their forearm in a filled with cold water (12°C) for 2 min to trigger the descending inhibitory pain response. The CPT was immediately followed by a second tonic heat test. The patient's capacity to endogenously inhibit pain was described previously as the diffuse noxious inhibitory control, and here measured as the CPM efficiency was then calculated as the percentage difference between the mean pain intensity of the test stimulus before and after the conditioning stimulus over the mean pain intensity during the test stimulus before the conditioning stimulus. A negative percentage result under −30% indicated an optimal inhibitory pain response, a negative percentage result between −10 and −30% indicated a suboptimal inhibitory pain response, and a negative percentage result above −10% or a positive percentage result indicated an inefficient or facilitatory pain response (Ferland et al., 2018a, 2018b; Teles et al., 2019). A 10%–30% reduction in pain was labelled to be a minimal improvement, whilst a 30% reduction in pain intensity was labelled to be a clinically important difference in pain intensity (Farrar et al., 2001) and is approximately the mean value of inhibitory CPM observed in previous studies (Ferland et al., 2018a, 2018b; Potvin & Marchand, 2016; Teles et al., 2019; Tousignant‐Laflamme et al., 2008).

2.4. Statistical analysis

Descriptive statistics were performed using the R Studio software to summarize the collected data regarding the patients’ characteristics, clinical data relative to pain, psychosocial factors and QST results. Sample size requirements for principal component analysis (PCA) are not definitive and are dependent on many factors. Therefore, the sample size was based on population proportion in which minimally 14% of the paediatric population is affected by chronic back pain. Based on this assumption, a sample size of 185 patients is required to achieve 90% statistical power at the 0.05 significance level.

An unsupervised cluster analysis was performed using the FactoMineR package in the R Studio software (Le et al., 2008) to subgroup patients into clinical profiles and potentially identify responders to specific therapeutic strategies. To profile the patients based on their psychophysical and psychosocial characteristics, the cluster analysis involved 17 indicator variables: sensory descriptors, affective descriptors, evaluative descriptors, temporal descriptors, DN4 total score, FDI total score, RCADS total T‐score, PQSI global score, mechanical detection threshold in the control and affected area, pressure pain threshold in the control and affected area, heat pain threshold, heat tolerance threshold, the average pain score during the cold pressor task, CPM efficiency score and the pain score during the thermal temporal summation of pain. Other quantitative and qualitative outcome measures were included as supplementary variables as they do not represent underlying mechanisms of pain and instead may represent consequences of chronic back pain: location of recruitment, age, sex, ethnicity, duration of pain, frequency of pain, duration of painful episodes, pathology, most painful location, average pain reported in the back, pain radiating down the legs and test temperature for the CPM assessment. Since all measures had different units, iterative PCA using the FactorMineR package in the R Studio software was first conducted as a data reduction technique standardizing all variables into Z‐scores. Principal component analysis was conducted to investigate interrelationships between and within psychophysical and psychosocial variables to determine whether a smaller number of principal components is representative of the total variation in the data. Standardization of all variables was to ensure equal importance of each variable in the PCA. Missing data for a maximum of two variables were observed for eight patients. No differences were observed in these eight patients in comparison to the rest of the sample regarding their demographic characteristics (data not shown). Therefore, these eight patients were kept in the analysis. Missing data were imputed for the indicator variables using the missMDA package which takes into account similarities between the values of the variables of each patient (Josse & Husson, 2016; Le et al., 2008). Principal components (PCs) with eigenvalues >1 were retained (Hair, 2010). Variable loading on each principal component was considered significant if >0.3 (Hair, 2010). Hierarchical clustering with k‐means consolidation was conducted on the principle components. The hierarchical clustering was, therefore, performed multiple times to minimize within‐cluster variability and maximize between‐cluster variability. The best partition of clusters was the one with the highest relative loss of inertia (Hair, 2010). An analysis of variance (ANOVA) model was conducted along with a Fisher test to determine which principal components best represent each cluster and determine cluster effect. Differences between clusters regarding their characteristics, clinical data relative to pain, psychosocial factors and QST results was conducted using the chi‐squared test and Kruskal‐Wallis one‐way ANOVA followed by Dunn's test depending on whether the variable was qualitative or quantitative, respectively.

3. RESULTS

Two hundred and four patients were recruited for this cross‐sectional study. However, six patients dropped out prior to the quantitative sensory tests. Therefore, the data of 198 patients with chronic back pain were analysed, in which 170 (85.9%) were recruited from the spine and orthopaedic outpatient clinics whilst 28 (14.1%) the chronic pain services of our institution. The mean age was 15.69 ± 2.25 years old and 81.8% of our cohort were females (Table 1). The majority of the patients were Caucasian (90.4%), experience pain for more than 12 months (72.2%), experience pain on a daily basis (65.2%) and experience constant painful episodes (55.6%). Moreover, 25.8% of the cohort reported back pain radiating down their legs, whilst 27.8% of our cohort self‐report their pain as most likely to be neuropathic. Among the cohort, 71.2% reported their most painful location along their spine. Furthermore, 54.6% of the cohort self‐reported mild to severe functional disability, 7.6% self‐reported borderline of above clinical threshold symptoms of anxiety and depression and 72.7% self‐reported poor sleep quality. Large variability was observed for the QST results in the cohort. The inhibitory pain control assessment revealed great variability among the cohort. The CPM efficiency was optimal in 51.5%, suboptimal in 22.7% and inefficient in 25.8% of the cohort. Moreover, 13.6% of the cohort displayed temporal summation of pain.

TABLE 1.

Demographics, clinical data relative to pain and psychosocial and psychophysical characteristics of cohort

Variable Total sample (n = 198)
Location of recruitment, n (%)
Spine and orthopaedic outpatient clinics 170 (85.9)
Chronic pain services 28 (14.1)
Age, Mean ± SD 15.69 ± 2.25
Sex, n (%)
Female 162 (81.8)
Male 36 (18.2)
Ethnicity, n (%)
Caucasian 179 (90.4)
Black or African American 10 (5.1)
Asian 4 (2.0)
Interracial 5 (2.5)
Duration of pain, n (%)
3–6 months 14 (7.1)
6–12 months 41 (20.7)
> 12 months 143 (72.2)
Frequency of pain, n (%)
Daily 129 (65.2)
Every 2nd day 43 (21.7)
Once a week 26 (13.1)
Duration of painful episodes, n (%)
Few seconds 8 (4.0)
Few minutes 36 (18.2)
One hour 44 (22.2)
Constant 110 (55.6)
Pathology, n (%)
Arthritic 6 (3.0)
Disc protrusion 8 (4.0)
Mechanical back pain 14 (7.1)
Scoliosis 115 (58.1)
Spondylolysis/Spondylolisthesis 13 (6.6)
Tight hamstrings 9 (4.5)
Non‐specific back pain 33 (16.7)
Most painful location, n (%)
Neck 3 (1.5)
Left upper back 6 (3.0)
Center upper back 38 (19.2)
Right upper back 11 (5.6)
Left middle back 8 (4.0)
Center middle back 37 (18.7)
Right middle back 12 (6.1)
Left lower back 12 (6.1)
Center lower back 63 (31.8)
Right lower back 6 (3.0)
Average pain reported, NRS (0–10), Mean (CI)
Neck 2.91 (2.50–3.32)
Left upper back 2.73 (2.32–3.14)
Center upper back 3.44 (3.00–3.87)
Right upper back 2.46 (2.05–2.87)
Left middle back 2.80 (2.39–3.21)
Center middle back 4.32 (3.90–4.74)
Right middle back 2.56 (2.16–2.95)
Left lower back 3.07 (2.61–3.53)
Center lower back 4.09 (3.63–4.54)
Right lower back 3.18 (2.73–3.64)
Pain radiating down legs, n (%)
Yes 51 (25.8)
No 140 (70.7)
Descriptors of pain used, Mean (%) ± SD
Sensory 18.04 ± 11.61
Affective 8.92 ± 12.00
Evaluative 34.13 ± 21.41
Temporal 23.94 ± 13.78
Neuropathic component, n (%)
Mean score of DN4 questionnaire, Mean ± SD 2.46 ± 2.08
Likely neuropathic 55 (27.8)
Not likely neuropathic 143 (72.2)
Functional Disability, n (%)
Mean score of FDI, Mean ± SD 15.43 ± 10.31
None or minimal 89 (44.9)
Mild 50 (25.3)
Moderate 37 (18.7)
Severe 21 (10.6)
Anxiety and Depression Symptoms, n (%)
Mean T‐score of RCADS, Mean ± SD 45.34 ± 12.39
Below clinical threshold 183 (92.4)
Borderline 5 (2.5)
Above clinical threshold 10 (5.1)
Sleep Quality, n (%)
Mean global score of PSQI, Mean ± SD 6.98 ± 3.48
Good sleep quality 54 (27.3)
Poor sleep quality 144 (72.7)
MDT (g), Mean ± SD
Control area 0.52 ± 1.65
Affected area 1.47 ± 12.37
PPT (N), Mean ± SD
Control area 27.62 ± 14.82
Affected area 26.38 ± 17.44
HPT (°C), Mean ± SD 39.24 ± 3.17
HTT (°C), Mean ± SD 45.16 ± 2.41
Test temperature for CPM assessment (°C), Mean ±SD 43.56 ± 2.51
CPT average pain score NRS (0–10), Mean ± SD 6.98 ± 2.32
CPM, n (%)
CPM efficiency (%), Mean ±SD −29.44 ± 42.87
Inefficient 51 (25.8)
Suboptimal 45 (22.7)
Optimal 102 (51.5)
TSP, n (%)
TSP pain score NRS (0–10), Mean ±SD 0.09 ± 2.07
No presence 171 (86.4)
Presence 27 (13.6)
Open in a new tab

Abbreviations: CI, 95% confidence interval; CPM, conditioned pain modulation; CPT, cold pressor task; DN4, douleur neuropathique 4 questionnaire; FDI, functional disability index; HPT, heat pain threshold; HTT, heat tolerance threshold; MDT, mechanical detection threshold; PPT, pressure pain threshold; PSQI, Pittsburgh sleep quality index; RCADS, revised children's anxiety and depression scale; SD, standard deviation; TSP, temporal summation of pain.

3.1. Principal component analysis

Iterative principal component analysis derived five principal components (PC) with eigenvalues >1 accounting for 59.2% of the total variation in the data. Variable loading on each principal component is summarized in Table 2. The PCs can be summarized as representing the dimensions of psychosocial factors (PC1), pressure pain and heat tolerance thresholds (PC2), mechanical detection threshold (PC4) and CPM efficiency (PC5). No significant variable loading was observed for PC3. The PC scores were calculated for each patient using the component loadings and were used to replace the indicator variables in the cluster analysis.

TABLE 2.

Principal component analysis of psychophysical variables

Variable Component 1 Component 2 Component 3 Component 4 Component 5
Sensory descriptors 0.556 0.115 0.054 0.007 0.003
Affective descriptors 0.463 0.053 0.017 0.050 0.008
Evaluative descriptors 0.445 0.022 0.103 0.001 0.009
Temporal descriptors 0.188 0.071 0.144 0.012 0.005
DN4 Total score 0.321 0.080 0.004 0.074 0.058
FDI Total score 0.521 0.058 0.048 0.008 0.005
Anxiety and Depression Total T‐score 0.303 0.003 0.169 0.075 0.005
PSQI score 0.295 0.054 0.209 0.037 0.020
MDT control 0.005 0.016 0.001 0.496 0.008
MDT affected 0.016 0.003 0.069 0.115 0.377
PPT control 0.140 0.401 0.121 0.036 0.007
PPT affected 0.136 0.319 0.202 0.090 0.000
HPT 0.142 0.289 0.072 0.039 0.007
HTT 0.164 0.446 0.116 0.015 0.006
CPT average pain score 0.130 0.259 0.012 0.000 0.181
CPM efficiency 0.003 0.050 0.002 0.156 0.451
TSP pain score 0.002 0.136 0.122 0.008 0.185
Open in a new tab

Variable loading on each component was considered significant if >0.3 (bolded).

Abbreviations: CPM, conditioned pain modulation; CPT, cold pressor task; DN4, douleur neuropathique 4 questionnaire; FDI, functional disability index; HPT, heat pain threshold; HTT, heat tolerance threshold; MDT, mechanical detection threshold; PPT, pressure pain threshold; PSQI, Pittsburgh sleep quality index; TSP, temporal summation of pain.

3.2. Cluster analysis

Hierarchal clustering partitioned our patients into three clusters accounting for 34.73% of the total variation in the data. Eighty‐nine patients (44.9%) were grouped in cluster 1, 71 patients (35.9%) and 38 patients (19.2%) were grouped in cluster 2 and cluster 3, respectively. Figure 1 displays the three clusters according to principal components 1 and 2. Patients grouped in cluster 1 are characterized by significantly low values for PC1 (t = 5.77, p < 0.001) and high values for PC2 (t = 5.43, p < 0.001) and we, therefore, named it the adaptive cluster. In contrast, patients grouped in cluster 2 are mainly characterized by significantly low values for PC2 (t = 7.54, p < 0.001) and was, therefore, named the pain‐sensitive cluster. Moreover, patients grouped in cluster 3 are mainly characterized by significantly high values for PC1 (t = 11.54, p < 0.001) and thus named the high somatic symptoms cluster.

FIGURE 1.

FIGURE 1

Open in a new tab

Results of the hierarchical clustering analysis displaying three clusters derived from the principal component (PC) scores of PC1 and PC2 representing the dimensions of psychosocial factors, and pressure pain and heat tolerance thresholds, respectively. The adaptive cluster is characterized by significantly higher values for pressure pain and heat tolerance thresholds. In contrast, the pain‐sensitive cluster is mainly characterized by lower values for pressure pain and heat tolerance thresholds. The high somatic symptoms cluster is mainly characterized by having the highest values for the dimension of psychological factors

3.3. Profiling of clusters

No significant differences were observed between clusters in regard to their age, sex, ethnicity, duration of pain, duration of painful episodes, pathology or location of pain (Table 3). However, a significant association was observed between cluster membership and the location of recruitment in the study (p < 0.001) and the reported frequency of pain (p = 0.008). A higher proportion of patients grouped in the high symptomatic symptoms cluster were recruited from the chronic pain services at our institution and all reported pain at least every second day. Moreover, the high somatic symptom cluster reported significantly higher pain intensity in all regions of the back (p < 0.05). Furthermore, a higher proportion of patients in the high somatic symptoms clusters reported their back pain radiating down their legs (p < 0.001).

TABLE 3.

Differences between clusters regarding demographics and clinical data relative to pain

Variable Adaptive cluster (n = 89) Pain‐sensitive cluster (n = 71) High somatic symptoms cluster (n = 38) χ2 value p‐value
Location of recruitment, n (%)
Spine and orthopaedic outpatient clinics 83 (93.3) 62 (87.3) 25 (65.8) 16.75* <0.001
Chronic pain services 6 (6.7) 9 (12.7) 13 (34.2)
Age, Mean ±SD 15.74 ± 2.15 15.32 ± 2.34 16.26 ± 2.25 2.23 † 0.110
Sex, n (%)
Female 69 (77.5) 61 (85.9) 32 (84.2) 2.05* 0.359
Male 20 (22.5) 10 (14.1) 6 (15.8)
Ethnicity, n (%)
Caucasian 82 (92.1) 60 (84.5) 37 (97.4) 11.00* 0.088
Black or African American 2 (2.2) 8 (11.3) 0 (0.0)
Asian 3 (3.4) 1 (1.4) 0 (0.0)
Interracial 2 (2.2) 2 (2.8) 1 (2.6)
Duration of pain, n (%)
3–6 months 7 (7.9) 6 (8.5) 1 (2.6) 1.91* 0.751
6–12 months 20 (22.5) 14 (19.7) 7 (18.4)
> 12 months 62 (69.7) 51 (71.8) 30 (78.9)
Frequency of pain, n (%)
Daily 55 (61.8) 40 (56.3) 34 (89.5) 13.71* 0.008
Every 2nd day 20 (22.5) 19 (26.8) 4 (10.5)
Once a week 14 (15.7) 12 (16.9) 0 (0.0)
Duration of painful episodes, n (%)
Few seconds 6 (6.7) 2 (2.8) 0 (0.0) 7.78* 0.255
Few minutes 16 (18.0) 16 (22.5) 4 (10.5)
One hour 20 (22.5) 17 (23.9) 7 (18.4)
Constant 47 (52.9) 36 (50.7) 27 (71.1)
Pathology, n (%)
Arthritic 3 (3.4) 1 (1.4) 2 (5.3) 16.48* 0.170
Disc protrusion 4 (4.5) 1 (1.4) 3 (7.9)
Mechanical back pain 9 (10.1) 3 (4.2) 2 (5.3)
Scoliosis 52 (58.4) 47 (66.2) 16 (42.1)
Spondylolysis/Spondylolisthesis 7 (7.9) 5 (7.0) 1 (2.6)
Tight hamstrings 4 (4.5) 3 (4.3) 2 (5.3)
Non‐specific back pain 10 (11.2) 11 (15.5) 12 (31.6)
Most painful location, n (%)
Neck 2 (2.2) 0 1 (2.6) 17.10* 0.516
Left upper back 2 (2.2) 1 (1.4) 3 (7.9)
Center upper back 13 (14.6) 15 (21.1) 10 (26.3)
Right upper back 5 (5.6) 5 (7.0) 1 (2.6)
Left middle back 5 (5.6) 3 (4.3) 0
Center middle back 15 (16.9) 14 (19.7) 8 (21.1)
Right middle back 6 (6.7) 5 (7.0) 1 (2.6)
Left lower back 6 (6.7) 6 (8.5) 0
Center lower back 32 (36.0) 20 (28.2) 11 (28.9)
Right lower back 3 (3.4) 1 (1.4) 2 (5.3)
Average pain reported, NRS (0–10), Mean (CI)
Neck 2.82 (2.22–3.42) c 2.28 (1.65–2.90) c 4.29 (3.21–5.37) a , b 10.84 † 0.004
Left upper back 2.30 (1.75–2.85) c 2.19 (1.58–2.81) c 4.75 (3.61–5.89) a , b 17.51 † <0.001
Center upper back 2.89 (2.29–3.48) c 3.32 (2.58–4.07) c 4.93 (3.87–6.00) a , b 10.63 † 0.005
Right upper back 2.08 (1.52–2.65) c 1.99 (1.37–2.60) c 4.22 (3.10–5.35) a , b 14.40 † <0.001
Left middle back 2.43 (1.86–3.00) c 2.56 (1.88–3.24) c 4.11 (3.03–5.19) a , b 8.58 † 0.014
Center middle back 3.87 (3.26–4.47) c 4.27 (3.55–4.99) 5.47 (4.53–6.42) a 7.68 † 0.021
Right middle back 2.15 (1.57–2.74) c 2.49 (1.87–3.11) 3.63 (2.60–4.66) a 7.64 † 0.022
Left lower back 2.65 (1.99–3.31) c 2.83 (2.09–3.57) c 4.50 (3.33–5.67) a , b 10.04 † 0.007
Center lower back 3.70 (3.06–4.33) c 3.65 (2.86–4.44) c 5.82 (4.77–6.87) a , b 13.54 † 0.001
Right lower back 2.37 (1.74–3.00) c 3.08 (2.36–3.81) c 5.28 (4.19–6.37) a , b 19.97 † <0.001
Pain radiating down legs, n (%)
Yes 18 (20.2) 14 (19.7) 19 (50.0) 18.00* <0.001
No 70 (78.7) 55 (77.5) 15 (39.5)
Open in a new tab

p‐values ≤0.05 are bolded.

Abbreviations: CI, 95% confidence interval; SD, standard deviation.

a

Significant difference with cluster 1.

b

Significant difference with cluster 2.

c

Significant difference with cluster 3.

*

Chi‐squared test statistic.

†

Kruskal‐Wallis one‐way analysis of variance chi‐squared test statistic.

Figure 2 displays the Z‐scores for the indicator variables for the respective three clusters. Significant between‐cluster differences in regard to the raw data of the indicator variables were observed (Table 4). The high somatic symptoms cluster was characterized to significantly have the highest scores for all the questionnaires completed (p < 0.001). The high somatic symptoms cluster were characterized to group patients who used more sensory, affective, evaluative and temporal descriptors of pain, more likely reported their pain as neuropathic of nature, reported more functional disability, reported symptoms of anxiety and depression, and reported poor sleep quality. The adaptive cluster, in comparison to the pain‐sensitive and high somatic symptoms clusters, was characterized to significantly have the highest pressure pain threshold in the control and affected area, highest heat pain and tolerance threshold, and lowest pain intensity reported during the cold pressor task (p < 0.001). Interestingly, patients in the adaptive cluster had a higher proportion of patients that display temporal summation pain than the pain‐sensitive cluster and the high somatic symptoms cluster (p = 0.005). The pain‐sensitive cluster, in general, displayed lower pressure pain threshold in the control and affected area, lower heat pain and tolerance threshold, and higher pain intensity reported during the cold pressor task than the adaptive cluster (p < 0.001), but also displayed lower scores for all the questionnaires completed than the high somatic symptoms cluster (p < 0.001).

FIGURE 2.

FIGURE 2

Open in a new tab

Z‐scores of the indicator variables. These plots show the mean values and associated 95% confidence intervals of the indicator variables included in the cluster analysis. Each variable displayed in the figure was normalized to have a mean of 0 and SD 1. Z‐scores >0 represent values higher than then the mean of the sample, and z‐scores <0 represent values lower than the mean of the sample. DN4, douleur neuropathique 4 questionnaire; FDI, functional disability index; PSQI, Pittsburgh sleep quality index; MDT, mechanical detection threshold; PPT, pressure pain threshold; HPT, heat pain threshold; HTT, heat tolerance threshold; CPT, cold pressor task; CPM, conditioned pain modulation; TSP, temporal summation of pain

TABLE 4.

Differences between clusters regarding psychosocial and psychophysical characteristics

Variable Adaptive cluster (n = 89) Pain‐sensitive cluster (n = 71) High somatic symptoms cluster (n = 38) χ2 value p‐value
Descriptors of pain used, Mean (%) ± SD
Sensory 14.98 ± 7.77 c 13.64 ± 7.18 c 33.21 ± 13.3 a , b 63.13 † <0.001
Affective 4.96 ± 6.79 c 5.27 ± 7.27 c 24.92 ± 15.11 a , b 59.01 † <0.001
Evaluative 27.06 ± 15.49 c 30.09 ± 18.70 c 58.16 ± 21.47 a , b 49.83 † <0.001
Temporal 21.66 ± 13.14 c 21.68 ± 10.27 c 33.53 ± 16.87 a , b 19.18 † <0.001
Neuropathic component, n (%)
Mean score of DN4 questionnaire, Mean ±SD 2.24 ± 1.84 b , c 1.59 ± 1.55 a , c 4.61 ± 2.03 a , b 50.82 † <0.001
Likely neuropathic 18 (20.2) 10 (14.1) 27 (71.1) 44.64* <0.001
Not likely neuropathic 71 (79.8) 61 (85.9) 11 (28.9)
Functional Disability, n (%)
Mean score of FDI, Mean ±SD 12.82 ± 8.58 c 11.52 ± 6.61 c 28.72 ± 8.91 a , b 63.06 † <0.001
None or minimal 48 (53.9) 40 (56.3) 1 (2.6) 94.12* <0.001
Mild 23 (25.8) 21 (29.6) 6 (15.8)
Moderate 16 (18.0) 9 (12.7) 12 (31.6)
Severe 2 (2.2) 0 19 (50.0)
Anxiety and Depression Symptoms, n (%)
Mean T‐score of RCADS, Mean ±SD 41.52 ± 9.95 c 44.30 ± 10.84 c 56.24 ± 14.17 a , b 28.36 † <0.001
Below clinical threshold 88 (98.9) 68 (95.8) 27 (71.1) 32.21* <0.001
Borderline 1 (1.1) 1(1.4) 3 (7.9)
Above clinical threshold 0 2 (2.8) 8 (21.1)
Sleep Quality, n (%)
Mean global score of PSQI, Mean ±SD 6.70 ± 3.25 c 5.68 ± 2.51 c 10.07 ± 3.79 a , b 32.50 † <0.001
Good sleep quality 26 (29.2) 25 (35.2) 3 (7.9) 9.83* 0.007
Poor sleep quality 63 (70.8) 46 (64.8) 35 (92.1)
MDT (g), Mean ± SD
Control area 0.42 ± 0.58 0.75 ± 2.66 0.31 ± 0.27 0.45 † 0.800
Affected area 0.56 ± 0.95 0.53 ± 1.14 5.34 ± 28.04 1.38 † 0.501
PPT (N), Mean ± SD
Control area 35.58 ± 14.05 b , c 20.75 ± 9.53 a 21.62 ± 15.68 a 53.12 † <0.001
Affected area 35.18 ± 19.41 b , c 19.94 ± 10.49 a 17.62 ± 12.92 a 43.36 † <0.001
HPT (°C), Mean ± SD 41.04 ± 2.83 b , c 37.74 ± 2.47 a 37.80 ± 2.92 a 51.18 † <0.001
HTT (°C), Mean ± SD 46.82 ± 1.30 b , c 43.75 ± 2.13 a 43.9 ± 2.48 a 94.47 † <0.001
CPT average pain score NRS (0–10), Mean ± SD 5.68 ± 2.28 b , c 8.08 ± 1.73 a 7.89 ± 1.85 a 52.13 † <0.001
CPM, n (%)
CPM efficiency (%), Mean ±SD −38.37 ± 33.00 −18.73 ± 54.40 −28.53 ± 34.40 3.80 † 0.149
Inefficient 17 (19.1) 23 (32.4) 11 (28.9) 4.14* 0.387
Suboptimal 21 (23.6) 16 (22.5) 8 (21.1)
Optimal 51 (57.3) 32 (45.1) 19 (50.0)
TSP, n (%)
TSP pain score NRS (0–10), Mean ±SD 0.68 ± 1.99 b −0.69 ± 2.02 a 0.17 ± 1.94 15.31 † <0.001
No presence 70 (78.7) 66 (93.0) 35 (92.1) 10.43* 0.005
Presence 19 (21.3) 5 (7.0) 3 (7.9)
Open in a new tab

p‐values ≤0.05 are bolded.

Abbreviations: CPM, conditioned pain modulation; CPT, cold pressor task; DN4, douleur neuropathique 4 questionnaire; FDI, functional disability index; HPT, heat pain threshold; HTT, heat tolerance threshold; MDT, mechanical detection threshold; PPT, pressure pain threshold; PSQI, Pittsburgh sleep quality index; RCADS, revised children's anxiety and depression scale; SD, standard deviation; TSP, temporal summation of pain.

a

Significant difference with cluster 1

b

Significant difference with cluster 2.

c

Significant difference with cluster 3.

*

Chi‐squared test statistic.

†

Kruskal‐Wallis 1‐way analysis of variance chi‐squared test statistic.

4. DISCUSSION

The objective of this study was to identify specific psychophysical and psychosocial profiles among a cohort of paediatric patients with chronic back pain. A cluster analysis of these patients suggested three subgroups and were best described by two principal components representing the dimensions of psychosocial factors, and pressure pain and heat tolerance thresholds. Cluster membership did not vary significantly by age, sex, ethnicity, duration of pain, duration of painful episodes, pathology or most painful location as observed by Schurman et al. (2008) in a cluster analysis of children with recurrent abdominal pain (Schurman et al., 2008). Furthermore, no difference in tactile sensitivity or efficiency of their descending inhibitory pain response was observed among the groups. To our knowledge, this is the first cluster analysis performed with youth experiencing chronic back pain. Furthermore, our cluster model included QST results and psychosocial factors, building on prior work by Rabey et al. (2015) and Baron et al. (2017) who included only QST results in their cluster analysis (Baron et al., 2017; Rabey et al., 2015), and adult and paediatric studies who based their analysis on pain descriptors and psychological symptoms (Larsson et al., 2017; Scharff et al., 2005; Schurman et al., 2008; Wager et al., 2014). Moreover, we conducted an unsupervised approach to cluster analysis, unlike Bair et al. (2016) who conducted a supervised cluster analysis to determine risk factors for temporomandibular disorder in healthy individuals. Our unsupervised approach was appropriate for the cross‐sectional design of the study to identify clusters of patients with chronic back pain that may benefit from a tailored management based on their psychophysical and psychosocial profiles.

4.1. Profiling of clusters

The adaptive cluster represented 44.9% of the patients and was characterized by a higher thermal and pressure pain threshold. Subgroups of adult patients with chronic back pain presenting with similar psychophysical characteristics have been identified (Coronado et al., 2014; Rabey et al., 2015). However, the meaning of the low pressure and heat sensitivity in our cohort remains unclear. The results of sensory testing of the patients in this cluster are visually similar to reference values established in the hand and foot in the paediatric population (Blankenburg et al., 2010). This is unlike other paediatric population‐based studies that show lower pressure pain thresholds in adolescents with chronic pain (Tham et al., 2016). However, Tham et al. (2016) has shown in a large cohort of adolescents that the heat pain threshold and cold pressor data were not significantly different between those with and without chronic pain (Tham et al., 2016). Contrarily to Tham et al., Sethna et al. (2007) observed in paediatric patients with complex regional pain syndromes, an overall significant difference with healthy controls for cold and heat pain thresholds. However, a large percentage of patients were within normal reference intervals (Sethna et al., 2007). Our results highlight that despite the presence of chronic back pain, there is a subgroup of patients that do not display deep tissue sensitivity or thermal hyperalgesia in either the affected or control area of the body.

A larger proportion of patients that displayed the presence of thermal temporal summation of pain were found in the adaptive cluster. In a systematic review in children with chronic pain conducted by Pas et al. (2018), central hyperexcitability was shown to be present in several paediatric chronic pain conditions (Pas et al., 2018). Therefore, the sensitization to a tonic noxious heat stimulation in a region of the body remote from the primary area of pain may suggest that the chronic pain of the patients in the adaptive cluster arise or persist from central processes (Giesecke et al., 2004). In a systematic review on adult patients conducted by Hubscher et al. (2013), a fair association between spinal pain intensity and thermal temporal summation was observed (Hubscher et al., 2013). Although we did not conduct this analysis in our cohort, altogether, our results may suggest that the persistent back pain in the adaptive cluster may arise from central facilitation.

The pain‐sensitive cluster, representing 35.9% of the cohort was characterized to have lower thermal and pressure pain thresholds in comparison to the adaptive cluster. Lower pain thresholds in the affected region of the body have been observed in other chronic pain conditions in paediatrics (Cornelissen et al., 2014; Tham et al., 2016). We recently observed (Teles et al., 2019) in a subset of the cohort with idiopathic scoliosis with chronic back pain that the severity of their curve was significantly associated with deep tissue sensitivity in the back (Teles et al., 2019). Therefore, the diagnosis that may underlie that chronic back pain should not be ignored in this subgroup. Studies investigating strictly psychophysical profiles of adult chronic pain patients observe minimally a three‐group solution (Baron et al., 2012; Rabey et al., 2015), unlike our results revealing two psychophysical profiles. However, our results highlight that, in contrast to the adaptive cluster, there is a subgroup of patients that display maladaptive pain mechanisms suggesting possible involvement of central and peripheral pain mechanisms that can be targeted.

The adaptive cluster and the pain‐sensitive cluster were characterized to have lower scores for all questionnaires (i.e. use less descriptors of pain, not likely to report their pain as neuropathic in nature, none to mild functional disability, report less anxiety and depression symptoms below the clinical threshold and report better sleep quality). Similarly, to other studies conducting cluster analysis of psychological profiles among children with chronic pain, at least two subgroups can be observed (Scharff et al., 2005; Schurman et al., 2008). Scharff et al. (2005) observed in a subgroup of children with chronic pain (52.1%) whose questionnaire scores fell within established population norms and was distinguished by low levels of disability (Scharff et al., 2005). Schurman et al. (2008) conducted a similar cluster analysis and also observed more than half of their sample with better psychological functioning (Schurman et al., 2008). Therefore, our results are consistent with the chronic pain model where inter‐individual variability in the relative contributions of multiple elements of pain would be expected.

Despite their low thermal and pressure pain threshold similar to the pain‐sensitive cluster, the high somatic symptoms cluster, representing 19.2% of patients, displayed higher self‐report of pain intensity in the back, functional disability, anxiety and depression symptoms, and poor sleep quality. This is as observed by other research groups investigating variations in psychosocial profiles in children and adolescents with chronic pain (Scharff et al., 2005; Schurman et al., 2008). Functional disability, mental distress and sleeps problems have been shown to be associated with pain in the paediatric population (Eccleston et al., 2004; Lewandowski Holley et al., 2017; Long et al., 2008; Wojtowicz & Banez, 2015). However, the cause‐and‐effect relationship between pain and these outcomes is unclear. Furthermore, studies investigating strictly psychosocial subgroups of paediatric chronic pain patients observe minimally a three‐group solution (Scharff et al., 2005; Schurman et al., 2008; Wager et al., 2014), unlike our results revealing two psychosocial profiles. Therefore, future directions may include separate cluster analyses on psychophysical and psychosocial profiles to reveal more subgroups masked by our current cluster approach.

A higher proportion of patients in the high somatic symptoms cluster reported their back pain radiating down their leg and reported their back pain to display neuropathic‐like characteristics. Neuropathic pain, usually viewed only as to be a result of lesions affecting the somatosensory system, has also been shown to be triggered in parallel by psychological factors. In 2015, Dimova et al. demonstrated that healthy adults who displayed a pessimistic life attitude also displayed neuropathic‐like pain patterns after topical capsaicin application (Dimova et al., 2015). Therefore, it is hypothesized that the high proportion of patients reporting a neuropathic‐like component for their back pain in the high somatic symptoms cluster may be explained by a high tendency of the patients to focus on their pain‐related bodily sensations. However, along with reporting neuropathic‐like characteristics, patients in the high somatic symptoms cluster displayed similar thermal and pressure pain thresholds to the pain‐sensitive cluster, suggesting possible involvement of central and peripheral pain mechanisms. Without the presence of a lesion in the somatosensory system, it may be hypothesized that nociplastic pain may act as the dominant pain mechanism in this cluster of patients such that nociceptive and neuropathic pain are not entirely responsible for the pain (Kosek et al., 2021). Nociplastic pain is defined as “pain that arises from altered nociception despite no clear evidence of actual or threatened tissue damage causing the activation of peripheral nociceptors or evidence for disease or lesion of the somatosensory system causing the pain” (Kosek et al., 2016). Recently, clinically useful criteria for nociplastic pain were established such that chronic nociplastic pain was defined as: (1) pain duration >3 months, (2) a regional rather than discrete distribution, (3) not entirely explained by nociceptive or neuropathic pain mechanisms and (4) displaying clinical signs of pain hypersensitivity in the region of pain. The presence of a history of pain hypersensitivity in the region of pain and defined co‐morbidities (e.g. sleep disturbance and cognitive problems) strengthen the probability of nociplastic pain (Kosek et al., 2021). Some patients in the high somatic symptoms cluster meet the requirements of chronic nociplastic pain such that they may report regional pain distribution (i.e. variable pain intensity across the back), report pain that cannot entirely be explained by nociceptive or neuropathic mechanisms, show clinical signs of pain hypersensitivity (i.e. low thermal and pressure thresholds) and psychosocial co‐morbidities (Nijs et al., 2021).

4.2. Clinical implications

The management and treatment of chronic back pain may remain a challenge. Current back pain guidelines highlight multidisciplinary management using a biopsychosocial model as the standard of care. A comprehensive use of exercises, physical therapy, cognitive behavioural therapy, and medical treatments with a active commitment of the patients and parents are associated with positive clinical outcomes (Randall et al., 2018; Simons et al., 2018). Studies investigating quantitative sensory testing and psychosocial factors in relation to musculoskeletal pain have shown the importance of a multidimensional assessment (Georgopoulos et al., 2019; Holbech et al., 2016; Hwang et al., 2017; Teles et al., 2019; Tham et al., 2016). Georgopoulos et al. (2019) highlight that the baseline assessment with quantitative sensory testing was a valuable instrument to predict clinical outcomes including disability in patients with musculoskeletal pain. Improving the diagnostic process by identifying ‘clusters’ of patients with chronic back pain based on results of quantitative sensory testing, pain‐related outcomes and psychosocial factors may help clinicians provide an improved individualized care to patients (Vega et al., 2018).

Exercises, physical therapy and psychological therapies are aimed to focus on helping patients return to their desired level of functioning through progressive engagement in previously avoided activities and a self‐management approach to pain (Simons et al., 2018; Vega et al., 2018). Studies targeting the central pain processes have used physical activity to reduce the presence of temporal summation pain (Bishop et al., 2011; Pack et al., 2018). Therefore, the patients belonging to an adaptive cluster who display temporal summation of pain, possibly arising from central facilitation, may benefit from a multidimensional programme centred on physical activity (Mirek et al., 2019).

Psychological therapies, delivered individually or in groups in the paediatric chronic pain population, have been shown to reduce pain symptoms, disability and negative affect, but also modify social environmental factors to enhance functional status (Fisher et al., 2018). Hence, a multicomponent approach focused on psychological therapeutic interventions addressing anxiety, depression and poor sleep quality and on the probable pain hypersensitivity may be more beneficial for patients that are grouped in the high somatic symptoms cluster who display more functional disability, mental distress and sleeps problems.

Pharmacological treatments and interventional procedures are mainly supported through studies conducted in adults. Clinical trials in adults suggested that sodium channel modulators such as local anaesthetics could be useful to treat pain conditions associated with peripheral sensitization (Demant et al., 2014; Mainka et al., 2016). Moreover, patients with potential involvement of central pain processes could benefit more from gabapentinoids, inhibiting central neuronal sensitization (Granovsky & Yarnitsky, 2013). Therefore, patients belonging to the pain‐sensitive cluster with possible involvement of central and peripheral pain mechanisms may benefit from a multidimensional program centred on pharmacological or interventional strategies.

4.3. Limitations and conclusion

There are certain limitations to this study that should be explicit. First, healthy controls were not tested so it is unknown if all pain‐free children would fall into one cluster, a new cluster or have a variety of pain profiles as highlighted by Bair et al. (2016). Furthermore, the exclusion of healthy controls limits the extent of the involvement of the underlying nociceptive mechanisms in chronic musculoskeletal pain being clinically relevant. However, the objective of the study was to identify and describe profiles of patients to identify potential treatment responders and ultimately lead to personalized treatment. The second limitation was the cross‐sectional nature of the study such that the long‐term stability over weeks or months was not studied in this cohort. Therefore, it is unknown whether patients shift from one cluster to another depending on if a therapeutic intervention was given. Future work, conducting a prospective study that includes healthy controls to determine which psychophysical profile is a risk factor to chronic back pain and/or to determine whether a tailored treatment approach based on the clinical profile of the patient is beneficial, is warranted.

In conclusion, despite different pathologies, this study identified clusters of children and adolescents experiencing chronic back pain based on physical and psychosocial profiles. The assessment of chronic back pain should be comprehensive to assess multiple elements contributing to pain, including pathophysiology, somatosensory functioning, and psychosocial factors to improve multidisciplinary pain management.

Conflict of Interest

The Quebec Pain Research Network financially supported this study. This research analysis was also supported by an Edwards PhD Studentship in Pain Research from the Louise and Alan Edwards Foundation awarded to D.D. Ocay. The authors declare no conflict of interest related to this work.

Author Contributions

All authors contributed to data analysis, drafting or revising the article, gave final approval of the version to be published and agree to be accountable for all aspects of the work.

ACKNOWLEDGEMENTS

The authors would like to thank the participants, Ms. Sheila Bote, Ms. Dee‐Anne Naylor, Ms. My‐Linh Ma, Ms. Diana‐Luk Ye, and all the clinical staff of the Shriners Hospitals for Children, Canada for their precious collaboration.

Ocay, D. D. , Loewen, A. , Premachandran, S. , Ingelmo, P. M. , Saran, N. , Ouellet, J. A. , & Ferland, C. E. (2022). Psychosocial and psychophysical assessment in paediatric patients and young adults with chronic back pain: A cluster analysis. European Journal of Pain, 26, 855–872. 10.1002/ejp.1912

DATA AVAILABILITY STATEMENT

The data and code used to support the findings of this study are available from the corresponding author upon request.

REFERENCES

  1. Altaf, F. , Heran, M. K. , & Wilson, L. F. (2014). Back pain in children and adolescents. The Bone & Joint Journal, 717–723. 10.1302/0301-620X.96B6.33075 [DOI] [PubMed] [Google Scholar]
  2. Bair, E. , Gaynor, S. , Slade, G. D. , Ohrbach, R. , Fillingim, R. B. , Greenspan, J. D. , Dubner, R. , Smith, S. B. , Diatchenko, L. , & Maixner, W. (2016). Identification of clusters of individuals relevant to temporomandibular disorders and other chronic pain conditions: the OPPERA study. Pain, 157, 1266–1278. 10.1097/j.pain.0000000000000518 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Balagué, F. , Skovron, M. L. , Nordin, M. , Dutoit, G. , Pol, L. R. , & Waldburger, M. (1995). Low back pain in schoolchildren. A study of familial and psychological factors. Spine, 20(11), 1265–1270. 10.1097/00007632-199506000-00012 [DOI] [PubMed] [Google Scholar]
  4. Balague, F. , Troussier, B. , & Salminen, J. J. (1999). Non‐specific low back pain in children and adolescents: risk factors. European Spine Journal, 8, 429–438. 10.1007/s005860050201 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Baron, R. , Forster, M. , & Binder, A. (2012). Subgrouping of patients with neuropathic pain according to pain‐related sensory abnormalities: A first step to a stratified treatment approach. The Lancet Neurology, 11, 999–1005. 10.1016/S1474-4422(12)70189-8 [DOI] [PubMed] [Google Scholar]
  6. Baron, R. , Maier, C. , Attal, N. , Binder, A. , Bouhassira, D. , Cruccu, G. , Finnerup, N. B. , Haanpaa, M. , Hansson, P. , Hullemann, P. , Jensen, T. S. , Freynhagen, R. , Kennedy, J. D. , Magerl, W. , Mainka, T. , Reimer, M. , Rice, A. S. , Segerdahl, M. , Serra, J. , … Treede, R. D. (2017). Peripheral neuropathic pain: a mechanism‐related organizing principle based on sensory profiles. Pain, 158, 261–272. 10.1097/j.pain.0000000000000753 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bishop, M. D. , Beneciuk, J. M. , & George, S. Z. (2011). Immediate reduction in temporal sensory summation after thoracic spinal manipulation. The Spine Journal, 11, 440–446. 10.1016/j.spinee.2011.03.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Blankenburg, M. , Boekens, H. , Hechler, T. , Maier, C. , Krumova, E. , Scherens, A. , Magerl, W. , Aksu, F. , & Zernikow, B. (2010). Reference values for quantitative sensory testing in children and adolescents: developmental and gender differences of somatosensory perception. Pain, 149, 76–88. 10.1016/j.pain.2010.01.011 [DOI] [PubMed] [Google Scholar]
  9. Bouhassira, D. , Attal, N. , Alchaar, H. , Boureau, F. , Brochet, B. , Bruxelle, J. , Cunin, G. , Fermanian, J. , Ginies, P. , Grun‐Overdyking, A. , Jafari‐Schluep, H. , Lanteri‐Minet, M. , Laurent, B. , Mick, G. , Serrie, A. , Valade, D. , & Vicaut, E. (2005). Comparison of pain syndromes associated with nervous or somatic lesions and development of a new neuropathic pain diagnostic questionnaire (DN4). Pain, 114, 29–36. 10.1016/j.pain.2004.12.010 [DOI] [PubMed] [Google Scholar]
  10. Brattberg, G. (2004). Do pain problems in young school children persist into early adulthood? A 13‐year follow‐up. European Journal of Pain, 8, 187–199. 10.1016/j.ejpain.2003.08.001 [DOI] [PubMed] [Google Scholar]
  11. Buysse, D. J. , Reynolds, C. F., 3rd , Monk, T. H. , Berman, S. R. , & Kupfer, D. J. (1989). The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Research, 28, 193–213. 10.1016/0165-1781(89)90047-4 [DOI] [PubMed] [Google Scholar]
  12. Chorpita, B. F. , Moffitt, C. E. , & Gray, J. (2005). Psychometric properties of the revised child anxiety and depression scale in a clinical sample. Behavior Research and Therapy, 43, 309–322. 10.1016/j.brat.2004.02.004 [DOI] [PubMed] [Google Scholar]
  13. Chorpita, B. F. , Yim, L. , Moffitt, C. , Umemoto, L. A. , & Francis, S. E. (2000). Assessment of symptoms of DSM‐IV anxiety and depression in children: a revised child anxiety and depression scale. Behavior Research and Therapy, 38, 835–855. 10.1016/S0005-7967(99)00130-8 [DOI] [PubMed] [Google Scholar]
  14. Claar, R. L. , & Walker, L. S. (2006). Functional assessment of pediatric pain patients: Psychometric properties of the functional disability inventory. Pain, 121, 77–84. 10.1016/j.pain.2005.12.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Cornelissen, L. , Donado, C. , Kim, J. , Chiel, L. , Zurakowski, D. , Logan, D. E. , Meier, P. , Sethna, N. F. , Blankenburg, M. , Zernikow, B. , Sundel, R. P. , & Berde, C. B. (2014). Pain hypersensitivity in juvenile idiopathic arthritis: a quantitative sensory testing study. Pediatric Rheumatology, 12, 39. 10.1186/1546-0096-12-39 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Coronado, R. A. , Bialosky, J. E. , Robinson, M. E. , & George, S. Z. (2014). Pain sensitivity subgroups in individuals with spine pain: potential relevance to short‐term clinical outcome. Physical Therapy, 94, 1111–1122. 10.2522/ptj.20130372 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. David, R. , Pontone, S. , Dugué, S. , Delivet, H. , Ilharreborde, B. , Mazda, K. , Nivoche, Y. , & Dahmani, S. (2015). Facteurs prédictifs de douleurs neuropathiques postopératoires après chirurgie de scoliose en pédiatrie. Anesthésie & Réanimation, 1, A128–A129. 10.1016/j.anrea.2015.07.198 [DOI] [Google Scholar]
  18. Davis, P. J. , & Williams, H. J. (2008). The investigation and management of back pain in children. Archives of Disease in Childhood ‐ Education and Practice, 93, 73–83. 10.1136/adc.2006.115535 [DOI] [PubMed] [Google Scholar]
  19. de Leeuw, T. G. , der Zanden, T. V. , Ravera, S. , Felisi, M. , Bonifazi, D. , Tibboel, D. , Ceci, A. , Kaguelidou, F. , & de Wildt, S. N. (2020). Diagnosis and treatment of chronic neuropathic and mixed pain in children and adolescents: Results of a survey study amongst practitioners. Children, 7(11), 208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Demant, D. T. , Lund, K. , Vollert, J. , Maier, C. , Segerdahl, M. , Finnerup, N. B. , Jensen, T. S. , & Sindrup, S. H. (2014). The effect of oxcarbazepine in peripheral neuropathic pain depends on pain phenotype: A randomised, double‐blind, placebo‐controlled phenotype‐stratified study. Pain, 155, 2263–2273. 10.1016/j.pain.2014.08.014 [DOI] [PubMed] [Google Scholar]
  21. Diatchenko, L. , Nackley, A. G. , Slade, G. D. , Fillingim, R. B. , & Maixner, W. (2006). Idiopathic pain disorders–pathways of vulnerability. Pain, 123, 226–230. 10.1016/j.pain.2006.04.015 [DOI] [PubMed] [Google Scholar]
  22. Dimova, V. , Oertel, B. G. , Kabakci, G. , Zimmermann, M. , Hermens, H. , Lautenbacher, S. , Ultsch, A. , & Lotsch, J. (2015). A more pessimistic life orientation is associated with experimental inducibility of a neuropathy‐like pain pattern in healthy individuals. The Journal of Pain, 16, 791–800. [DOI] [PubMed] [Google Scholar]
  23. Eccleston, C. , Crombez, G. , Scotford, A. , Clinch, J. , & Connell, H. (2004). Adolescent chronic pain: Patterns and predictors of emotional distress in adolescents with chronic pain and their parents. Pain, 108, 221–229. 10.1016/j.pain.2003.11.008 [DOI] [PubMed] [Google Scholar]
  24. Farrar, J. T. , Young, J. P., Jr , LaMoreaux, L. , Werth, J. L. , & Poole, R. M. (2001). Clinical importance of changes in chronic pain intensity measured on an 11‐point numerical pain rating scale. Pain, 94, 149–158. 10.1016/S0304-3959(01)00349-9 [DOI] [PubMed] [Google Scholar]
  25. Ferland, C. E. , Teles, A. R. , Ingelmo, P. , Saran, N. , Marchand, S. , & Ouellet, J. A. (2018a). Blood monoamines as potential biomarkers for conditioned pain modulation efficacy: An exploratory study in paediatrics. European Journal of Pain, 23(2), 327–340. 10.1002/ejp.1307 [DOI] [PubMed] [Google Scholar]
  26. Ferland, C. E. , Villemure, C. , Michon, P. E. , Gandhi, W. , Ma, M. L. , Chouchou, F. , Parent, A. J. , Bushnell, M. C. , Lavigne, G. , Rainville, P. , Ware, M. A. , Jackson, P. L. , Schweinhardt, P. , & Marchand, S. (2018b). Multi‐center assessment of quantitative sensory testing (qst) for the detection of neuropathic‐like pain responses using the topical capsaicin model. Canadian Journal of Pain. 10.1080/24740527.2018.1525682 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Fernandes, A. M. , De Campos, C. , Batalha, L. , Perdigao, A. , & Jacob, E. (2014). Pain assessment using the adolescent pediatric pain tool: A systematic review. Pain Research and Management, 19, 212–218. 10.1155/2014/979416 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Fisher, E. , Law, E. , Dudeney, J. , Palermo, T. M. , Stewart, G. , & Eccleston, C. (2018). Psychological therapies for the management of chronic and recurrent pain in children and adolescents. Cochrane Database Systematic Review, 9, Cd003968. 10.1002/14651858.CD003968.pub5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Georgopoulos, V. , Akin‐Akinyosoye, K. , Zhang, W. , McWilliams, D. F. , Hendrick, P. , & Walsh, D. A. (2019). Quantitative sensory testing and predicting outcomes for musculoskeletal pain, disability, and negative affect: a systematic review and meta‐analysis. Pain, 160(9), 1920–1932. 10.1097/j.pain.0000000000001590 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Giesecke, T. , Gracely, R. H. , Grant, M. A. , Nachemson, A. , Petzke, F. , Williams, D. A. , & Clauw, D. J. (2004). Evidence of augmented central pain processing in idiopathic chronic low back pain. Arthritis and Rheumatism, 50, 613–623. 10.1002/art.20063 [DOI] [PubMed] [Google Scholar]
  31. Granovsky, Y. , & Yarnitsky, D. (2013). Personalized pain medicine: The clinical value of psychophysical assessment of pain modulation profile. Rambam Maimonides Medical Journal, 4, e0024. 10.5041/RMMJ.10131 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Haidar, R. , Saad, S. , Khoury, N. J. , & Musharrafieh, U. (2011). Practical approach to the child presenting with back pain. European Journal of Pediatrics, 170, 149–156. 10.1007/s00431-010-1220-9 [DOI] [PubMed] [Google Scholar]
  33. Hair, J. F. (2010). Multivariate data analysis: a global perspective. Pearson Education. [Google Scholar]
  34. Hestbaek, L. , Leboeuf‐Yde, C. , Kyvik, K. O. , & Manniche, C. (1976). The course of low back pain from adolescence to adulthood: Eight‐year follow‐up of 9600 twins. Spine, 31(4), 468–472. 10.1097/01.brs.0000199958.04073.d9 [DOI] [PubMed] [Google Scholar]
  35. Hirschfeld, G. , Zernikow, B. , Kraemer, N. , Hechler, T. , Aksu, F. , Krumova, E. , Maier, C. , Magerl, W. , & Blankenburg, M. (2012). Development of somatosensory perception in children: a longitudinal QST‐study. Neuropediatrics, 43, 10–16. 10.1055/s-0032-1307450 [DOI] [PubMed] [Google Scholar]
  36. Holbech, J. V. , Bach, F. W. , Finnerup, N. B. , Jensen, T. S. , & Sindrup, S. H. (2016). Pain phenotype as a predictor for drug response in painful polyneuropathy‐a retrospective analysis of data from controlled clinical trials. Pain, 157, 1305–1313. 10.1097/j.pain.0000000000000563 [DOI] [PubMed] [Google Scholar]
  37. Hubscher, M. , Moloney, N. , Leaver, A. , Rebbeck, T. , McAuley, J. H. , & Refshauge, K. M. (2013). Relationship between quantitative sensory testing and pain or disability in people with spinal pain‐a systematic review and meta‐analysis. Pain, 154, 1497–1504. 10.1016/j.pain.2013.05.031 [DOI] [PubMed] [Google Scholar]
  38. Huguet, A. , & Miro, J. (2008). The severity of chronic pediatric pain: An epidemiological study. The Journal of Pain, 9, 226–236. 10.1016/j.jpain.2007.10.015 [DOI] [PubMed] [Google Scholar]
  39. Hwang, P. S. , Ma, M. L. , Spiegelberg, N. , & Ferland, C. E. (2017). Current methodological approaches in conditioned pain modulation assessment in pediatrics. Journal of Pain Research, 10, 2797–2802. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Jacob, E. , Mack, A. K. , Savedra, M. , Van Cleve, L. , & Wilkie, D. J. (2014). Adolescent pediatric pain tool for multidimensional measurement of pain in children and adolescents. Pain Management Nursing, 15, 694–706. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Jeffries, L. J. , Milanese, S. F. , & Grimmer‐Somers, K. A. (2007). Epidemiology of adolescent spinal pain: A systematic overview of the research literature. Spine, 32(23), 2630–2637. 10.1097/BRS.0b013e318158d70b [DOI] [PubMed] [Google Scholar]
  42. Josse, J. , & Husson, F. (2016). missMDA: A package for handling missing values in multivariate data analysis. Journal of Statistical Software, 70. [Google Scholar]
  43. Kosek, E. , Clauw, D. , Nijs, J. , Baron, R. , Gilron, I. , Harris, R. E. , Mico, J. A. , Rice, A. S. , & Sterling, M. (2021). Chronic nociplastic pain affecting the musculoskeletal system: Clinical criteria and grading system. Pain, 162(11), 2629–2634. 10.1097/j.pain.0000000000002324 [DOI] [PubMed] [Google Scholar]
  44. Kosek, E. , Cohen, M. , Baron, R. , Gebhart, G. F. , Mico, J. A. , Rice, A. S. C. , Rief, W. , & Sluka, A. K. (2016). Do we need a third mechanistic descriptor for chronic pain states? Pain, 157, 1382–1386. 10.1097/j.pain.0000000000000507 [DOI] [PubMed] [Google Scholar]
  45. Larche, C. L. , Plante, I. , Roy, M. , Ingelmo, P. M. , & Ferland, C. E. (2021). The pittsburgh sleep quality index: Reliability, factor structure, and related clinical factors among children, adolescents, and young adults with chronic pain. Sleep Disorders, 2021, 5546484. 10.1155/2021/5546484 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Larsson, B. , Gerdle, B. , Bernfort, L. , Levin, L. A. , & Dragioti, E. (2017). Distinctive subgroups derived by cluster analysis based on pain and psychological symptoms in Swedish older adults with chronic pain ‐ a population study (PainS65+). BMC Geriatrics, 17, 200. 10.1186/s12877-017-0591-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Le, S. , Josse, J. , & Husson, F. (2008). FactoMineR: An R package for multivariate analysis. Journal of Statistical Software, 25, 1–18. [Google Scholar]
  48. Lewandowski Holley, A. , Wilson, A. C. , Cho, E. , & Palermo, T. M. (2017). Clinical phenotyping of youth with new‐onset musculoskeletal pain: A controlled cohort study. Clinical Journal of Pain, 33, 28–36. 10.1097/AJP.0000000000000371 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Long, A. C. , Krishnamurthy, V. , & Palermo, T. M. (2008). Sleep disturbances in school‐age children with chronic pain. Journal of Pediatric Psychology, 33, 258–268. 10.1093/jpepsy/jsm129 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Mainka, T. , Malewicz, N. M. , Baron, R. , Enax‐Krumova, E. K. , Treede, R. D. , & Maier, C. (2016). Presence of hyperalgesia predicts analgesic efficacy of topically applied capsaicin 8% in patients with peripheral neuropathic pain. European Journal of Pain, 20, 116–129. 10.1002/ejp.703 [DOI] [PubMed] [Google Scholar]
  51. Mathieson, S. , Maher, C. G. , Terwee, C. B. , Folly de Campos, T. , & Lin, C. W. (2015). Neuropathic pain screening questionnaires have limited measurement properties. A systematic review. Journal of Clinical Epidemiology, 68, 957–966. 10.1016/j.jclinepi.2015.03.010 [DOI] [PubMed] [Google Scholar]
  52. McGrath, P. J. , Walco, G. A. , Turk, D. C. , Dworkin, R. H. , Brown, M. T. , Davidson, K. , Eccleston, C. , Finley, G. A. , Goldschneider, K. , Haverkos, L. , Hertz, S. H. , Ljungman, G. , Palermo, T. , Rappaport, B. A. , Rhodes, T. , Schechter, N. , Scott, J. , Sethna, N. , Svensson, O. K. , … Zeltzer, L. (2008). Core outcome domains and measures for pediatric acute and chronic/recurrent pain clinical trials: PedIMMPACT recommendations. The Journal of Pain, 9, 771–783. 10.1016/j.jpain.2008.04.007 [DOI] [PubMed] [Google Scholar]
  53. Mikkelsson, M. , El‐Metwally, A. , Kautiainen, H. , Auvinen, A. , Macfarlane, G. J. , & Salminen, J. J. (2008). Onset, prognosis and risk factors for widespread pain in schoolchildren: A prospective 4‐year follow‐up study. Pain, 138, 681–687. 10.1016/j.pain.2008.06.005 [DOI] [PubMed] [Google Scholar]
  54. Mirek, E. , Logan, D. , Boullard, K. , Hall, A. M. , Staffa, S. J. , & Sethna, N. (2019). Physical therapy outcome measures for assessment of lower extremity chronic pain‐related function in pediatrics. Pediatric Physical Therapy, 31, 200–207. 10.1097/PEP.0000000000000587 [DOI] [PubMed] [Google Scholar]
  55. Mollayeva, T. , Thurairajah, P. , Burton, K. , Mollayeva, S. , Shapiro, C. M. , & Colantonio, A. (2016). The Pittsburgh sleep quality index as a screening tool for sleep dysfunction in clinical and non‐clinical samples: A systematic review and meta‐analysis. Sleep Medicine Reviews, 25, 52–73. 10.1016/j.smrv.2015.01.009 [DOI] [PubMed] [Google Scholar]
  56. Moreno, M. A. (2017). Low back pain in children and adolescents. JAMA Pediatrics, 171, 312. 10.1001/jamapediatrics.2016.3091 [DOI] [PubMed] [Google Scholar]
  57. Nijs, J. , Lahousse, A. , Kapreli, E. , Bilika, P. , Saracoglu, I. , Malfliet, A. , Coppieters, I. , De Baets, L. , Leysen, L. , Roose, E. , Clark, J. , Voogt, L. , & Huysmans, E. (2021). Nociplastic pain criteria or recognition of central sensitization? Pain phenotyping in the past, present and future. Journal of Clinical Medicine, 10(15), 3203. 10.3390/jcm10153203 [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. O'Sullivan, P. , Beales, D. , Jensen, L. , Murray, K. , & Myers, T. (2011). Characteristics of chronic non‐specific musculoskeletal pain in children and adolescents attending a rheumatology outpatients clinic: a cross‐sectional study. Pediatric Rheumatology, 9, 3. 10.1186/1546-0096-9-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Pack, R. , Gilliland, R. , & Mecham, A. (2018). The treatment of central sensitization in an adolescent using pain neuroscience education and graded exposure to activity: A case report. Physiotherapy Theory and Practice, 36(10), 1164–1174. 10.1080/09593985.2018.1551454 [DOI] [PubMed] [Google Scholar]
  60. Palermo, T. M. (2009). Assessment of chronic pain in children: current status and emerging topics. Pain Research and Management, 14(1), 21–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Pas, R. , Ickmans, K. , Van Oosterwijck, S. , Van der Cruyssen, K. , Foubert, A. , Leysen, L. , Nijs, J. , & Meeus, M. (2018). Hyperexcitability of the central nervous system in children with chronic pain: A systematic review. Pain Med, 19, 2504–2514. 10.1093/pm/pnx320 [DOI] [PubMed] [Google Scholar]
  62. Potvin, S. , & Marchand, S. (2016). Pain facilitation and pain inhibition during conditioned pain modulation in fibromyalgia and in healthy controls. Pain, 157, 1704–1710. 10.1097/j.pain.0000000000000573 [DOI] [PubMed] [Google Scholar]
  63. Rabey, M. , Slater, H. , O'Sullivan, P. , Beales, D. , & Smith, A. (2015). Somatosensory nociceptive characteristics differentiate subgroups in people with chronic low back pain: a cluster analysis. Pain, 156, 1874–1884. 10.1097/j.pain.0000000000000244 [DOI] [PubMed] [Google Scholar]
  64. Randall, E. T. , Smith, K. R. , Conroy, C. , Smith, A. M. , Sethna, N. , & Logan, D. E. (2018). Back to living: Long‐term functional status of pediatric patients who completed intensive interdisciplinary pain treatment. Clinical Journal of Pain, 34, 890–899. 10.1097/AJP.0000000000000616 [DOI] [PubMed] [Google Scholar]
  65. Raniti, M. B. , Waloszek, J. M. , Schwartz, O. , Allen, N. B. , & Trinder, J. (2018). Factor structure and psychometric properties of the Pittsburgh Sleep Quality Index in community‐based adolescents. Sleep, 41. 10.1093/sleep/zsy066 [DOI] [PubMed] [Google Scholar]
  66. Rolke, R. , Baron, R. , Maier, C. , Tolle, T. R. , Treede, R. D. , Beyer, A. , Binder, A. , Birbaumer, N. , Birklein, F. , Botefur, I. C. , Braune, S. , Flor, H. , Huge, V. , Klug, R. , Landwehrmeyer, G. B. , Magerl, W. , Maihofner, C. , Rolko, C. , Schaub, C. , … Wasserka, B. (2006). Quantitative sensory testing in the German Research Network on Neuropathic Pain (DFNS): standardized protocol and reference values. Pain, 123, 231–243. 10.1016/j.pain.2006.01.041 [DOI] [PubMed] [Google Scholar]
  67. Savedra, M. C. , Holzemer, W. L. , Tesler, M. D. , & Wilkie, D. J. (1993). Assessment of postoperation pain in children and adolescents using the adolescent pediatric pain tool. Nursing Research, 42, 5–9. 10.1097/00006199-199301000-00002 [DOI] [PubMed] [Google Scholar]
  68. Savedra, M. C. , Tesler, M. D. , Holzemer, W. L. , Wilkie, D. J. , & Ward, J. A. (1989). Pain location: validity and reliability of body outline markings by hospitalized children and adolescents. Research in Nursing & Health, 12, 307–314. 10.1002/nur.4770120506 [DOI] [PubMed] [Google Scholar]
  69. Scharff, L. , Langan, N. , Rotter, N. , Scott‐Sutherland, J. , Schenck, C. , Tayor, N. , McDonald‐Nolan, L. , & Masek, B. (2005). Psychological, behavioral, and family characteristics of pediatric patients with chronic pain: a 1‐year retrospective study and cluster analysis. Clinical Journal of Pain, 21, 432–438. 10.1097/01.ajp.0000130160.40974.f5 [DOI] [PubMed] [Google Scholar]
  70. Schurman, J. V. , Danda, C. E. , Friesen, C. A. , Hyman, P. E. , Simon, S. D. , & Cocjin, J. T. (2008). Variations in psychological profile among children with recurrent abdominal pain. Journal of Clinical Psychology in Medical Settings, 15, 241–251. 10.1007/s10880-008-9120-0 [DOI] [PubMed] [Google Scholar]
  71. Sethna, N. F. , Meier, P. M. , Zurakowski, D. , & Berde, C. B. (2007). Cutaneous sensory abnormalities in children and adolescents with complex regional pain syndromes. Pain, 131, 153–161. 10.1016/j.pain.2006.12.028 [DOI] [PubMed] [Google Scholar]
  72. Simons, L. E. , Sieberg, C. B. , Conroy, C. , Randall, E. T. , Shulman, J. , Borsook, D. , Berde, C. , Sethna, N. F. , & Logan, D. E. (2018). Children with chronic pain: Response trajectories after intensive pain rehabilitation treatment. J Pain, 19, 207–218. 10.1016/j.jpain.2017.10.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. Siu, Y. F. , Chan, S. , Wong, K. M. , & Wong, W. S. (2012). The comorbidity of chronic pain and sleep disturbances in a community adolescent sample: Prevalence and association with sociodemographic and psychosocial factors. Pain Medicine, 13, 1292–1303. 10.1111/j.1526-4637.2012.01473.x [DOI] [PubMed] [Google Scholar]
  74. Teles, A. R. , Ocay, D. D. , Bin Shebreen, A. , Tice, A. , Saran, N. , Ouellet, J. A. , & Ferland, C. E. (2019). Evidence of impaired pain modulation in adolescents with idiopathic scoliosis and chronic back pain. Spine Journal, 19, 677–686. 10.1016/j.spinee.2018.10.009 [DOI] [PubMed] [Google Scholar]
  75. Tham, S. W. , Palermo, T. M. , Holley, A. L. , Zhou, C. , Stubhaug, A. , Furberg, A. S. , & Nielsen, C. S. (2016). A population‐based study of quantitative sensory testing in adolescents with and without chronic pain. Pain, 157, 2807–2815. 10.1097/j.pain.0000000000000716 [DOI] [PubMed] [Google Scholar]
  76. Thibault, A. , Forget, R. , & Lambert, J. (1994). Evaluation of cutaneous and proprioceptive sensation in children: A reliability study. Developmental Medicine and Child Neurology, 36, 796–812. 10.1111/j.1469-8749.1994.tb08190.x [DOI] [PubMed] [Google Scholar]
  77. Tousignant‐Laflamme, Y. , Page, S. , Goffaux, P. , & Marchand, S. (2008). An experimental model to measure excitatory and inhibitory pain mechanisms in humans. Brain Research, 1230, 73–79. 10.1016/j.brainres.2008.06.120 [DOI] [PubMed] [Google Scholar]
  78. Treede, R.‐D. , Rief, W. , Barke, A. , Aziz, Q. , Bennett, M. I. , Benoliel, R. , Cohen, M. , Evers, S. , Finnerup, N. B. , First, M. B. , Giamberardino, M. A. , Kaasa, S. , Korwisi, B. , Kosek, E. , Lavand'homme, P. , Nicholas, M. , Perrot, S. , Scholz, J. , Schug, S. , … Wang, S.‐J. (2019). Chronic pain as a symptom or a disease: the IASP classification of chronic pain for the international classification of diseases (ICD‐11). Pain, 160, 19–27. 10.1097/j.pain.0000000000001384 [DOI] [PubMed] [Google Scholar]
  79. Vega, E. , Beaulieu, Y. , Gauvin, R. , Ferland, C. , Stabile, S. , Pitt, R. , Gonzalez Cardenas, V. H. , & Ingelmo, P. M. (2018). Chronic non‐cancer pain in children: we have a problem, but also solutions. Minerva Anestesiologica, 84, 1081–1092. 10.23736/S0375-9393.18.12367-4 [DOI] [PubMed] [Google Scholar]
  80. Wager, J. , Zernikow, B. , Darlington, A. , Vocks, S. , & Hechler, T. (2014). Identifying subgroups of paediatric chronic pain patients: a cluster‐analytic approach. European Journal of Pain, 18, 1352–1362. 10.1002/j.1532-2149.2014.497.x [DOI] [PubMed] [Google Scholar]
  81. Walker, L. S. , & Greene, J. W. (1991). The functional disability inventory: measuring a neglected dimension of child health status. Journal of Pediatric Psychology, 16, 39–58. 10.1093/jpepsy/16.1.39 [DOI] [PubMed] [Google Scholar]
  82. Walker, L. S. , Sherman, A. L. , Bruehl, S. , Garber, J. , & Smith, C. A. (2012). Functional abdominal pain patient subtypes in childhood predict functional gastrointestinal disorders with chronic pain and psychiatric comorbidities in adolescence and adulthood. Pain, 153, 1798–1806. 10.1016/j.pain.2012.03.026 [DOI] [PMC free article] [PubMed] [Google Scholar]
  83. Watson, K. D. , Papageorgiou, A. C. , Jones, G. T. , Taylor, S. , Symmons, D. P. , Silman, A. J. , & Macfarlane, G. J. (2002). Low back pain in schoolchildren: occurrence and characteristics. Pain, 97, 87–92. 10.1016/S0304-3959(02)00008-8 [DOI] [PubMed] [Google Scholar]
  84. Wojtowicz, A. A. , & Banez, G. A. (2015). Adolescents with chronic pain and associated functional disability: A descriptive analysis. Journal of Child Health Care, 19, 478–484. 10.1177/1367493514523157 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The data and code used to support the findings of this study are available from the corresponding author upon request.

Articles from European Journal of Pain (London, England) are provided here courtesy of Wiley

RESOURCES