Brain structure, psychosocial, and physical health in acute and chronic back pain: a UK Biobank study

Influence of chronic pain on regional brain volume reduction in a general older Japanese population: a longitudinal imaging analysis from the Hisayama Study

Longitudinal analyses of the influence of chronic pain on pain-related regional brain volumes in general populations are warranted. This prospective cohort study investigated the association between the presence of chronic pain at baseline and the subsequent changes in pain-related regional brain volumes among Japanese community-dwelling older residents. Participants aged 65 years or older who underwent brain magnetic resonance imaging (MRI) scans in both 2012 and 2017 were included. According to the presence or absence of chronic pain (defined as pain lasting for longer than 3 months) in 2012, participants were categorized into a 'chronic pain' group and 'no chronic pain' group. Region-of-interest analyses for the ventrolateral prefrontal cortex, dorsolateral prefrontal cortex, orbitofrontal cortex, postcentral gyrus, insular cortex, thalamus, anterior cingulate cortex, posterior cingulate cortex, amygdala and hippocampus were performed using FreeSurfer software. Whole-brain analysis was conducted by voxel-based morphometry. Rates of change in regional brain volume at 5 years after baseline were estimated using analysis of covariance. Among the 766 participants included in the FreeSurfer analysis, 444 (58%) were female and 287 (37%) were categorized into the chronic pain group. The results of FreeSurfer analysis showed that the chronic pain group had significantly greater decreases in regional volume in the postcentral gyrus (-2.187% in the chronic pain group versus -1.681% in the no chronic pain group, P = 0.01), thalamus (-4.400% versus -3.897%, P = 0.006), anterior cingulate cortex (-2.507% versus -1.941%, P = 0.004) and amygdala (-4.739% versus -4.022%, P = 0.03) compared to the no chronic pain group after adjusting for age, sex, education attainment, marital status, hypertension, diabetes, serum total cholesterol level, body mass index, current smoking, current drinking, regular exercise, cerebrovascular lesions on MRI, activities in daily living disability and depressive symptoms. Among the 730 participants included in the voxel-based morphometry analysis, 433 (59%) were female and 272 (37%) were categorized into the chronic pain group. The voxel-based morphometry analysis showed that the chronic pain group had a significantly greater regional volume decrease in the right anterior insula than the no chronic pain group. Our findings suggest that the presence of chronic pain at baseline is associated with a significantly greater decrease in the volume of pain-related brain regions at 5 years after baseline in community-dwelling older Japanese.

Exploring the associations between the presence, characteristics, and biopsychosocial covariates of pain and lifetime depression in adolescents: A cross-sectional ABCD study analysis

INTRODUCTION

Depression and pain co-occur, even during adolescence. However, there is limited knowledge on the association between pain and lifetime depression, and which biopsychosocial measures are associated with this co-occurrence.

METHODS

Cross-sectional analysis of the Adolescent Brain and Cognitive Development (ABCD) two-year follow-up. We explored associations between the presence and characteristics of past month pain (intensity, duration, activity limitations, and number of pain sites) and lifetime depression using logistic regression. We explored associations of brain structure, physical, behavioural, emotional, social, and cognitive measures with lifetime depression and past month pain compared to having had one or neither condition using multinomial logistic regression.

RESULTS

A total of 5211 adolescents (mean age = 12.0 years) who had: (1) no lifetime mental ill-health and no pain (n = 3327); (2) pain only (n = 1407); (3) lifetime depressive disorder but no pain (n = 272); and (4) lifetime depressive disorder and pain (n = 205) were included. Pain presence was associated with lifetime depression (OR[95%CI]: 1.76 [1.45, 2.13], p < 0.001). Pain-related activity limitations (1.13 [1.06, 1.21], p < 0.001) and the number of pain sites (1.06 [1.02, 1.09], p < 0.001) were associated with lifetime depression. Various behavioural, emotional, social, and cognitive, but not brain structure or physical measures, were associated with lifetime depression and past month pain.

LIMITATIONS

Longitudinal analyses should validate prognostic markers for predicting co-occurring depression and pain.

CONCLUSIONS

Results support an association between the presence and characteristics of pain and lifetime depression during adolescence and could indicate the need for more integrated recognition and clinical care of youth experiencing both depression and pain.

Gray matter volume of limbic brain structures during the development of chronic back pain: a longitudinal cohort study

ABSTRACT

This study set out to investigate in a population-based longitudinal cohort, whether chronification of back pain (BP) is related to structural gray matter changes in corticolimbic brain structures. Gray matter volume (GMV) was measured in participants with chronic BP (CBP, n = 168) and controls without chronic pain (n = 323) at 2 time points with an interval of 7 years (baseline t1, follow-up t2). Over this time period, participants with CBP showed an increase of GMV in the left ventral striatum, whereas controls showed a decrease. By contrast, participants with CBP had a GMV decrease in the left parahippocampal gyrus. Within the CBP group, pain duration was negatively associated with GMV in the left caudate. Those with emerging CBP had less GMV in the right entorhinal area, right amygdala, and left medial frontal cortex. Additional variables differing between those who had BP at t1 and later developed CBP or not were pain intensity, body mass index, and depression score. In sum, these findings are in accordance with the notion that limbic brain properties are both predisposing risk factors and drivers of brain reorganization during the development of CBP.

Does sedentary time and physical activity predict chronic back pain and morphological brain changes? A UK biobank cohort study in 33,402 participants

BACKGROUND

The relationship between sedentary time, physical activity, and chronic back pain remains unclear. The study aims to investigate whether sedentary time and physical activity predict chronic back pain and morphological brain changes.

METHODS

This cohort study recruited adults aged 37-73 years enrolled between 2006 and 2010, with follow-up until 2014. The total cohort comprised 33,402 participants (mean age: 54.53). Data were collected on daily sedentary time, physical activity, lifestyle factors, and health outcomes.

RESULTS

After nearly 8-year follow-up, 3,006 individuals (9.00%) reported chronic back pain in total. Individuals with daily sedentary time exceeding 6 h had a 33% higher risk of chronic back pain compared to those with sedentary time of 2 h or less (RR, 1.33, 95%CI, 1.17-1.52). Sedentary time was also associated with decreased grey matter volume in several brain regions, including bilateral primary somatosensory cortex (S1), secondary somatosensory cortex, putamen, primary motor cortex (M1), insula, hippocampus, amygdala, as well as right supplementary motor area, left medial frontal cortex, and right anterior cingulate cortex (FDR-corrected p-value < 0.05). Compared to individuals who sat for more than 6 h with light physical activity, those engaging in moderate physical activity with sedentary time of 2 h or less (RR, 0.71, 95%CI, 0.52-0.99) exhibited a significant decrease in chronic back pain risk. In addition, replacing sedentary time with equivalent amount of physical activity also demonstrated a reduction in the risk of chronic back pain (RR, 0.87, 95%CI, 0.77-0.99) and increased the reginal grey matter volumes including the amygdala, insula, M1, putamen and S1.

CONCLUSIONS

Prolonged sedentary time is associated with heightened risks of chronic back pain and deterioration in brain health.

Characteristics of locus coeruleus functional connectivity network in patients with comorbid migraine and insomnia

BACKGROUND

Migraine and insomnia are prevalent conditions that often co-occur, each exacerbating the other and substantially impacting the quality of life. The locus coeruleus (LC), a brainstem region responsible for norepinephrine synthesis, participates in pain modulation, sleep/wake cycles, and emotional regulation, rendering it a potential nexus in the comorbidity of migraine and insomnia. Disruptions in the LC-noradrenergic system have been hypothesized to contribute to the comorbidities of migraine and insomnia, although neuroimaging evidence in humans remains scarce. In this study, we aimed to investigate the intrinsic functional connectivity (FC) network of the LC in patients with comorbid migraine and subjective chronic insomnia and patients with migraine with no insomnia (MnI) using resting-state functional magnetic resonance imaging (rs-fMRI) and seed-based FC analyses.

METHODS

In this cross-sectional study, 30 patients with comorbid migraine and chronic insomnia (MI), 30 patients with MnI, and 30 healthy controls (HCs) were enrolled. Participants underwent neuropsychological testing and rs-fMRI. The LC-FC network was constructed using seed-based voxel-wise FC analysis. To identify group differences in LC-FC networks, voxel-wise covariance analysis was conducted with sex and age as covariates. Subsequently, a partial correlation analysis was conducted to probe the clinical relevance of aberrant LC-FC in patients with MI and MnI.

RESULTS

Except for the insomnia score, no other significant difference was detected in demographic characteristics and behavioral performance between the MI and MnI groups. Compared with HCs, patients with MI exhibited altered LC-FC in several brain regions, including the dorsomedial prefrontal cortex (DMPFC), anterior cerebellum, dorsolateral prefrontal cortex (DLPFC), thalamus, and parahippocampal gyrus (PHG). Lower FC between the LC and DLPFC was associated with greater insomnia severity, whereas higher FC between the LC and DMPFC was linked to longer migraine attack duration in the MI group.

CONCLUSION

Our findings reveal the presence of aberrant LC-FC networks in patients with MI, providing neuroimaging evidence of the interplay between these conditions. The identified LC-FC alterations may serve as potential targets for therapeutic interventions and highlight the importance of considering the LC-noradrenergic system in the management of MI.

Association between seated trunk control and cortical sensorimotor white matter brain changes in patients with chronic low back pain

Trunk control involves integration of sensorimotor information in the brain. Individuals with chronic low back pain (cLBP) have impaired trunk control and show differences in brain structure and function in sensorimotor areas compared with healthy controls (HC). However, the relationship between brain structure and trunk control in this group is not well understood. This cross-sectional study aimed to compare seated trunk control and sensorimotor white matter (WM) structure in people with cLBP and HC and explore relationships between WM properties and trunk control in each group. Thirty-two people with cLBP and 35 HC were tested sitting on an unstable chair to isolate trunk control; performance was measured using the 95% confidence ellipse area (CEA95) of center-of-pressure tracing. A WM network between cortical sensorimotor regions of interest was derived using probabilistic tractography. WM microstructure and anatomical connectivity between cortical sensorimotor regions were assessed. A mixed-model ANOVA showed that people with cLBP had worse trunk control than HC (F = 12.96; p < .001; ηp2 = .091). There were no differences in WM microstructure or anatomical connectivity between groups (p = 0.564 to 0.940). In the cLBP group, WM microstructure was moderately correlated (|r| = .456 to .565; p ≤ .009) with trunk control. Additionally, the cLBP group demonstrated stronger relationships between anatomical connectivity and trunk control (|r| = .377 to .618 p < .034) compared to the HC group. Unique to the cLBP group, WM connectivity between right somatosensory and left motor areas highlights the importance of interhemispheric information exchange for trunk control. Parietal areas associated with attention and spatial reference frames were also relevant to trunk control. These findings suggest that people with cLBP adopt a more cortically driven sensorimotor integration strategy for trunk control. Future research should replicate these findings and identify interventions to effectively modulate this strategy.

Understanding the Mechanosensitivity of the Median Nerve in Pre-Surgical Carpal Tunnel Syndrome Patients: A Correlational Study

(1) Background: Neurodynamic tests are recommended for the diagnosis of entrapment neuropathies such as carpal tunnel syndrome (CTS). However, their association with clinical variables in severe patients or patients with associated comorbidities is poorly documented. This study aims to analyze the association between the mechanosensitivity of the median nerve and symptoms, function and psycho-social variables in moderate and severe carpal tunnel syndrome patients with comorbidities; (2) Methods: Correlational study. In total, 42 pre-surgical patients (24 females; 59.1 ± 12.7 years) included in the Spanish Public Healthcare System with an electrodiagnostic of CTS were selected. Sociodemographic variables and clinical features (symptoms, function, sensitivity and quality of life evaluated with the 36-item Short Form Survey (SF-36) and the Medical Outcomes Study Sleep Scale (MOS-sleep) were recorded. Upper Limb Neurodynamic Test 1 was used to evaluate neural mechanosensitivity; (3) Results: The 81% had a severe CTS and 78.6% had some comorbidity. The average time from the first medical visit to the surgeon's visit was 365.5 days. Median nerve mechanosensitivity correlated weakly with the SF-36 subscale, General Health, (Spearman's rho = 0.367) and MOS sleep scale, Awaken Short of Breath or with headache dimension (Spearman's rho = -0.353) and moderately with SF-36 subscale, Social Functioning (Spearman's rho = 0.553); (4) Conclusions: No associations were observed for median nerve mechanosensitivity, except for quality of life and sleep. Both social determinants and clinical variables should be considered when examining and treating these patients.

How to Interpret Effect Sizes for Biopsychosocial Outcomes and Implications for Current Research

Biopsychosocial factors are associated with pain, but they can be difficult to compare. One way of comparing them is to use standardized mean differences. Previously, these effects sizes have been termed as small, medium, or large, if they are bigger than or equal to, respectively, .2, .5, or .8. These cut-offs are arbitrary and recent evidence showed that they need to be reconsidered. We argue it is necessary to determine cut-offs for each biopsychosocial factor. To achieve this, we propose 3 potential approaches: 1) examining, for each factor, how the effect size differs depending upon disease severity; 2) using an existing minimum clinically important difference to anchor the large effect size; and 3) define cut-offs by comparing data from people with and without pain. This is important for pain research, as exploring these methodologies has potential to improve comparability of biopsychosocial factors and lead to more directed treatments. We note assumptions and limitations of these methods that should also be considered. PERSPECTIVE: Standardized mean differences can estimate effect sizes between groups and could theoretically allow for comparison of biopsychosocial factors. However, common thresholds to define effect sizes are arbitrary and likely differ based on outcome. We propose methods that could overcome this and be used to derive biopsychosocial outcome-specific effect sizes.

Association of persistent musculoskeletal pain with dementia risk score in adults aged 45 years or older: The China health and retirement longitudinal study

BACKGROUND

Recent studies have confirmed an association between pain and dementia. Whether musculoskeletal pain in the spine, upper limbs, and lower limbs is associated with dementia risk remains unclear. The longitudinal effect of musculoskeletal pain on dementia risk also remains unclear.

AIMS

This work aimed to investigate the association between musculoskeletal pain and dementia risk score.

METHODS

We conducted cross-sectional and longitudinal analyses using data from the China Health and Retirement Longitudinal Study. Participants aged 45 years or older were recruited in 2011. A total of 10,759 participants with complete pain information at baseline were eligible for the cross-sectional analysis, and 5,855 were eligible for the longitudinal analyses. We utilized the Rotterdam Study Basic Dementia Risk Model (BDRM) to assess dementia risk. Generalized estimating equations were used to investigate the associations.

RESULTS

Compared with participants without persistent musculoskeletal pain, those with persistent musculoskeletal pain (standardized, β = 0.83; 95 % CI: 0.06, 1.61, p = 0.036), multisite pain (sites≧5; β = 1.52; 95 % CI: 0.13, 2.91, p = 0.032), neck pain (β = 2.33; 95 % CI: 0.41, 4.25, p = 0.018), back pain (β = 2.12; 95 % CI: 0.43, 3.82, p = 0.014), waist pain (β = 1.09; 95 % CI: 0.07, 2.11, p = 0.037), shoulder pain (β = 1.74; 95 % CI: 0.46, 3.02, p = 0.008), wrist pain (β = 2.72; 95 % CI: 0.42, 5.02, p = 0.021), and knee pain (β = 1.91; 95 % CI: 0.70, 3.13, p = 0.002) had a higher BDRM score during 4 years of follow-up.

CONCLUSIONS

Promoting the management of musculoskeletal pain may be beneficial in reducing the dementia risk score.

Chronic pain is associated with less grey matter volume in the anterior cingulum, anterior and posterior insula and hippocampus across three different chronic pain conditions

BACKGROUND

Chronic pain of different aetiologies and localization has been associated with less grey matter volume (GMV) in several cortical and subcortical brain areas. Recent meta-analyses reported low reproducibility of GMV alterations between studies and pain syndromes.

METHODS

To investigate GMV in common chronic pain conditions defined by body location (chronic back pain, n = 174; migraine, n = 92; craniomandibular disorder, n = 39) compared to controls (n = 296), we conducted voxel-based morphometry and determined GMV from high-resolution cranial MRIs obtained in an epidemiologic survey. Mediation analyses were performed between the presence of chronic pain and GMV testing the mediators stress and mild depression. The predictability of chronic pain was investigated with binomial logistic regression.

RESULTS

Whole-brain analyses yielded reduced GMV within the left anterior insula and the anterior cingulate cortex, for a ROI approach additionally the left posterior insula and left hippocampus showing less GMV across all patients with chronic pain. The relationship of pain with GMV in the left hippocampus was mediated by self-reported stressors in the last 12 months. Binomial logistic regression revealed a predictive effect for GMV in the left hippocampus and left anterior insula/temporal pole for the presence of chronic pain.

CONCLUSIONS

Chronic pain across three different pain conditions was characterized by less GMV in brain regions consistently described for different chronic pain conditions before. Less GMV in the left hippocampus mediated by experienced stress during the last year might be related to altered pain learning mechanisms in chronic pain patients.

SIGNIFICANCE

Grey matter reorganization could serve as a diagnostic biomarker for chronic pain. In a large cohort, we here replicated findings of less grey matter volume across three pain conditions in the left anterior and posterior insula, anterior cingulate and left hippocampus. Less hippocampal grey matter was mediated by experienced stress.

Big data, big consortia, and pain: UK Biobank, PAINSTORM, and DOLORisk

Chronic pain (CP) is a common and often debilitating disorder that has major social and economic impacts. A subset of patients develop CP that significantly interferes with their activities of daily living and requires a high level of healthcare support. The challenge for treating physicians is in preventing the onset of refractory CP or effectively managing existing pain. To be able to do this, it is necessary to understand the risk factors, both genetic and environmental, for the onset of CP and response to treatment, as well as the pathogenesis of the disorder, which is highly heterogenous. However, studies of CP, particularly pain with neuropathic characteristics, have been hindered by a lack of consensus on phenotyping and data collection, making comparisons difficult. Furthermore, existing cohorts have suffered from small sample sizes meaning that analyses, especially genome-wide association studies, are insufficiently powered. The key to overcoming these issues is through the creation of large consortia such as DOLORisk and PAINSTORM and biorepositories, such as UK Biobank, where a common approach can be taken to CP phenotyping, which allows harmonisation across different cohorts and in turn increased study power. This review describes the approach that was used for studying neuropathic pain in DOLORisk and how this has informed current projects such as PAINSTORM, the rephenotyping of UK Biobank, and other endeavours. Moreover, an overview is provided of the outputs from these studies and the lessons learnt for future projects.

Evidence- and data-driven classification of low back pain via artificial intelligence: Protocol of the PREDICT-LBP study

In patients presenting with low back pain (LBP), once specific causes are excluded (fracture, infection, inflammatory arthritis, cancer, cauda equina and radiculopathy) many clinicians pose a diagnosis of non-specific LBP. Accordingly, current management of non-specific LBP is generic. There is a need for a classification of non-specific LBP that is both data- and evidence-based assessing multi-dimensional pain-related factors in a large sample size. The "PRedictive Evidence Driven Intelligent Classification Tool for Low Back Pain" (PREDICT-LBP) project is a prospective cross-sectional study which will compare 300 women and men with non-specific LBP (aged 18-55 years) with 100 matched referents without a history of LBP. Participants will be recruited from the general public and local medical facilities. Data will be collected on spinal tissue (intervertebral disc composition and morphology, vertebral fat fraction and paraspinal muscle size and composition via magnetic resonance imaging [MRI]), central nervous system adaptation (pain thresholds, temporal summation of pain, brain resting state functional connectivity, structural connectivity and regional volumes via MRI), psychosocial factors (e.g. depression, anxiety) and other musculoskeletal pain symptoms. Dimensionality reduction, cluster validation and fuzzy c-means clustering methods, classification models, and relevant sensitivity analyses, will classify non-specific LBP patients into sub-groups. This project represents a first personalised diagnostic approach to non-specific LBP, with potential for widespread uptake in clinical practice. This project will provide evidence to support clinical trials assessing specific treatments approaches for potential subgroups of patients with non-specific LBP. The classification tool may lead to better patient outcomes and reduction in economic costs.

Towards data-driven biopsychosocial classification of non-specific chronic low back pain: a pilot study

The classification of non-specific chronic low back pain (CLBP) according to multidimensional data could guide clinical management; yet recent systematic reviews show this has not been attempted. This was a prospective cross-sectional study of participants with CLBP (n = 21) and age-, sex- and height-matched pain-free controls (n = 21). Nervous system, lumbar spinal tissue and psychosocial factors were collected. Dimensionality reduction was followed by fuzzy c-means clustering to determine sub-groups. Machine learning models (Support Vector Machine, k-Nearest Neighbour, Naïve Bayes and Random Forest) were used to determine the accuracy of classification to sub-groups. The primary analysis showed that four factors (cognitive function, depressive symptoms, general self-efficacy and anxiety symptoms) and two clusters (normal versus impaired psychosocial profiles) optimally classified participants. The error rates in classification models ranged from 4.2 to 14.2% when only CLBP patients were considered and increased to 24.2 to 37.5% when pain-free controls were added. This data-driven pilot study classified participants with CLBP into sub-groups, primarily based on psychosocial factors. This contributes to the literature as it was the first study to evaluate data-driven machine learning CLBP classification based on nervous system, lumbar spinal tissue and psychosocial factors. Future studies with larger sample sizes should validate these findings.

Abnormal activation of brain regions in idiopathic trigeminal neuralgia patients by fMRI: An activation likelihood estimation meta-analysis

BACKGROUND

Idiopathic trigeminal neuralgia (ITN) is one of the most common types of neuropathic pain, severely affecting the physiological and psychological wellbeing of patients. Recently, fMRI has been used to examine abnormal activation of brain regions in patients with ITN. However, sample sizes have been small in these few studies, and the abnormally activated brain regions remain unclear. Therefore, in the present study, we retrieved and analyzed literature on the brain areas with abnormal or reduced activation in ITN patients, with the aim of providing insight into the neuropathological basis of the disease and to provide new targets for treatment.

METHODS

We retrieved resting state fMRI studies on trigeminal neuralgia patients from PubMed, the Web of Science and Scopus databases until November 2022, and we extracted the coordinates of the sites with increased or decreased activation. We used activation likelihood estimation (ALE) meta-analysis to identify regions of abnormal activation in ITN patients.

RESULTS

ALE meta-analysis revealed that the left caudate nucleus and right anterior ventral nucleus of the thalamus are abnormally hyperactivated in ITN patients. Moreover, ITN patients showed reduced activation in the left precuneus, middle temporal gyrus, lingual gyrus, and medial frontal gyrus.

CONCLUSION

ALE meta-analysis identified several brain regions with abnormally high or decreased activation in ITN patients. Sites with altered activation may be potential targets for non-invasive brain stimulation as adjunct therapy for ITN.

The functional connectivity of the basal ganglia subregions changed in mid-aged and young males with chronic prostatitis/chronic pelvic pain syndrome

Background

The Basal ganglia (BG) played a crucial role in the brain-level mechanisms of chronic pain disorders. However, the functional changes of BG in chronic prostatitis/chronic pelvic pain syndrome (CP/CPPS) are still poorly understood. This study investigated the BG subregions’ resting-state functional connectivity (rs-FC) in CP/CPPS patients compared with healthy controls.

Methods

Twenty eight patients with CP/CPPS and 28 age- and education-matched healthy males underwent clinical measurements and 3T brain MR imaging, including T1-weighted structural images and resting-state functional imaging. The data were analyzed by the seeded-based rs-FC analysis. Then, a machine learning method was applied to assess the feasibility of detecting CP/CPPS patients through the changed rs-FC.

Results

Compared with healthy males, patients presented decreased rs-FC between the BG subregions and right middle cingulate cortex, and correlated with pain (r = 0.51, p-uncorrected = 0.005) and urinary symptoms (r = –0.4, p-uncorrected = 0.034). The left superior temporal gyrus and right supramarginal gyrus showed decreased rs-FC with the BG subregions as well. The area under the receiver operating characteristic curve of 0.943 (accuracy = 80%, F1-score = 80.6%) was achieved for the classification of CP/CPPS patients and healthy males with support vector machine (SVM) based on the changed rs-FC.

Conclusion

These findings provide evidence of altered BG subregions’ rs-FC in CP/CPPS, which may contribute to our understanding of the BG’s role in CP/CPPS.

Chronic back pain sub-grouped via psychosocial, brain and physical factors using machine learning

Chronic back pain (CBP) is heterogenous and identifying sub-groups could improve clinical decision making. Machine learning can build upon prior sub-grouping approaches by using a data-driven approach to overcome clinician subjectivity, however, only binary classification of pain versus no-pain has been attempted to date. In our cross-sectional study, age- and sex-matched participants with CBP (n = 4156) and pain-free controls (n = 14,927) from the UkBioBank were included. We included variables of body mass index, depression, loneliness/social isolation, grip strength, brain grey matter volumes and functional connectivity. We used fuzzy c-means clustering to derive CBP sub-groups and Support Vector Machine (SVM), Naïve Bayes, k-Nearest Neighbour (kNN) and Random Forest classifiers to determine classification accuracy. We showed that two variables (loneliness/social isolation and depression) and five clusters were optimal for creating sub-groups of CBP individuals. Classification accuracy was greater than 95% for when CBP sub-groups were assessed only, while misclassification in CBP sub-groups increased to 35-53% across classifiers when pain-free controls were added. We showed that individuals with CBP could sub-grouped and accurately classified. Future research should optimise variables by including specific spinal, psychosocial and nervous system measures associated with CBP to create more robust sub-groups that are discernible from pain-free controls.