|Year : 2018 | Volume
| Issue : 3 | Page : 180-185
A classification analysis of musculoskeletal complaints and its association with anxiety, depression and psychological distress: Results from a large-scale cross-sectional study of adult Iranian population
Maryam Yazdi1, Awat Feizi2, Ammar Hassanzadeh Keshteli3, Hamid Afshar4, Peyman Adibi5
1 Department of Biostatistics and Epidemiology, School of Health, Isfahan University of Medical Sciences, Isfahan, Iran
2 Department of Biostatistics and Epidemiology, School of Health, Isfahan University of Medical Sciences; Psychosomatic Research Center, Isfahan University of Medical Sciences; Integrative Functional Gastroenterology Research Center, Isfahan University of Medical Sciences, Isfahan, Iran
3 Integrative Functional Gastroenterology Research Center, Isfahan University of Medical Sciences, Isfahan, Iran; Department of Medicine, University of Alberta, Edmonton, Alberta, Canada
4 Psychosomatic Research Center, Isfahan University of Medical Sciences, Isfahan, Iran
5 Integrative Functional Gastroenterology Research Center; Department of Internal Medicine, School of Medicine, Isfahan University of Medical Sciences, Isfahan, Iran
|Date of Web Publication||24-Sep-2018|
Isfahan University of Medical Sciences, P. O. Box 319, Hezar-Jerib Ave., Isfahan 81746 73461
Source of Support: None, Conflict of Interest: None
Objectives: The current study aimed to classify study population in homogenous groups based on musculoskeletal complaints and investigate the association of common psychological disorders with musculoskeletal complaints in a large sample of Iranian adults. Methods: Using data from a cross-sectional sample of adult Iranian population (n = 4762), individuals were classified in meaningful subgroups based on their musculoskeletal complaints and the association of common psychological disorders with musculoskeletal complaints was investigated in identified classes using structural equation mixture model. Musculoskeletal complaints score extracted through eight items concerning site-specific or widespread musculoskeletal symptoms. Psychological distress was measured by General Health Questionnaire (GHQ-12). Depression and anxiety were measured using Persian validated version of the Hospital Anxiety and Depression Scale (HADS). Results: Two classes characterised by high (15.5%) and low (84.5%) levels of musculoskeletal complaints, were identified using structural equation mixture modeling. All musculoskeletal symptoms had a higher prevalence among participants allocated to 'high musculoskeletal complaints' class compared to 'low musculoskeletal complaints' class. Severe fatigue and back pain were the most reported complaints. Anxiety, depression and psychological distress were positively associated with musculoskeletal complaints score in identified classes, controlling for sex and age. Anxiety showed a stronger association with musculoskeletal complaints score compared to depression and psychological distress. Conclusions: Musculoskeletal complaints can be summarised in a categorical and dimensional structure in the adult study population. Common psychological disorders including anxiety, depression and psychological distress are significantly associated with musculoskeletal complaints. These findings could be useful for dealing with prevention and treatment programmes.
Keywords: Anxiety, depression, latent class, musculoskeletal complaints, psychological distress, structural equation modeling
|How to cite this article:|
Yazdi M, Feizi A, Keshteli AH, Afshar H, Adibi P. A classification analysis of musculoskeletal complaints and its association with anxiety, depression and psychological distress: Results from a large-scale cross-sectional study of adult Iranian population. Adv Hum Biol 2018;8:180-5
|How to cite this URL:|
Yazdi M, Feizi A, Keshteli AH, Afshar H, Adibi P. A classification analysis of musculoskeletal complaints and its association with anxiety, depression and psychological distress: Results from a large-scale cross-sectional study of adult Iranian population. Adv Hum Biol [serial online] 2018 [cited 2020 Oct 23];8:180-5. Available from: https://www.aihbonline.com/text.asp?2018/8/3/180/241921
| Introduction|| |
Musculoskeletal complaints (MSC) are a common problem in the general population. Musculoskeletal symptoms and disorders encompass pain and other symptoms at specific anatomical sites, as well as more generalised symptoms as seen in fibromyalgia and chronic widespread pain. Chronic MSC affects between 11% and 50% of the general population. Such variation in prevalence estimates is partly due to differences in the definition of musculoskeletal symptoms and the methods used for its evaluation.
Clinical and epidemiological studies indicate musculoskeletal symptoms could be the manifestations of mental illnesses in the form of physical symptoms. Several studies have reported a significant association between psychological disorders and pain in sites of back, neck, head, joints or face.,, A systematic review indicated that depressive symptoms is associated with higher levels of pain intensity, more functional limitation and disability and worse prognosis. However, previous researches not accurately addressed this question that which psychological disorders will be most strongly associated with musculoskeletal symptoms.
In addition, previous studies have been evaluated MSC mostly site-specific, or only classify persons based on major patterns of site-specific complaints in different class using latent class and cluster analysis.,, In the latent class analysis approach, individuals assigned to the most probable latent classes based on their observed symptoms so that the resulted latent classes can then elucidate diagnostic subgroups or subtypes. This categorical view of MSC has advantage that clarify needs of clinical remedies and reporting for healthcare authorities. However, latent class analysis and the categorical approach to MSC do not consider the range in the severity of musculoskeletal impairment within and across diagnostic classes. Continues and dimensional nature of disorders has its counterpart in factor analysis. Here, continuous latent variables, called factors, are extracted from correlated symptoms. Extracted factors provides quantitative scores, representing the severity of the impairment in each individual. Factor mixture model uses a hybrid of both categorical and continuous latent variables, which allows the underlying structure of MSC to be simultaneously dimensional and categorical. This structure is considered categorical because factor mixture model categorised the individuals into homogenous subgroups and it is as well considered dimensional because of accounting the within groups heterogeneity by continuous latent variables., An extension of factor mixture models known by structural equation mixture model (SEMM) allows to explanatory variables affect both latent variables and group membership.,
The present study aimed to construct a continues measurement as MSC profile (latent factor) explaining the severity of musculoskeletal symptoms and classify the study population into more homogeneous subgroups (latent class) based on it. Simultaneously, the association of common psychological disorders, i.e., anxiety, depression and psychological distress with constructed MSC profile across identified subgroups, controlling for demographic variables was evaluated in a comprehensive modeling framework; SEMM.
| Methods|| |
Study design and participants
The current study was conducted within the part of the Study of the Epidemiology of Psychological, Alimentary Health and Nutrition (SEPAHAN) project performed among a large sample of adult Iranian population in 2011. The data were sampled in two phases using multistage random sampling and convenient in the last stage. In the first phase, demographic, lifestyle and nutritional factors related questionnaires were delivered to 10087 invited individuals in which, 8691 participants took part (response rate: 86.16%). At the second phase, questionnaires designed to obtain information about psychological disorders, gastrointestinal and somatoform symptoms, were distributed and 6239 questionnaires were completed (response rate: 64.64%). According to national identification numbers of participants, gathered data of both phases were merged. Finally, information of 4762 subjects was remained in the current analysis. The study protocol was explained for all the participants, and a signed informed consent was obtained from them. The Bioethics Committee of Isfahan University of Medical Sciences approved the study protocol (Project numbers: 189069, 189082 and 189086). More details could be found elsewhere.
Since in SEPAHAN project there was not a separate questionnaire to assess MSC so we referred to Lacourt psychosomatic questionnaire, consist of 47 items based on body feeling and psychiatric diagnosis and patient health questionnaire (PHQ-15); a standard 15 items questionnaire that measures a range of physical signs related to sickness. Eight items that were related to musculoskeletal symptoms and complaints were extracted as a tool to evaluate the frequency and severity of MSC. The items were included feeling headache, back pain, pain in joints, eyesore, severe fatigue, dizziness and confusion, chills and extreme cold and hot flashes in the last three mounts range on a four-point scale (never, sometimes, often and always). Internal consistency of this stablished tool assessed using Cronbach α was 0.81, indicating that there was strong reliability.
Psychological distress was measured by a self-administered 12-item general health questionnaire (GHQ-12) through that participants answer having experienced a special feeling or behaviour 'less than usual, no more than usual, fairly more than usual, or much more than usual' in the past few weeks., A participant could score between 0 and 12 points and a threshold score of four or more was used to specify an individual with high-distress level. The internal consistency of GHQ-12 calculated with Cronbach α coefficient was found 0.87 in Iranian population.
Anxiety and depression
Anxiety and depression were assessed by Persian validated version of the Hospital Anxiety and Depression Scale (HADS). HADS consists of fourteen items, seven items for anxiety and seven for depression. Items were measured on a 4-point Likert scale ranging from 0 (not present) to 3 (considerable). The maximum score for anxiety and depression is 21. A threshold score of 8 or more on each subscale are considered as disorder. The validity and reliability of HADS was evaluated in Iranian population. The Cronbach α coefficients were 0.78 and 0.86 for HADS anxiety and depression subscales, respectively.
Assessment of other variables
Demographic variables consist of sex (male, female), age in year, marital status (single, married), educational level (under diploma, diploma (12-year formal education) and university graduate) and lifestyle variables included body mass index (kg/m2) and physical activity (inactive and moderately inactive/moderately active and active) measured by General Practice Physical Activity Questionnaire.
MSC score extracted through eight items concerning site-specific or widespread MSC using factor analysis. To ensure that the resulted factor is valid and could be replicated, we used a random split sample method and divided the sample in half. Confirmatory factor analysis performed in the second half of the sample and replicability of exploratory results was evaluated using the Tucker–Lewis coefficient (TLI), the comparative fit index (CFI), as well as the root mean square error of approximation (RMSEA). The RMSEA values <0.10 and TLI and CFI values >0.9 are indicating an acceptable fitness., Furthermore, internal consistency measured using Cronbach α in the second half sample was computed. Values near one indicate sufficient reliability of the instrument.
For following up our main study objective, i.e., if the studied population could be classified into meaningful subgroups based on MSC, SEMM approach was used. In the SEMM framework, we could simultaneously investigate the association of common psychological disorders (i.e., anxiety, depression and psychological distress) with MSC in identified classes. The optimum number of classes was determined through comparing the Bayesian information criteria (BIC) and entropy indices across models. Higher entropy and lower BIC imply better class discriminant and model fitting., Due to skewness and departure from normality assumption, we also considered skew normal distribution as an alternative to a normal distribution assumption for latent response, i.e., MSC score. To improve the classification, additional covariates; sex and age were used to model the variation in class membership. SEMM was conducted by Mplus version 8, and the model parameters were estimated using maximum likelihood method.
| Results|| |
Of 4762 adults participated in the study 2657 (55.8%) were females and 3776 (81.2%) were married. The mean age was 36.63 ± 8.93 years. About 57.2% of participants had college education, 28.3% had diploma and 13.8% were under diploma. Nearly 34.8% of participants reported having a regular physical activity. About 3.5% of individuals were underweight, 37.1% were overweight and 9.4% were obese. Psychological distress, anxiety and depression were identified in 23.1%, 5.8% and 10.4% of participants, respectively.
One factor was extracted from musculoskeletal symptoms-related items, explaining 45% of the variance. Factor loadings corresponded to eight considered items were as the following: back pain (0.68), pain in joint (0.71), eyesore (0.6), severe fatigue (0.75) dizziness and confusion (0.75), chills and extreme cold (0.62), headache (0.67) and hot flashes (0.61). In a primary analysis to ensure considered items and consequently extracted factor, could represent a valid instrument to quantify MSC, the sample randomly was spilt half and exploratory factor analysis was performed on the first half and confirmatory analysis on the second half. Fitness criteria of confirmatory factor analysis performed in the second half sample were in acceptable range, indicating that extracted factor in the exploratory factor analysis could be replicated (CFI = 0.97, TLI = 0.96 and RMSEA = 0.039). Internal consistency measured using Cronbach α in the second half sample (alpha = 0.83) indicated adequate reliability of the instrument.
The latent structure or unobserved heterogeneity of the study population regarding MSC score and its association with psychological disorders was recognised using SEMM. Due to considerable skewness of MSC score (0.90), skew normal distribution was also considered as an alternative to normal distribution assumption in building model process. Two-class model with skew normal distribution assumption and unequal mean, variance and skewness of MSC score across classes provided the best fit (BIC = 132564.89, entropy = 0.91). This fitness criteria indicate that individuals are correctly classified by our fitted model. The two recognised classes were labelled as 'high MSC' and 'low MSC;' since participants in the first class got higher scores of musculoskeletal symptoms (mean: 1.07 in the first-class vs. 0.64 in the second class). There were 4009 individuals (84.5%) in the 'low MSC' class and 737 participants (15.5%) in the 'high MSC' class. According to the two-class SEMM solution approximately, all items were significantly loaded on the extracted MSC factor.
[Figure 1] illustrates the frequency of reporting 'often' and 'always' MSC across two identified class. Severe fatigue, headache and back pain were respectively the most prevalent complains in two classes and chills and extreme cold were the lowest ones. The prevalence of all eight musculoskeletal symptoms for participants assigned to the 'high MSC' class was significantly higher than another class (P < 0.001). Among participants assigned to 'high MSC', 60.3% reported often or always feel severe fatigue, 34.2% headache and 33.4% back pain. In contrast, in participants assigned to 'low MSC', 29.8%, 18.5% and 18.7% reported often or always feel severe fatigue, headache and back pain, respectively.
|Figure 1: Self-reported frequency (‘often’ and ‘always’) of different musculoskeletal complaints across two identified class.|
Click here to view
[Table 1] represents regression coefficients regarding the association of psychological disorders; psychological distress, anxiety and depression with MSC score in the two identified classes.
|Table 1: Association of psychological distress, anxiety and depression with musculoskeletal complaints score in two identified classes|
Click here to view
In both classes, the anxiety and depression scores had significant direct associations with MSC score. The psychological distress as well has a positive significant association in 'low MSC' class, but not significant in 'high MSC' class, that might be arising from smaller sample size in that class. Furthermore, according to the magnitude of regression coefficients and standard errors, anxiety had the strongest impact and psychological distress weakest impact on MSC score. After adjustment of class membership for age and sex, the results did not show a significant change.
[Figure 2] illustrates the difference in mean of psychological disorders scores between two extracted latent classes. Mean of all psychological disorders score was higher in 'high MSC' class, indicating participant assigned to 'high MSC' class involved a higher level of distress in comparison with 'low MSC' class.
|Figure 2: Distribution of psychological disorders in term of mean and standard deviation across ‘low musculoskeletal complaints’ class (dash line) and ‘high musculoskeletal complaints’ class (solid line).|
Click here to view
| Discussion|| |
In this cross-sectional population-based study, two classes characterised by high (15.5%) and low (84%) levels of MSC, were identified from a large sample of Iranian adults using structural equation mixture modeling (SEMM). Prevalence of all MSC for participants assigned to 'high MSC' class was significantly higher than 'low MSC' class. Severe fatigue and back pain were the most prevalent complaints reported by participants. In the current study, three common psychological disorders, i.e., anxiety, depression and psychological distress were positively associated with the MSC score in identified classes. Anxiety showed a stronger association with MSC score compared to depression and psychological distress.
The current study is the first one that classified individuals based on their MSC and evaluate psychological condition in homogenous groups using SEMM.
To the best of our knowledge, there is not any study which classified a general large population into homogeneous subgroups based on musculoskeletal disorders using SEMM. However, other statistical approaches, i.e., clustering, factor analysis and latent class analysis in some studies were used to classify musculoskeletal disorders.,, In Gold et al.'s study, among patients with upper extremity musculoskeletal disorders, using cluster analysis, the patients were classified to mild thorough severe diffuse musculoskeletal disorders subgroups. Furthermore, the cluster analysis was used to classified patients on the basis of their individual course of low back pain over a 6 months period. Hartvigsen et al. investigated patterns of musculoskeletal pain based on primary pain site in the 4817 Danes using latent class analysis.
On the other hand, previous studies on population classification according to musculoskeletal symptoms have not accounted simultaneously the impact of psychological disorders. Classification method based on SEMM in our study permitted us to inter some psychological determinants that explain the variation of MSC in the modeling framework. There is evidence that psychological illnesses usually manifest themselves as forms of physical symptoms. Several systematic reviews have indicated the psychological disorders are risk factors for musculoskeletal symptoms., In a 10-year follow-up on metal industry employees, it was found musculoskeletal symptoms were associated with the change in the stress symptoms in men, as did the clinical findings in the neck-shoulder and low back regions. de Heer et al. using separate multinomial logistic regressions in a general-based study population showed that those with a current depressive disorder, remitted depressive or anxiety disorder have high odds for having musculoskeletal pain in at least one sites of back, neck, head, joints, or face. Several studies have reported an increased prevalence of headache in patients with major depression and anxiety disorders., Psychological distress in Reme et al.'s study was found to be higher in patient care workers with musculoskeletal pain. One question, not accurately settled in previous researches in this area, which of the psychological disorders will be most strongly associated with musculoskeletal symptoms, particularly when adjusting for demographic variables and other psychological disorders. In evaluating the simultaneous association of depression, anxiety and psychological distress scores with MSC score, our findings implied anxiety score had the strongest association with MSC score.
The link between psychological disorders and musculoskeletal disorders can be explained from biological perspective. Some investigators have reported the possible role of neurotransmitters  and cytokine receptors. However, there is still no proved neurochemical explanation for the association of low mood and musculoskeletal condition.
There are evidences that support sex difference in musculoskeletal disorders. Furthermore, age-related changes in the musculoskeletal system is well documented. SEMM allowed us to consider the impact of age and sex difference to find subgroups. Consistent with previous studies, being female and higher age significantly increased the chance of allocating to 'high MSC' class.
It is important to mention some strengths and limitations of the present study. A major strength of our large population-based study was the application of SEMM for identifying a certain measurement as MSC score and stratifying study population based on it. In this way because of skewness of extracted score, we assumed skew normal distribution instead relevant normal assumption to get better fitness and estimations. Furthermore, common psychological disorders, i.e., psychological distress, anxiety and depression were evaluated in identified classes through a unique modeling framework. However, due to the cross-sectional nature of study design, we could not infer cause–effect association from the findings. Furthermore since the data collected by self-reported questionnaires, that could lead to misspecification of participants to correct class.
| Conclusions|| |
The results of the present study suggested that MSC had a dimensional-categorical structure within the study population. In addition, we showed that the common psychological disorders including anxiety, depression and psychological distress are significantly associated with MSC. This finding could be useful for dealing with prevention and treatment programmes. For example, categorising treatment of musculoskeletal impairments according to patients' psychological indicators may make cost-effective development in the treatment process. Finally, prevention and early detection of psychological symptoms need to be highlighted in adults' mental healthcare programmes since impose a heavy burden on persons and community through consequent physical illnesses and disabilities.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
McBeth J, Jones K. Epidemiology of chronic musculoskeletal pain. Best Pract Res Clin Rheumatol 2007;21:403-25.
Cimmino MA, Ferrone C, Cutolo M. Epidemiology of chronic musculoskeletal pain. Best Pract Res Clin Rheumatol 2011;25:173-83.
Vargas-Prada S, Coggon D. Psychological and psychosocial determinants of musculoskeletal pain and associated disability. Best Pract Res Clin Rheumatol 2015;29:374-90.
de Heer EW, Gerrits MM, Beekman AT, Dekker J, van Marwijk HW, de Waal MW, et al.
The association of depression and anxiety with pain: A study from NESDA. PLoS One 2014;9:e106907.
Leino P, Magni G. Depressive and distress symptoms as predictors of low back pain, neck-shoulder pain, and other musculoskeletal morbidity: A 10-year follow-up of metal industry employees. Pain 1993;53:89-94.
Reme SE, Dennerlein JT, Hashimoto D, Sorensen G. Musculoskeletal pain and psychological distress in hospital patient care workers. J Occup Rehabil 2012;22:503-10.
Bair MJ, Robinson RL, Katon W, Kroenke K. Depression and pain comorbidity: A literature review. Arch Intern Med 2003;163:2433-45.
Hartvigsen J, Davidsen M, Hestbaek L, Sogaard K, Roos EM. Patterns of musculoskeletal pain in the population: A latent class analysis using a nationally representative interviewer-based survey of 4817 danes. Eur J Pain 2013;17:452-60.
Axén I, Bodin L, Bergström G, Halasz L, Lange F, Lövgren PW, et al.
Clustering patients on the basis of their individual course of low back pain over a six month period. BMC Musculoskelet Disord 2011;12:99.
Gold JE, Piligian G, Glutting JJ, Hanlon A, Frings-Dresen MH, Sluiter JK, et al.
Cluster analysis of symptoms among patients with upper extremity musculoskeletal disorders. J Occup Rehabil 2010;20:526-36.
Muthén B. Should substance use disorders be considered as categorical or dimensional? Addiction 2006;101 Suppl 1:6-16.
Clark SL, Muthén B, Kaprio J, D'Onofrio BM, Viken R, Rose RJ, et al.
Models and strategies for factor mixture analysis: An example concerning the structure underlying psychological disorders. Struct Equ Modeling 2013;20: 681-703.
Lubke GH, Muthén B. Investigating population heterogeneity with factor mixture models. Psychol Methods 2005;10:21-39.
Jedidi K, Jagpal HS, DeSarbo WS. Finite-mixture structural equation models for response-based segmentation and unobserved heterogeneity. Mark Sci 1997;16:39-59.
Vermunt JK, Magidson J. Structural Equation Modeling: Mixture Models. Wiley StatsRef: Statistics Reference Online 2014.
Adibi P, Keshteli AH, Esmaillzadeh A, Afshar H, Roohafza H, Bagherian-Sararoudi R, et al
. The study on the epidemiology of psychological, alimentary health and nutrition (SEPAHAN): Overview of methodology. J Res Med Sci 2012;17:1.
Lacourt T, Houtveen J, van Doornen L. Functional somatic syndromes, one or many? An answer by cluster analysis. J Psychosom Res 2013;74:6-11.
Spitzer RL, Kroenke K, Williams JB. Validation and utility of a self-report version of PRIME-MD: The PHQ primary care study. Primary care evaluation of mental disorders. Patient health questionnaire. JAMA 1999;282:1737-44.
Montazeri A, Harirchi AM, Shariati M, Garmaroudi G, Ebadi M, Fateh A, et al.
The 12-item general health questionnaire (GHQ-12): Translation and validation study of the Iranian version. Health Qual Life Outcomes 2003;1:66.
Goldberg DP, Gater R, Sartorius N, Ustun TB, Piccinelli M, Gureje O, et al.
The validity of two versions of the GHQ in the WHO study of mental illness in general health care. Psychol Med 1997;27:191-7.
Montazeri A, Vahdaninia M, Ebrahimi M, Jarvandi S. The Hospital Anxiety and Depression Scale (HADS): Translation and validation study of the Iranian version. Health Qual Life Outcomes 2003;1:14.
National Collaborating Centre for Nursing Supportive Center. National Institute for Health and Clinical Excellence: Guidance. Irritable Bowel Syndrome in Adults: Diagnosis and Management of Irritable Bowel Syndrome in Primary Care. London: Royal College of Nursing (UK); 2008.
Osborne JW, Fitzpatrick DC. Replication analysis in exploratory factor analysis & 58; what it is and why it makes your analysis better. Pract Assess 2012;17:1-8.
Hooper D, Coughlan J, Mullen M. Structural equation modeling: Guidelines for determining model fit. Electron J Bus Res Methods: 2008;6:53-60.
Hair JF, Black WC, Babin BJ, Anderson RE. Multivariate Data Analysis. UK: Pearson Education Limited; 2013.
Browne MW, Cudeck R, Bollen KA, Long JS. Alternative ways of assessing model fit. Sage Focus Ed 1993;154:136.
Taber KS. The use of cronbach's alpha when developing and reporting research instruments in science education. Res Sci Educ 2017;1-24.
Muthén LK, Muthén B. Mplus. The comprehensive modelling program for applied researchers: user's guide. 2015;5.
Phyomaung PP, Dubowitz J, Cicuttini FM, Fernando S, Wluka AE, Raaijmaakers P, et al.
Are depression, anxiety and poor mental health risk factors for knee pain? A systematic review. BMC Musculoskelet Disord 2014;15:10.
Pincus T, Burton AK, Vogel S, Field AP. A systematic review of psychological factors as predictors of chronicity/disability in prospective cohorts of low back pain. Spine (Phila Pa 1976) 2002;27:E109-20.
Beghi E, Allais G, Cortelli P, D'Amico D, De Simone R, d'Onofrio F, et al.
Headache and anxiety-depressive disorder comorbidity: The HADAS study. Neurol Sci 2007;28 Suppl 2:S217-9.
Pompili M, Di Cosimo D, Innamorati M, Lester D, Tatarelli R, Martelletti P, et al.
Psychiatric comorbidity in patients with chronic daily headache and migraine: A selective overview including personality traits and suicide risk. J Headache Pain 2009;10:283-90.
Trivedi MH. The link between depression and physical symptoms. Prim Care Companion J Clin Psychiatry 2004;6:12-6.
Walker AK, Kavelaars A, Heijnen CJ, Dantzer R. Neuroinflammation and comorbidity of pain and depression. Pharmacol Rev 2014;66:80-101.
Wijnhoven HA, de Vet HC, Picavet HS. Explaining sex differences in chronic musculoskeletal pain in a general population. Pain 2006;124:158-66.
Roberts S, Colombier P, Sowman A, Mennan C, Rölfing JH, Guicheux J, et al
. Ageing in the musculoskeletal system: Cellular function and dysfunction throughout life. Acta Orthop 2016;87 Suppl 363:15-25.
[Figure 1], [Figure 2]