• Matt Fernandez, Lucia Colodro-Conde, Jan Hartvigsen, Manuela L Ferreira, Kathryn M Refshauge, Marina B Pinheiro, Juan R Ordoñana, and Paulo H Ferreira.
• Faculty of Health Sciences, The University of Sydney, 75 East St, Lidcombe, Sydney, NSW 2141, Australia. Electronic address: mfer4234@uni.sydney.edu.au.
• Spine J. 2017 Jul 1; 17 (7): 905-912.

Background ContextPain is commonly associated with symptoms of depression or anxiety, although this relationship is considered bidirectional. There is limited knowledge regarding causal relationships.PurposeThis study aims to investigate whether chronic low back pain (LBP) increases the risk of depression or anxiety symptoms, after adjusting for shared familial factors.Study DesignThis is a longitudinal, genetically informative study design from the Murcia Twin Registry in Spain.Patient SampleThe patient sample included 1,269 adult twins with a mean age of 53 years.Outcome MeasuresThe outcome of depression or anxiety symptoms was evaluated with EuroQol questionnaire.MethodsUsing logistic regression analyses, twins were initially assessed as individuals in the total sample analysis, followed by a co-twin case-control, which was partially (dizygotic [DZ] twins) and fully (monozygotic [MZ] twins) adjusted for shared familial factors. There was no external funding for this study and no conflict of interest was declared.ResultsThere was a significant association between chronic LBP and the risk of depression or anxiety symptoms in the unadjusted total sample analysis (odds ratio [OR]: 1.81, 95% confidence interval [CI]: 1.34-2.44). After adjusting for confounders, the association remained significant (OR: 1.43, 95% CI: 1.05-1.95), although the adjusted co-twin case-control was non-significant in DZ (OR: 1.03, 95% CI: 0.50-2.13) and MZ twins (OR: 1.86, 95% CI: 0.63-5.51).ConclusionsThe relationship between chronic LBP and the future development of depression or anxiety symptoms is not causal. The relationship is likely to be explained by confounding from shared familial factors, given the non-statistically significant associations in the co-twin case-control analyses.Copyright © 2017 Elsevier Inc. All rights reserved.

21 keywords...

hide…