Background/Purpose: Adverse childhood experiences (ACEs), including maltreatment (abuse and neglect), have been associated with chronic pain, but the evidence is largely cross-sectional. Existing evidence is limited in its exploration of the potential mechanisms and mediators, as well as being subject to recall bias. Using the UK Biobank enabled us to investigate multiple mediators and also a suitable negative control exposure (NCE) – childhood sunburn – with which to assess the potential influence of recall bias.

Using causal counterfactual methods (g-methods), we aimed to estimate the following: the causal effect on the risk of chronic pain all over, of reporting childhood maltreatment; the indirect effect mediated by deprivation in adulthood, depression/anxiety, social support and sleep problems; the joint interactive effect of reporting both childhood maltreatment and recent trauma

<

Methods: We used the UK Biobank (application 1144, www.ukbiobank.ac.uk), a database of 500,000 people aged 40-69 recruited 2006-2010, some of whom were invited to complete questionnaires in 2016 (including 5 questions from the Childhood Trauma Screener) and 2019 (including pain items). Participants reported if they were “troubled by pain or discomfort, either all the time or on and off, that has been present for more than 3 months” and if this was “all over the body”. A directed acyclic graph identified key variables. Doubly-robust models, stratified by sex, estimated risk differences using inverse probability weights (IPW) and the teffects commands in Stata. IPW included ethnicity and age on leaving education as a proxy for childhood socioeconomic status. To estimate the joint interactive effect, we derived a 4-category combined exposure variable. We looked at mediation by depression/anxiety, sleep issues, deprivation and social support, with inverse odds weighting (IOW) in a generalized linear model.

Results: Of 118,347 participants, 41.8% reported childhood maltreatment, and 5.2% reported chronic pain all over (males 3.4%; females 6.6%). The risk of chronic pain all over was higher with exposure to childhood maltreatment, both in men (3.9% vs 1.3%; absolute risk difference 2.6% (95% CI 2.4%, 2.8%)) and women (8.1% vs 5.0%; absolute risk difference 3.0% (95% CI 2.6%, 3.5%)). One third of the risk increase was mediated by the included variables. There were independent (and similar) effects of childhood maltreatment and adult stressful life events, with a supra-additive interaction between the two exposures. A small risk difference was found for women (0.8%, 95% CI 0.3%, 1.3%) but not men using the NCE. Analyses on imputed data were confirmatory.

Conclusion: Exposure to childhood maltreatment increases the risk of chronic pain all over by 2-3% and the use of a negative control exposure suggest that this is not attributable to differential recall. Trauma in adulthood in combination with childhood trauma further increased risk.

About a third of the increased risk was indirect, largely through depression/anxiety. Identifying and treating survivors of maltreatment could help to prevent an important proportion of cases of chronic widespread pain and stressful life events in adulthood may represent an important timepoint in which to intervene.

ACR Meeting Abstracts - https://acrabstracts.org/abstract/childhood-maltreatment-increases-risk-of-chronic-widespread-pain-counterfactual-analysis-of-uk-biobank/