Chronic multisite pain is associated with an increased risk of incident fracture at all major sites: a prospective cohort study — ASN Events

Chronic multisite pain is associated with an increased risk of incident fracture at all major sites: a prospective cohort study (#15)

Nikola Markovic 1 , Fatemeh Vazirian 2 , Michael Thompson 1 , Feng Pan 2
  1. Endocrinology, Royal Hobart Hospital, Hobart, Tasmania, Australia
  2. Menzies Institute for Medical Research, Hobart , Tasmania

Multisite pain (MSP) is associated with adverse health outcomes. Prior studies have linked MSP to self-reported fracture. However, the relationship between MSP and incident fracture using hospitalisation data has not been investigated.

Prospective cohort analysis of 350,665 participants (mean age 56.5 years) from the UK Biobank with baseline data on chronic pain and covariates were included. Chronic pain, defined as lasting more than 3 months, was self-reported across anatomical sites and categorised by number of pain sites (none, 1-2 sites, 3-4 sites and ‘pain all over the body’). Incident fractures (vertebral, hip, major, non-vertebral, and any fracture) were identified using self-report and hospitalisation records. Multivariable Cox regression models, with participants without chronic pain the reference group, were used to examine the association between number of chronic pain sites and incident fractures. Follow-up was calculated from baseline assessment to first fracture event, death, or study censoring. Sensitivity analyses for competing risk of death and loss to follow up were performed using Fine-Gray proportional sub hazards models and inverse probability weighting, respectively.


Over a median follow-up of 13.28 years, 17,237 incident fractures occurred (1,262 vertebral, 3,258 hip, 16,233 non-vertebral, and 9,594 major fractures). Increasing number of chronic pain sites was associated with significantly more incident fractures, including vertebral (ptrend=1.36), hip (ptrend=1.12) , major (ptrend=1.07) and any (ptrend=1.05) fracture in a dose-response fashion. Pain all over the body was associated with an increased risk of vertebral (HR 2.17; 95% CI 1.54-3.06), hip (HR 1.53; 95% CI 1.21-1.93), major (HR 1.35, 95% CI 1.13-1.62), non-vertebral (HR 1.19; 95% CI 1.03-1.38) and any (HR 1.24; 95% CI 1.08-1.43) fracture. All sensitivity analyses yielded consistent results.

 
Chronic MSP was independently associated with incident fracture, particularly hip and vertebral fracture. Incorporating MSP into fracture risk assessment may improve risk stratification.