Data linkage and pain medication in people with cerebral palsy: a cross-sectional study

Aim To explore data linkage and pain medication as a proxy for pain, to assess differences in pain medication between the cerebral palsy (CP) and the general populations, and to identify factors associated with pain medication in CP. Method This cross-sectional study linked the Northern Ireland CP Register and two administrative health care databases for people resident in Northern Ireland born between 1981 and 2008. Pain medication as a proxy was validated by replicating analyses from the Study of Participation of Children with Cerebral Palsy Living in Europe (SPARCLE) studies. Logistic regression compared pain medication in the CP and general populations. Multi-level regression models assessed factors associated with pain medication in the CP cohort. Results The sample size was 701 075, of whom 1430 (0.2%) were people with CP. There were 358 969 males and 340 677 females in the general population, and 810 males and 620 females in the CP population, with an age range of 4 to 31 years in both groups. The validation exercise produced results similar to the SPARCLE studies. More people with CP received pain medication (61% vs 50.9%) and had twice the odds of being prescribed opioid analgesics (odds ratio [OR]=2.81, 95% confidence interval [CI] 2.32-3.40). Among those with CP, the odds of being prescribed pain medication were higher for: females (OR=1.34, 95% CI 1.06-1.70), younger age (OR=1.60, 95% CI 1.02-2.51), Gross Motor Function Classification System level V (OR=2.60, 95% CI 1.52-4.47), seizures (OR=2.55, 95% CI 1.68-3.87), and higher deprivation score (OR=2.06, 95% CI 1.41-3.24). Interpretation Pain medication is an effective proxy for pain. More people with CP were prescribed pain medication than the general population. Pain medication for people with CP is not only dependent on physiological and clinical characteristics, but also environmental factors.