Aims We examined the impact of off-label and unlicensed (OLUL) prescribing on adverse drug reaction (ADR) risk in paediatric inpatients. We hypothesised that OLUL prescribing may be a risk factor for ADRs in paediatric inpatients.

Methods This was a nested case-control study within a 12 month prospective cohort study of paediatric inpatients. Cases were children who had experienced at least one probable or definite ADR during their admission, controls were children who had not experienced any probable or definite ADRs during their admission. An OLUL category was assigned to each medicine administered by comparison of its use with the terms of its marketing authorisation found in the SmPC¹ using a system based on the categories proposed by Turner et al. ² The OR of OLUL medicines being implicated in a probable or definite ADR was calculated. A multivariate Cox proportional hazards regression model was fit to the data to assess the influence that off label and unlicensed medicine use had on the hazard of an ADR occurring.

Results Of the 10 699 medicine courses administered to 1388 patients, 10 145 could be categorised. 68.8% were approved, 23.7% were off-label and 7.5% were unlicensed. 6.2% of all approved medicine courses were implicated in at least one ADR compared with 12.4% of off-label medicine courses and 14.9% of unlicensed medicine courses. The OR of an OLUL medicine being implicated in an ADR when compared with an approved medicine course was 2.3 (95% CI 2.0 to 2.6). Medicines licensed in children but given to a child below the minimum age or weight had the greatest risk of being implicated in an ADR when compared to approved medicines (OR 3·5, 95% CI 2.8 to 4.4). Fentanyl via any route excluding epidural had the greatest proportion of courses implicated (49.3%) and 98.6% of courses were off-label. Epidural fentanyl had 43.4% of courses implicated, 100% of courses were unlicensed. Multivariate analysis showed that age on admission and receipt of a general anaesthetic each had a significant effect on ADR risk. Each additional OLUL medicine given per day significantly increased the risk of an ADR (HR 1.267 95% CI 1.201 to 1.336 p<0.001). Similarly, each unit increase in the number of approved medicines in a single day also significantly increased the hazard of an ADR (HR 1.217 95% CI 1.171 to 1.263 p<0.001).

Conclusions The increase in ADR risk associated with OLUL prescribing must be considered in the context of the types of medicine implicated, the frequency of their use, the frequency with which they are implicated in ADRs and the frequency with which they are categorised as off-label or unlicensed. Some of the most commonly implicated medicines in this study are frequently OLUL (eg, fentanyl). Both the number of approved AND the number of OLUL medicines administered significantly increase the hazard of an ADR.