Unplanned medication discontinuation as a potential pharmacovigilance signal: a nested young person cohort study

BACKGROUND: Because of relatively small treatment numbers together with low adverse drug reaction (ADR) reporting rates the timely identification of ADRs affecting children and young people is problematic. The primary objective of this study was to assess the utility of unplanned medication discontinuation as a signal for possible ADRs in children and young people. METHODS: Using orlistat as an exemplar, all orlistat prescriptions issued to patients up to 18 years of age together with patient characteristics, prescription duration, co-prescribed medicines and recorded clinical (Read) codes were identified from the Primary Care Informatics Unit database between 1st Jan 2006-30th Nov 2009. Binary logistic regression was used to assess association between characteristics and discontinuation. RESULTS: During the study period, 79 patients were prescribed orlistat (81% female, median age 17 years). Unplanned medication discontinuation rates for orlistat were 52% and 77% at 1 and 3-months. Almost 20% of patients were co-prescribed an anti-depressant. One month unplanned medication discontinuation was significantly lower in the least deprived group (SIMD 1–2 compared to SIMD 9–10 OR 0.09 (95% CI0.01 – 0.83)) and those co-prescribed at least one other medication. At 3 months, discontinuation was higher in young people (≥17 yr versus, OR 3.07 (95% CI1.03 – 9.14)). Read codes were recorded for digestive, respiratory and urinary symptoms around the time of discontinuation for 24% of patients. Urinary retention was reported for 7.6% of patients. CONCLUSIONS: Identification of unplanned medication discontinuation using large primary care datasets may be a useful tool for pharmacovigilance signal generation and detection of potential ADRs in children and young people.