Impact of Pharmacovigilance Interventions Targeting Diclofenac on Pain Medication Use in the United Kingdom

Background: In 2013, the European Medicines Agency introduced new regulatory risk minimisation measures (RMMs) addressing the elevated risk of cardiovascular events associated with the use of diclofenac. The RMMs included safety warnings cautioning about prescribing diclofenac to patients with specific risk factors and new contraindications.

Objectives: To assess the impact of RMMs targeting diclofenac on analgesic drug use based on patients’ risk profile.

Methods: CPRD GOLD was used to select patients diagnosed with musculoskeletal disorders during 2010–2019 in the UK who were not receiving pain treatment prior to the date of diagnosis. The main outcome was quarterly treatment incidence rates per person-years per quarter (PYpQ). Incidence rates were assessed for three patient risk profiles: (1) patients without RMM specified risk factors, (2) Patients with risk factors (e.g., hypertension, hyperlipidaemia, diabetes mellitus, and smoking) and (3) Patients with contraindications (prior cardiovascular events). Impact of RMMs on analgesic drug use was assessed using Interrupted time series analysis.

Results: Out of 41,109 patients included 50.8% had ≥ 1 risk factor, and 10.0% had ≥ 1 contraindication for diclofenac. Diclofenac was the initial painkiller prescribed following diagnosis for 3.5% of the patients, NSAIDs for 24.1%, paracetamol for 6.0%. In patients with risk factors and contraindications, the trend of quarterly incidence remained relatively stable prior to RMMs: 0.001 (95% CI < 0.001–0.003) and −0.004 (95% CI −0.01–0.003) per 100,000 PYpQ, respectively; while in patients with no risk, it was decreasing prior to RMMs (−0.14 per 100,000 PYpQ, 95% CI −0.18 to −0.12). No significant changes were seen following RMMs implementation in patients with risk factors. For patients with contraindications, quarterly incidence of systemic diclofenac dropped immediately by 0.03 (95% CI −0.04 to −0.01) per 100,000 PYpQ. Quarterly incidence in patients with no associated risk decelerated significantly by 0.05 (95% CI 0.03–0.06) per 100,000 PTiQ. Incidence rates for all other treatments were increasing for patients with ≥ 1 risk factor or ≥ 1 contraindication and decreasing for patients without additional risk. However, these changes did not seem to be associated with RMMs.

Conclusions: Decreasing trend in diclofenac incidence rate was already seen prior RMMs. The RMM had an impact, mainly on those patients suffering from musculoskeletal diseases without risk factors indicated in the diclofenac RMMs. Impact on prescribing of other alternative analgesics was not seen. Fewer patients with additional risk were treated with any analgesic after RMMs (during the first three months following diagnosis).