Endocrine Abstracts

Anonymised therapeutic data has been available for all patients in NI across 364 general practices since 2008. The database records age, gender, trusts and postcodes, which allows for spatial deprivation analyses. Patients were categorised in deciles (1-most deprived, 10-least) according to published criteria. We examined carbimazole (CBZ) and propylthiouracil (PTU) prescriptions over 10 years (2010 to 2019). Patients treated definitively with surgery/radioiodine are invariably pre-treated with anti-thyroid medication in NI and so are captured in this database. The median age of the population in NI in mid-2019 was 38.9 years (vs 43.7y for European Union). The number of patients in NI prescribed CBZ/PTU was ~4000 patients/year or ~0.2% population (3,760 in 2010 with population of 1.80million; 4,025 in 2019 with pop of 1.89million). This is lower than published prevalence estimates (including subclinical disease) of ~0.8-1%. There was a modest reduction (4%) over 10 years from what would have been expected, when controlled for change in demographic structure. In 2019, females accounted for 80% of patients. The standardised incidence ratio was maximal in 35-44 year group. Patients aged 0-24 years accounted for 32% of the general pop vs 3% (113) of patients on treatment. CBZ and PTU represented 93% and 7% of prescription items, respectively. Of those prescribed PTU, 89% were female. The proportion of patients on anti-thyroid medications was modestly but persistently higher in the more deprived cohort deciles (Z testing P<0.05). This is in keeping with a previous report from Birmingham describing deprivation as an independent risk for thyroid dysfunction among 20 general practices. The reason(s) for changing medical treatment rates and deprivation differences require exploration (eg definitive treatment rates, environmental factors). This database offers a caveated resource for following overt and treated hyperthyroidism rates, generating hypotheses and commissioning services such as transition clinics.