Endocrine Abstracts

The Thyroid Epidemiology and Audit Research Study (TEARS) involves electronic linkage of various clinical datasets to establish a comprehensive thyroid dataset (“TEARS database”). This includes linking a thyroid follow-up register, all regional biochemistry results, regional prescribing data, a radioactive iodine database, hospital admissions and the General Register of Deaths, all of which use a common patient identification number. Specific algorithms were used to determine patients with primary hypothyroidism, and those who had hyperthyroidism originally. Direct case-note assessment identified that the electronic system was accurate with a 98% positive predictive value for hyperthyroidism and a 96% PPV for hypothyroidism. TEARS demonstrated that the point prevalence of treated hyperthyroidism increased from 0.51% in 1993 to 0.63% in 1996. The prevalence of hypothyroidism increased from 2.2% to 3.0% in the same period. In Tayside the incidence of hyperthyroidism was 6-fold less than that for hypothyroidism (approximately 50 vs 300 per 100,000 people).

Mortality and morbidity data were examined from 1993 to 2001 for the TEARS dataset i.e. patients treated and stabilised from either hyper- or hypothyroidism. This included 15,889 patients with primary hypothyroidism and 3,888 with initial hyperthyroidism. All cause mortality and stroke deaths were not increased for either condition. Cardiovascular mortality was increased in patients following hypothyroidism (SMR 1.15; 1.02–1.31 (95% confidence interval)) but not for hyperthyroid patients. Cardiovascular morbidity was increased in treated hypothyroidism (SIR 1.25; 1.16–1.35), due to ischaemic heart disease (1.23; 1.1–1.36) and dysrhythmias (1.32; 1.11–1.57). Overall cardiovascular disease and strokes were not increased in treated stabilised hyperthyroidism, despite an increase in dysrhythmias (2.71; 1.63–4.24). The incidence of these conditions did not change over the first 5 years of follow-up.

TEARS is a useful epidemiological resource which can be used to address specific questions on treated thyroid conditions.