Background Both epidemiology and planning of health care puts high demands on the ability to record and monitor data. Disease registries have contributed significantly the previous years, however, an enormous amount of healthcare data is spread among hospitals, primary care providers, researchers, health insurers, with each of these usually acting as a silo, preventing effective use of data.
Objectives Since Greece is among the first countries that developed an extensive electronic prescription system, we aimed to identify all patients with prescribed pharmacological treatment for RA, SLE and SSc among 7.742.629 Greek citizens (72% of the population, >99% Caucasians) who were included in the system during the first semester of 2014.
Methods The database of the electronic prescription platform of the Greek National Organization for Provision of Healthcare Services (EOPYY) was used to provide analytics on these patients (date of birth and gender based on the unique citizens' social security numbers and the relevant ICD-10 codes). Permission for use of anonymized data was obtained by the administration of EOPYY together with the positive recommendation of the General Secretariat for Public Health of the Ministry of Health of Greece, in accordance to the national legislation on the Protection of Individuals and Personal Data.
Results This “Big Data” analysis revealed that RA prevalence is 0.84%, SLE prevalence is 0.075% and SSc prevalence is 0.016%. Female:male ratio is approximately 4:1 in RA, and 9:1 in both SLE and SSc, with slight differences across age groups (15-24, 25-34, 35-44, 45-54, 55-64, 65-74, 75+). The peak RA prevalence is observed after the age of 75 years; in contrast, the peak prevalence of SLE and SSc is observed between 45-54 and 55-64 years, respectively, which is compatible with the earlier/higher mortality of these patients compared to RA. The highest female preponderance (94%) is noted in patients older than 74 years with SSc, supporting the previously suspected earlier mortality of men compared to women in SSc.
Conclusions These data provide reliable estimates of the epidemiology of both common and rare autoimmune rheumatic diseases. Analysis of such large databases overrides any incorrect diagnosis-associated limitations that an electronic prescription system may have. However, a proportion of patients may be missed because of mild disease or not receiving (prescribed) treatment, therefore, the true prevalence is likely to be higher than that calculated. Further analyses of data deriving from the second semester of 2014 which covers 95% of the Greek population should confirm these results and reveal the distinct pharmacological approaches, including biologic agents, as well as the co-morbidities. These analyses are in progress.
Disclosure of Interest None declared