"Burden of osteoporotic fractures in primary health care in Catalonia (Spain): a population-based study".
Pagès-Castellà A., Carbonell-Abella C., Avilés FF., Alzamora M., Baena-Díez JM., Laguna DM., Nogués X., Nogués X., Díez-Pérez A., Prieto-Alhambra D.
BACKGROUND: Knowledge on the epidemiology of non-hip fractures in Spain is limited and somewhat outdated. Using computerized primary care records from the SIDIAP database, we derived age and sex-specific fracture incidence rates for the region of Catalonia during the year 2009. METHODS: The SIDIAP database contains quality-checked clinical information from computerized medical records of a representative sample of >5,800,000 patients (80% of the population of Catalonia). We conducted a retrospective cohort study including all patients aged ≥50 years, and followed them from January 1 to December 31, 2009. Major osteoporotic fractures registered in SIDIAP were ascertained using ICD-10 codes and validated by comparing data to hospital admission and patient-reported fractures records. Incidence rates and 95% confidence intervals were calculated. RESULTS: In total, 2,011,430 subjects were studied (54.6% women). Overall fracture rates were 10.91/1,000 person-years (py) [95%CI 10.89-10.92]: 15.18/1,000 py [15.15-15.21] in women and 5.78/1,000 py [5.76-5.79] in men. The most common fracture among women was wrist/forearm (3.86/1,000 py [3.74-3.98]), while among men it was clinical spine (1.25/1,000 py [1.18-1.33]). All fracture rates increased with age, but varying patterns were observed: while most of the fractures (hip, proximal humerus, clinical spine and pelvis) increased continuously with age, wrist and multiple rib fractures peaked at age 75-80 and then reached a plateau. CONCLUSIONS: Our study provides local estimates of age, sex and site-specific fracture burden in primary health care, which will be helpful for health-care planning and delivery. A proportion of fractures are not reported in primary care records, leading to underestimation of fracture incidence rates in these data.