Quality of registration of dementia diagnosis in primary care: The situation in Spain in 2002-2011

Objective: To ascertain the diagnosis- associated with specific treatment for dementia in the Primary Care Electronic Clinical Record (PC-ECR) and to analyse the factors associated with the quality of registration. Methods: Descriptive study of patients taking cholinesterase inhibitors or memantine registered in Database for pharmacoepidemiological research in PC (BIFAP) 2011: 24,575 patients between 2002 and 2011. Diagnoses associated with first prescription of these drugs were grouped into 5 categories: "dementia", "memory impairment", "dementia-related diseases", "intercurrent processes" and "convenience codes". We calculated the prevalence of each category by age and sex for each study year (95% CI) and analysed the associations and trend for 2002-2011 using difference in proportions in independent samples and binary logistic regression. Results: A code of "dementia" was associated with first prescription in 56.5% (95% CI: 55.8-57.1) of patients. It was higher in women [OR 1.09 (95% CI: 1.03-1.15)] and with increasing follow-up time [OR 1.07 (95% CI: 1.06-1.08) for each year of follow-up]. "Convenience codes" [16.3% (95% CI: 15.8-16.7)] were coded more frequently in women and in those >= 80 years; "Memory impairment" [12.4% (95% CI: 12.0-12.8)], "related diseases" [4.6% (95% CI: 4.4-4.8)] and "intercurrent processes" [10.3% (95% CI: 9.9-10.6)] were used more in men and in persons < 80 years. Between 2002 and 2011 improved the use of "convenience codes". Conclusions: Almost half of the patients taking cholinesterase inhibitors or memantine do not have a diagnosis of dementia registered in their PC-ECR. Registration improves with increasing time of follow-up. Improvements are needed in'the PC-ECR, adequate care coordination, and proactive approach to increase the quality of dementia registration. (C) 2014 Elsevier Espana, S.L.U.