[Concordance in the registry of dementia among the main sources of clinical information]. / Concordancia del registro de demencia en las principales fuentes de información clínica.

BACKGROUND: The objective of this work was to analyse the concordance in the registry of dementia among the main sources of clinical information, with the aim of determining their usefulness for epidemiological and clinical research. METHODS: Descriptive study of patients assigned to the Aragon Health Service in 2010 (n=1,344,891). DATABASES USED: (i)the pharmacy billing database (n=9,392); (ii)Primary Care electronic health records (EHR) (n=9,471), and (iii)the hospital minimum basic data set (n=3,289). When studying the concordance of the databases, the group of patients with a specific treatment for dementia (i.e., acetylcholinesterase inhibitors and/or memantine) was taken as the reference. RESULTS: The diagnosis in Primary Care was missing for 47.3% of patients taking anti-dementia drugs. The same occurred with 38.3% of dementia patients admitted to hospital during the study year. Among patients with a diagnosis of dementia in the EHR, only half (52.3%) was under treatment for this condition. This percentage decreased to 34.4% in patients with the diagnosis registered in the hospital database. CONCLUSIONS: The weak concordance in the registry of the dementia diagnosis between the main health information systems makes their use and analysis more complex, and supports the need to include all available health data sources in order to gain a global picture of the epidemiological and clinical reality of this health condition.

Concordancia del registro de demencia en las principales fuentes de información clínica / Concordance in the registry of dementia among the main sources of clinical information

Introducción. El objetivo de este trabajo es analizar la concordancia del registro del diagnóstico de demencia entre las principales fuentes de información clínica, con el fin de explorar su utilidad en investigación epidemiológica y clínica. Material y métodos. Estudio descriptivo de pacientes adscritos al Servicio Aragonés de Salud en 2010 (n=1.344.891). La identificación de pacientes con demencia se realizó en 3 bases de datos distintas: a)cinta de facturación de farmacia (n=9.392); b)historia clínica electrónica (HCE) de atención primaria (n=9.471), y c)conjunto mínimo básico de datos (CMBD) hospitalarios (n=3.289). Para analizar la concordancia entre las diferentes bases de datos se toman como referencia los pacientes con tratamiento específico para demencia (p.ej., inhibidores de la acetilcolinesterasa y/o memantina). Resultados. En el 47,3% de los pacientes con tratamiento específico no constaba el diagnóstico en la HCE. Lo mismo sucedió en el 38,3% de los pacientes con demencia que ingresaron en el hospital. De entre los pacientes con diagnóstico de demencia en la HCE, la mitad (52,3%) tenían prescrito tratamiento específico. Este porcentaje descendió hasta un 34,4% en el caso de los pacientes con el diagnóstico reflejado en el informe de alta hospitalaria. Conclusiones. Existe una escasa concordancia en el registro de este diagnóstico en las principales fuentes de información asistenciales, lo que cuestiona la utilidad de su empleo de forma aislada para estudios epidemiológicos y obliga a consultar las distintas bases de datos disponibles si queremos obtener una visión global de la realidad epidemiológica y clínica de esta enfermedad (AU)

Background. The objective of this work was to analyse the concordance in the registry of dementia among the main sources of clinical information, with the aim of determining their usefulness for epidemiological and clinical research. Methods. Descriptive study of patients assigned to the Aragon Health Service in 2010 (n=1,344,891). Databases used: (i)the pharmacy billing database (n=9,392); (ii)Primary Care electronic health records (EHR) (n=9,471), and (iii)the hospital minimum basic data set (n=3,289). When studying the concordance of the databases, the group of patients with a specific treatment for dementia (i.e., acetylcholinesterase inhibitors and/or memantine) was taken as the reference. Results. The diagnosis in Primary Care was missing for 47.3% of patients taking anti-dementia drugs. The same occurred with 38.3% of dementia patients admitted to hospital during the study year. Among patients with a diagnosis of dementia in the EHR, only half (52.3%) was under treatment for this condition. This percentage decreased to 34.4% in patients with the diagnosis registered in the hospital database. Conclusions. The weak concordance in the registry of the dementia diagnosis between the main health information systems makes their use and analysis more complex, and supports the need to include all available health data sources in order to gain a global picture of the epidemiological and clinical reality of this health condition (AU)