RESUMEN
Electronic patient records remain a rather unexplored, but potentially rich data source for discovering correlations between diseases. We describe a general approach for gathering phenotypic descriptions of patients from medical records in a systematic and non-cohort dependent manner. By extracting phenotype information from the free-text in such records we demonstrate that we can extend the information contained in the structured record data, and use it for producing fine-grained patient stratification and disease co-occurrence statistics. The approach uses a dictionary based on the International Classification of Disease ontology and is therefore in principle language independent. As a use case we show how records from a Danish psychiatric hospital lead to the identification of disease correlations, which subsequently can be mapped to systems biology frameworks.
Asunto(s)
Recolección de Datos/métodos , Minería de Datos/métodos , Registros Electrónicos de Salud , Análisis por Conglomerados , Estudios de Cohortes , Comorbilidad , Biología Computacional/métodos , Humanos , Clasificación Internacional de Enfermedades , Reproducibilidad de los ResultadosRESUMEN
A large body of research suggests a relationship between maternal influenza and the development of schizophrenia in the adult offspring. Some researchers, however, have questioned this association. A study by Crow and Done (1992) asserts that prenatal exposure to influenza does not cause schizophrenia. The methodology employed by Crow and Done may account for their null findings. Crow and colleagues assessed influenza by asking mothers at the time of birth to recall influenza infections experienced during pregnancy. Such retrospective recall may bias reporting. We assessed influenza symptoms during pregnancy in a group of 136 mothers at the twenty-fifth week of pregnancy, and again one or two days after birth. We compared accounts of influenza at the twenty-fifth week to recollection of influenza after birth. Results suggest that mothers tend to under-report infections when recalling infections after birth. Retrospective assessment of influenza symptoms at birth may be an inaccurate method of assessing influenza during pregnancy.