Premium
Automated databases and the need for fieldwork in pharmacoepidemiology
Author(s) -
Hemmelgarn Brenda,
Blais Lucie,
Collet JeanPaul,
Ernst Pierre,
Suissa Samy
Publication year - 1994
Publication title -
pharmacoepidemiology and drug safety
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.023
H-Index - 96
eISSN - 1099-1557
pISSN - 1053-8569
DOI - 10.1002/pds.2630030506
Subject(s) - medicine , pharmacoepidemiology , confounding , asthma , medical record , medical diagnosis , cohort , cohort study , medical emergency , database , family medicine , emergency medicine , pediatrics , pathology , nursing , computer science , medical prescription
Abstract The Saskatchewan Asthma Epidemiologic Project (SAEP) is used as an example to describe the fieldwork necessary to supplement computerized data sources. Steps were implemented beyond the computerized sources of information to identify case events, validate outcome diagnoses, and obtain clinical information on potential confounding variables. Validation of the cause of death as actually being a consequence of asthma was particularly beneficial and prevented the misclassification of 11 of the 165 deaths identified from a geographically defined cohort of asthmatic patients in the province of Saskatchewan. Hospital medical records and physician questionnaires were used in the SAEP as an external source of information in an attempt to measure and control asthma severity, a potential confounding variable. The availability of a variety of variables which contributed to the measure of asthma severity from hospital medical records and physician questionnaires was highly variable, ranging from 4 per cent to 86 per cent. A considerable amount of time and expense was associated with obtaining additional information beyond the automated databases.