Aletta Dijkstra1, Eelko Hak, Fanny Janssen. 1. Unit of PharmacoEpidemiology & PharmacoEconomics (PE(2)), Department of Pharmacy, University of Groningen, The Netherlands. a.dijkstra@rug.nl
Abstract
PURPOSE: Although current reviews of the use of spatial analysis in general epidemiologic research illustrate an important and well-established role in exploring and predicting health, its application has not been reviewed in the subspecialty field of pharmacoepidemiology. METHODS: We systematically reviewed the scientific literature to assess to what extent spatial analysis has been applied in pharmacoepidemiologic research and explored its potential added value. RESULTS: A systematic search in PubMed and Embase/MEDLINE yielded 823 potentially relevant articles; 45 articles met our criteria for review. The studies were reviewed on study objective, applied spatial methods and units of analysis, and author-reported added value of the geographic approach used. Of the 45 included studies, 34 (76%) reported a geographic research objective. Comparative spatial methods were most often used (n = 25; 56%). Eleven studies used spatial statistics (32%); cluster analysis (n = 5) and aggregate data analysis (n = 4) being most common. Mapping was done in 15 studies (33%). The most common added value reported was to aid the planning of health policies and interventions (n = 24; 53%). A minority of pharmacoepidemiologic studies used a geographic approach and the applied methods were less advanced compared with the broader field of epidemiology. CONCLUSIONS: Further advancements are needed to incorporate currently available spatial techniques to impact health care planning.
PURPOSE: Although current reviews of the use of spatial analysis in general epidemiologic research illustrate an important and well-established role in exploring and predicting health, its application has not been reviewed in the subspecialty field of pharmacoepidemiology. METHODS: We systematically reviewed the scientific literature to assess to what extent spatial analysis has been applied in pharmacoepidemiologic research and explored its potential added value. RESULTS: A systematic search in PubMed and Embase/MEDLINE yielded 823 potentially relevant articles; 45 articles met our criteria for review. The studies were reviewed on study objective, applied spatial methods and units of analysis, and author-reported added value of the geographic approach used. Of the 45 included studies, 34 (76%) reported a geographic research objective. Comparative spatial methods were most often used (n = 25; 56%). Eleven studies used spatial statistics (32%); cluster analysis (n = 5) and aggregate data analysis (n = 4) being most common. Mapping was done in 15 studies (33%). The most common added value reported was to aid the planning of health policies and interventions (n = 24; 53%). A minority of pharmacoepidemiologic studies used a geographic approach and the applied methods were less advanced compared with the broader field of epidemiology. CONCLUSIONS: Further advancements are needed to incorporate currently available spatial techniques to impact health care planning.
Authors: Keiki Hinami; Michael J Ray; Kruti Doshi; Maria Torres; Steven Aks; John J Shannon; William E Trick Journal: J Gen Intern Med Date: 2019-08-16 Impact factor: 5.128