The Use of Pharmacoepidemiological Databases to Assess Disease Burdens: Application to Diabetes
Assessing the disease burden of diabetes in general populations is important for planning and monitoring public health initiatives. Prevalence is the relative frequency of diseased subjects in a population, and is the default measure of disease burden. T
- PDF / 275,816 Bytes
- 14 Pages / 439.37 x 666.142 pts Page_size
- 34 Downloads / 150 Views
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 672
2
What is a Pharmacoepidemiologic Database? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 674
3
Gross Volume Studies: Diabetes Prevalence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 675
4 4.1 4.2 4.3
Individual Level Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 676 Individual Treatment Status Ascertainment Based on Decision Rules . . . . . . . . . . . . . . 676 Waiting Time Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 679 Beyond Prevalence: Incidence and Mortality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 679
5
Summary and Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 682 Summary Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 683 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 683
#
Springer Science+Business Media LLC 2010 (USA)
672
39
The Use of Pharmacoepidemiological Databases to Assess Disease Burdens
Abstract: Assessing the disease burden of diabetes in general populations is important for planning and monitoring public health initiatives. > Prevalence is the relative frequency of diseased subjects in a population, and is the default measure of disease burden. Traditional epidemiological methods for estimating prevalence are however costly, and so it has been suggested to exploit the advent of large scale pharmacoepidemiological databases for estimating diabetes prevalence, as these databases are available at nominal costs. In this paper we review how information on either gross volume sales of anti-diabetic agents or on prescription redemptions have been used to estimate the prevalence of diabetes, as well as > incidence and > mortality among treated. For studies based on gross volume sales, the critical prerequisite is to obtain valid estimates of average drug consumption among treated diabetics. Studies based on individual level information on prescription redemptions requires determination of treatment status, either explicitly via decision rules or implicitly via a stochastic model. For the latter, the > Waiting Time Distribution has been suggested. While prevalence estimates based on > pharmacoepidemiologic databases are found to underestimate overall prevalence of diabetes – by definition, they only concern diabetics treated pharmacologically – good agreement is found with respect to trend estimates of both prevalence, incidence, and mortality among diabetics. Pharmacoepidemiologic da
Data Loading...
