Abstract
A fundamental principle of good scientific conduct is that the data used for research are validated. This also applies to epidemiologic studies, including pharmacoepidemiologic studies. Therefore, validation studies are needed to validate data from registries for use in epidemiologic research [1].
Highlights
A fundamental principle of good scientific conduct is that the data used for research are validated
This article presents how to apply or calculate the positive predictive value (PPV) based on validation studies, when the validation study uses another population than the population of interest
We acknowledge that other more advanced methods, such as Bayesian or likelihood-based methods may be preferable to the RoganGladen estimator [9,11], but the simplicity of the RoganGladen estimator approach may be appealing
Summary
The PPV depends on the true prevalence of the condition in the studied population, and the sensitivity and specificity of the condition in the registry: [4]. The 2 × 2 table is used to explain the different measures associated with accuracy for dichotomous measures These measures include the positive predictive value (PPV), the negative predictive value, sensitivity, specificity, and observed ( called apparent) and true prevalence (Table 1). A high PPV ensures that the classification of patients into a given group is correct, while a low PPV results in the opposite The aim of this commentary is to briefly illustrate how validation studies can be used to estimate the PPV of registry data for use in epidemiologic research. It is only suitable to apply the PPV from an external validation study, when the population in the validation study resembles the studied population, especially with regard to the prevalence of the condition; this assumption is seldom reasonable
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.