Introduction Drug shortages are a public health concern which increase costs to health-systems and expose patients to adverse events. Stakeholders, including the US Food and Drug Administration (FDA), have prioritized reducing the incidence and impact of drug shortages. The FDA identifies shortages for “medically necessary” drugs with the intention of resolving their shortages faster. As the costs and patient risks due to shortages persist as long as the shortage continue, this study attempts to model the duration of drug shortages once they have occurred.
Methods Data on resolved drug shortages were captured and extracted on November 24, 2016 from the FDA Center for Drug Evaluation and Research (CDER) website, and the American Society of Health-Systems Pharmacists (ASHP) website. Descriptive statistics and univariate analyses are conducted. Drug price, marketing status, whether the shortage was “medically necessary,” the number of unique presentations, and the number of distinct manufacturers, along with other terms, were considered. Three modelling approaches were built on 75% of the data: ordinary least squares (OLS), Poisson regression, and negative binomial regression. Appropriate models were then compared for their accuracy in predicting the duration of the shortages in the remaining 25% of the data set. The model with lowest mean squared prediction error (MSPE) was considered the model that best fit the data. The selected model was rebuilt on the full data set.
Results The OLS model was not appropriate for drug shortage duration as it violated the assumption of normality. A Poisson model and negative binomial model were compared and the negative binomial model had the lowest MSPE. The negative binomial model was the model that best fit the data and retained fewer terms than the Poisson model. The parameter estimates for the negative binomial model built on the full data set were reported. The negative binomial model retained the number of manufacturers, the solution form, three routes of administration, one therapeutic category, as well as three manufacturers.
Conclusions Using these terms, the negative binomial model was the superior modelling approach for drug shortage duration. The FDA policy of labelling drugs as “medically necessary” was not shown to reduce the duration of those shortages and was not retained in the negative binomial model.
|Commitee:||Issa, Amalia, McGhan, William, Pontiggia, Laura|
|School:||University of the Sciences in Philadelphia|
|School Location:||United States -- Pennsylvania|
|Source:||DAI-A 78/11(E), Dissertation Abstracts International|
|Subjects:||Kinesiology, Public policy|
|Keywords:||Drug, Duration, Shortages|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be