Category: Pharmaceuticals

Pharmaceutical Supply Chains and Drug Shortages

By Hui Zhao📧

In Tutorials in Operations Research: Advancing the Frontiers of OR/MS: From Methodologies to Applications, 2023, 228–245. https://doi.org/10.1287/educ.2023.0258

Although the pharmaceutical industry is vital to the economy and the efficiency of pharmaceutical supply chains directly affects the quality and cost of patient care, pharmaceutical supply chains have been largely under-researched compared with the thriving research on medical services/hospital operations by the INFORMS community. At the same time, the pharmaceutical industry faces a unique economic and regulatory environment with many supply chain challenges. In this tutorial, I aim to provide a basic understanding of the complicated pharmaceutical supply chain and the challenges it faces. Using drug shortages (a persistent problem facing the pharmaceutical industry, government, and the society) and many other examples, I demonstrate that the richness and uniqueness of the pharmaceutical supply chains provide great opportunities for impactful research.

Keywords: Pharmaceutical supply chains; Drug shortages; Supply chain management; Public health

Stochastic Simulation Uncertainty Analysis to Accelerate Flexible Biomanufacturing Process Development

By Wei Xie, R. R. Barton📧, Barry L. Nelson, and Keqi Wanga

In European Journal of Operational Research, 2023, 310 (1): 238–248. https://doi.org/10.1016/j.ejor.2023.01.055

Motivated by critical challenges and needs from biopharmaceuticals manufacturing, we propose a general metamodel-assisted stochastic simulation uncertainty analysis framework to accelerate the development of a simulation model with modular design for flexible production processes. There are often very limited process observations. Thus, there exist both simulation and model uncertainties in the system performance estimates. In biopharmaceutical manufacturing, model uncertainty often dominates. The proposed framework can produce a confidence interval that accounts for simulation and model uncertainties by using a metamodel-assisted bootstrapping approach. Furthermore, a variance decomposition is utilized to estimate the relative contributions from each source of model uncertainty, as well as simulation uncertainty. This information can be used to improve the system mean performance estimation. Asymptotic analysis provides theoretical support for our approach, while the empirical study demonstrates that it has good finite-sample performance.

Keywords: Hybrid Simulation Model; Biomanufacturing Systems; Uncertainty Quantification (UQ); Sensitivity Analysis (SA); Gaussian Process (GP)

Not a Box of Nuts and Bolts: Distribution Channel Decisions for Specialty Drugs

By Liang Xu, Vidya Mani, and Hui Zhao📧

In Production and Operations Management, 2023, 32 (7): 2283–2303. https://doi.org/10.1111/poms.13973

One of the most important trends in the pharmaceutical industry is the rapid growth of specialty drugs. Specialty drugs, mostly bio-based, tend to be high-risk, high-priced, and more regulated than traditional drugs, resulting in unprecedented challenges in distribution. Such challenges lead to the emergence of specialty distributors (SDs), which, compared with traditional wholesalers (WSs), manage more controlled networks and carry a smaller variety of drugs. Using a unique dataset assembled from multiple proprietary and public data sources on transactions, inventory, and chargebacks for 419 specialty drugs across 11 manufacturers (including 8 of the top 15 pharmaceutical manufacturers),161 distributors, and129,911 POCs (point-of-care) in 2012-2015, we investigate unique factors associated with manufacturers’ SD usage (vs. WSs). We also develop a nested logit model to examine factors that drive manufacturers’ choice of specific SDs if they were to use an SD channel. We find that the restrictive access element of regulations (instead of regulations in general), downstream POC’s required drug variety, and distributors’ experience in providing critical value-added services for manufacturers (i.e., managing chargebacks) are associated with higher SD usage. Moreover, if a manufacturer were to use an SD channel, it is more likely to choose one if it has used this SD or the SD-affiliated WS before, or if this SD has better performance. WSs and SDs represent distinct approaches to balance channel accessibility and channel control. Our results provide important insights and guidance to manufacturers, regulators, and downstream POCs, while contributing to the limited empirical research on B2B distribution decisions.

Keywords: Distribution channel decision; Pharmaceutical industry; Specialty drugs; Regulations

When to Onshore? A Framework for the Manufacturing of Active Pharmaceutical Ingredients

By Emily Irvin, supervised by Robert A. Novack📧 (Thesis Supervisor) and John C. Spychalski📧 (Honors Advisor) (2022)

On February 24, 2021, President Joseph Biden issued Executive Order 14017 stating that he intends to strengthen “America’s Supply Chains.” This involves an in-depth look at risks within the supply chain, specifically related to the production of active pharmaceutical ingredients (APIs). The goal of this research is to analyze API production, understand why certain drugs are in shortage, and develop a framework for when the US Government should onshore critical APIs. This framework will also offer alternative solutions to onshoring, such as stockpiles, subsidies, and advanced manufacturing practices. It is a commonly accepted statistic that seventy to eighty percent of APIs are produced overseas, and copious research has been done to analyze the impact of disasters abroad on the United States’ ability to provide critical medication to its citizens. This thesis will build on prior research through interviews and case studies to develop a final framework that the government may use to determine the feasibility of onshoring API production. The utilization of this framework in manufacturing decisions will lower the risk of API shortages along the supply chain in times of future disaster.

Access the paper at Electronic Theses for Schreyer Honors College (ETDA) website here.

Outcome-Based Drug Reimbursement: The Solution to High Drug Spending?

By L. Xu, H. Li, and H. Zhao📧

In Manufacturing & Service Operations Management, 2022, 24 (4): 2029–2047. https://doi.org/10.1287/msom.2021.1051

Problem definition: The continuously soaring prices of new drugs and their uncertain effectiveness in clinical practice have put substantial risks on insurers/payers. To induce insurer coverage of their new drugs, manufacturers start to propose an innovative outcome-based reimbursement (OBR) scheme under which manufacturers refund insurers (and possibly patients) if the drugs fail to achieve a prespecified treatment target. We investigate the impact of OBR on insurers, manufacturers, and patients. Academic/practical relevance: Although OBR sounds intuitively appealing, its true impact is under much debate and depends particularly on the design of OBR. Our study sheds light on the optimal design of OBR and the debate around OBR, considering key trade-offs and key elements not covered in prior literature. Methodology: We develop a Stackelberg game under which the manufacturer designs a rebate scheme for its drug, either non-OBR or OBR, considering the trade-off between a favorable formulary position and the rebate provided. The insurer subsequently determines its formulary for the drug as well as other alternative drugs within the same disease category considering the trade-off between its spending and patient health benefits. Using data on 14 drugs treating a common disease, hyperlipidemia, we estimate through a multinomial logit model the demand of the 14 drugs and conduct counterfactual analyses on the impact of OBR. Results: Under the optimal OBR, the manufacturer lowers the insurer’s risk but inflates the wholesale price (hence, may not reduce insurer spending). OBR also induces a better formulary position for the manufacturer, which, hence, improves patient access to new drugs. Further, rebates to the insurer and patients affect demand through different mechanisms. Including patient rebates in OBR lowers patient expenses and increases drug demand but further increases insurer spending. Managerial implications: We demonstrate the structure of an optimal formulary and its application in practice. We caution insurers/payers who are seeking OBR to reduce their spending.

Keywords: Outcome-based reimbursement; Risk sharing; Drug pricing; Formulary design; Risk aversion; Multinomial logit models

Demand Forecasting with Supply Chain Information and Machine Learning: Evidence in the Pharmaceutical Industry

By X. Zhu, A. Ninh, H. Zhao📧, and Z. Liu

In Production and Operations Management, 2021, 30 (9): 3231–3252. https://doi.org/10.1111/poms.13426

Accurate demand forecasting is critical for supply chain efficiency, especially for the pharmaceutical (pharma) supply chain due to its unique characteristics. However, limited data have prevented forecasters from pursuing advanced models. Such problems exist even when long history of demand data is available because historical data in the distant past may bring little value as market situation changes. In the meantime, demands are also affected by many hidden factors that again require a large amount of data and more sophisticated models to capture. We propose to overcome these challenges by a novel demand forecasting framework which “borrows” time series data from many other products (cross-series training) and trains the data with advanced machine learning models (known for detecting patterns). We further improve performance of the cross-series models through various “grouping” schemes, and learning from non-demand features such as downstream inventory data across different products, information of supply chain structure, and relevant domain knowledge. We test our proposed framework with many modeling possibilities on two large datasets from major pharma manufacturers and our results show superior performance. Our work also provides empirical evidence of the value of downstream inventory information in the context of demand forecasting. We conduct prior and post-hoc field work to ensure the applicability of the proposed forecasting approach.

Keywords: Demand forecasting; Pharmaceutical; Machine learning

Managing Illicit Online Pharmacies Through Web Analytics and Predictive Models

By H. Zhao📧, S. Muthupandi, and S. Kumara

In Journal of Medical Internet Research, 2020, 22 (8): e17239. https://doi.org/10.2196/17239

Background: Online pharmacies have grown significantly in recent years, from US $29.35 billion in 2014 to an expected US $128 billion in 2023 worldwide. Although legitimate online pharmacies (LOPs) provide a channel of convenience and potentially lower costs for patients, illicit online pharmacies (IOPs) open the doors to unfettered access to prescription drugs, controlled substances (eg, opioids), and potentially counterfeits, posing a dramatic risk to the drug supply chain and the health of the patient. Unfortunately, we know little about IOPs, and even identifying and monitoring IOPs is challenging because of the large number of online pharmacies (at least 30,000-35,000) and the dynamic nature of the online channel (online pharmacies open and shut down easily). Objective: This study aims to increase our understanding of IOPs through web data traffic analysis and propose a novel framework using referral links to predict and identify IOPs, the first step in fighting IOPs. Methods: We first collected web traffic and engagement data to study and compare how consumers access and engage with LOPs and IOPs. We then proposed a simple but novel framework for predicting the status of online pharmacies (legitimate or illicit) through the referral links between websites. Under this framework, we developed 2 prediction models, the reference rating prediction method (RRPM) and the reference-based K-nearest neighbor. Results: We found that direct (typing URL), search, and referral are the 3 major traffic sources, representing more than 95% traffic to both LOPs and IOPs. It is alarming to see that direct represents the second-highest traffic source (34.32%) to IOPs. When tested on a data set with 763 online pharmacies, both RRPM and R2NN performed well, achieving an accuracy above 95% in their predictions of the status for the online pharmacies. R2NN outperformed RRPM in full performance metrics (accuracy, kappa, specificity, and sensitivity). On implementing the 2 models on Google search results for popular drugs (Xanax [alprazolam], OxyContin, and opioids), they produced an error rate of only 7.96% (R2NN) and 6.20% (RRPM). Conclusions: Our prediction models use what we know (referral links) to tackle the many unknown aspects of IOPs. They have many potential applications for patients, search engines, social media, payment companies, policy makers or government agencies, and drug manufacturers to help fight IOPs. With scarce work in this area, we hope to help address the current opioid crisis from this perspective and inspire future research in the critical area of drug safety.

Keywords: Classification; Illicit online pharmacies; Online pharmacy; Online traffic analysis; Web analytics

Inducing Compliance with Postmarket Studies for Drugs Under FDA’s Accelerated Approval Pathway

By Liang (Leon) Xu, H. Zhao📧, and Nicholas C. Petruzzi

In Manufacturing & Service Operations Management, 2021, 23 (1): 170–190. https://doi.org/10.1287/msom.2019.0822

Problem definition: In 1992, the Food and Drug Administration (FDA) instituted the accelerated approval pathway (AP) to allow promising drugs to enter the market based on limited evidence of efficacy, thereby permitting manufacturers to verify true clinical benefits through postmarket studies. However, most postmarket studies have not been completed as promised. We address this noncompliance problem. Academic/practical relevance: The prevalence of this noncompliance problem poses considerable public health risk, thus compromising the original purpose of a well-intentioned AP initiative. We provide an internally consistent and implementable solution to the problem through a comprehensive analysis of the myriad complicating factors and trade-offs facing the FDA. Methodology: We adopt a Stackelberg framework in which the regulator, which cannot observe the manufacturer’s private cost information or level of effort, leads by imposing a postmarket study deadline. The profit-maximizing manufacturer then follows by establishing its level of effort to invest in its postmarket study. In establishing its deadline, the regulator optimizes the trade-off between providing public access to potentially effective drugs and mitigating public health risks from ineffective drugs. Results: We develop a deadline-dependent user fee menu as a screening mechanism that establishes an incentive for manufacturer compliance. We show that its effectiveness in inducing compliance depends fundamentally on the enforceability of sanction, a drug-specific measure that indicates how difficult it is to withdraw an unproven drug from the market, and the drug’s success probability: The higher either is, the higher is the probability that the mechanism induces compliance. Managerial implications: We synthesize and distill the salient trade-offs and nuances facing the FDA’s noncompliance problem and provide an implementable solution. We quantify the value of the solution as a function of a drug’s success probability and enforceability. From a public policy perspective, we provide guidance for the FDA to increase the viability and effectiveness of AP.

Keywords: Health public policy; Drug approval policy; Pharmaceutical industry; Asymmetric information; Moral hazard

Demand Forecasting for Pharmaceuticals

By Sarah Elizabeth Heininger, supervised by Robert A. Novack📧 (Thesis Supervisor) and John C. Spychalski📧 (Honors Advisor) (2020)

This thesis analyzes the demand an animal pharmaceutical company sees and compares forecasting methods to find the most accurate one. Companies collect large amounts of data and a forecast model can utilize the collected data to predict future demand. There are various forecasting techniques that are better suited for different industries. In this research, error terms identify which forecast technique a company should be using by pinpointing which method produces the forecast closest to the historical data. In order to create this analysis, eighteen months of data were provided by Company Z and three forecast models were set up in Excel. The Excel model relies heavily on data input and subject matter experts from the company to ensure that the correct data is utilized in creating the forecasts. To conclude, this thesis provides information on which forecast method was most accurate for Company Z and how they will continue to monitor and update the forecast to see what their demand will be for the next three to six months. The findings for the animal pharmaceutical company in this thesis identified that exponential smoothing is the best forecasting method to utilize. The major findings confirm that for a company that sees a relatively low amount of seasonality and year after year sees the same trends, exponential smoothing creates the most accurate demand forecast.

Access the paper at Electronic Theses for Schreyer Honors College (ETDA) website here.

Drone Optimization for Medical Use in Sub-Saharan Africa

By Katie Gustas, supervised by Robert A. Novack📧 (Thesis Supervisor) and John C. Spychalski📧 (Honors Advisor) (2020)

Sub-Saharan Africa is struggling to provide access of life-saving medical resources to its citizens, creating widespread health concerns. The region has seen a high maternal mortality rate, correlating to the lack of medical accessibility. Specifically, there is a lack of adequate blood supply for mothers suffering from post-partum hemorrhaging. By conducting internet research provided by company executives and area experts, as well as public data sources, opportunities in drone transportation of blood are being explored. Industry leader in unmanned aerial vehicles (UAVs), Zipline, is evaluated for its successful implementation in the country of Rwanda. Throughout this thesis the reasons for the company’s success are identified and areas for potential improvement are explored. A centralized supply chain, lack of adequate transportation routes and limiting physical conditions, a localized workforce, supporting government regulations, and efficient technology all were found to contribute to Zipline’s success in saving lives through drone transportation in Rwanda. The lessons learned from Zipline in Rwanda are applicable across sub-Saharan Africa and help to conceptualize the positive impact drone usage can create in various regions of the continent through the same or similar processes. The recommended strategies provide a sustainable business opportunity, and also serve as guidelines to acting on a humanitarian issue in a socially responsible manner.

Access the paper at Electronic Theses for Schreyer Honors College (ETDA) website here.