Keywords

1 Introduction

Trust in the integrity of pharmaceutical products is one of the cornerstones of sustainable international pharma networks [1]. This trust is increasingly threatened by counterfeit medicine, which causes significant economic and human damage worldwide making it very difficult to achieve the UN Sustainable Development Goal 3 (“Ensure healthy lives and promote well-being for all at all ages”) [2]. The main reason for the occurrence of non-legitimate medicine is the complexity of heavily fragmented pharma networks. Throughout all value-adding processes, numerous transfer points between various actors in the logistics network exist with different informational interfaces leading to highly fragmented and inconsistent data flows [3]. This reduces transparency and trust and significantly complicates authenticity checks throughout the logistics network [4, 5]. Although this problem is known for decades, the need for a technology that addresses compatibility and credibility issues while allowing the alignment with legal requirements is still imminent.

Since blockchain combines the properties of tamper-proof data storage and secure information transfer, its deployment as a traceability solution in logistics networks is discussed in several fields [6]. Also, in the pharma industry several ideas emerge that promise a high potential for the industry. However, there is a variety of conditions that should be met to ensure that the technological benefits outweigh the techno-logical shortcomings [7]. In general, blockchain allows meeting all legal requirements that are mandatory for pharmaceutical traceability solutions [8]. As a result, the technology seems to be the ideal foundation for automating transactions and increase operational efficiency [9]. Despite all those theoretical benefits, adoption in pharma logistics networks is slow and only few blockchain solutions in the pharma logistics network exceed a prototypical implementation [9]. The vast amount of blockchain research focuses on the identification and analysis of blockchain use cases, but research regarding the transfer from science to practice is scarce. In this regard, most blockchain literature is not industry-specific and therefore doesn’t apply the proprietary mechanics and limitations of the pharmaceutical industry to their respective models. Furthermore, the specific challenges of the pharma logistics network and respective mitigation strategies remain unclear [10]. To contribute to this under-investigated area and to support future managers in the pharma industry, this study seeks to develop an implementation framework that outlines industry-specific implementation barriers, mitigation measures and enablers of blockchain implementations to tackle drug counterfeiting in pharma logistics networks. More specifically we aim at investigating the following research question: What are the most relevant enablers and barriers as well as implementation strategies for adopting blockchain in the pharma logistics network and how do they interconnect and interdepend?

To contribute to the RQs, the study analyses nine interviews with experts from the industry applying Grounded Theory (GT) methodology.

2 Theoretical Background: The Pharma Logistics Network and the Problem of Counterfeit Drugs

The term counterfeit medicines commonly refers to drugs in which ingredients are either not contained at all, not in the specified concentration, or in which components have been replaced by harmful substances [8]. Criminals usually focus on counterfeiting the most lucrative medications, which often have the most severe health effects on patients. These include painkillers, antibiotics, contraceptives as well as medicines for cancer and cardiovascular diseases [11]. The pharma logistics network is highly complex and is composed of several actors. A simplified version of those actors involved is shown in Fig. 1.

Fig. 1.
figure 1

Pharmaceutical logistics networks

Full size image

Contrary to this simplified illustration, the pharmaceutical industry comprises many multinational stakeholders involved in the processes. Concerning the in-sourcing of counterfeit medicines, several points in the logistics network are vulnerable to corruption. There is the risk that suppliers do not deliver the right ingredients. Those can be outdated, altered, or not contained in the right concentration [4]. Also, a completely different raw material may be supplied, which will not only fail to induce the desired pharmaceutical effect but may even cause harm to the patient [12]. The majority of counterfeit medicines enter the logistics network at the manufacturer’s level. The wrong ingredients may be used or the right ingredients in the wrong concentration. Also, there is a risk during the packaging processes, as a counterfeit might be packaged with legitimate cartonnage, making it almost indistinguishable from authentic medicine [12].

Also, risks occur at the interface between manufacturers and wholesalers/distributors. While primary wholesalers indeed source their products from manufacturers, this is not necessarily the case with secondary wholesalers. In reality, a company’s position as a secondary or primary wholesaler is not immediately apparent. For example, a wholesaler may purchase a type of drug from one manufacturer and at the same time from a secondary wholesaler if market demand exceeds the manufacturer’s production capacity. This transfer of drugs between different distributors is not uncommon in the pharmaceutical industry. Often, products are repackaged at each handling before being forwarded to the next company. This recursive and non-transparent transport of goods creates a very high degree of opacity regarding the provenance of medications [5].

At the wholesale level, counterfeit drugs may be mixed with legitimate products. This process is called sating and can happen without the malicious intent of a wholesaler. Sating occurs, for example, when a primary wholesaler purchases drugs from a secondary wholesaler who has unwittingly purchased counterfeit drugs. Subsequently, counterfeit drugs may receive authenticity labels during repackaging at the primary wholesaler and are identified as legitimate in subsequent tracking processes. Repackaging at the wholesaler’s level poses additional risks. Since manufacturers often sell their drugs in fraud-protected packaging, repackaging may remove authenticity features [5]. Finally, illegal drugs are often insourced at the pharmacy level. In particular online pharmacies represent a compelling opportunity to acquire medicines at low cost. At the same time, they are one of the most significant sources of counterfeit medicine in-sourcing [1].

3 Methodology

GT was chosen as a suitable methodology as it allows to understand the process behind a research object and to explore creative perspectives and insights regarding human interactions and business practices. In this regard, GT facilitates the development of theoretical models grounded in empirical data and their systematic analysis. Particularly in the research discipline of logistics and supply chain management, and even more so in the identification of critical success factors, GT has been widely accepted as a valid research methodology. Knowledge about blockchain and pharma logistics networks was gained and used during data collection, data analysis and theory generation. Within the GT methodology, both qualitative and quantitative data sources can be used to build theory. Data can be collected through focus groups, questionnaires, surveys, transcripts, letters, government reports, documents, grey literature, music, artifacts, videos, blogs and memos [13]. Especially interviews are a well-recognized method for data collection [14].

Table 1. Overview of interview partners
Full size table

In total 9 interviews were successfully conducted with an average of 50 min of interview length. In total, 413 min of interview recordings were analyzed. This resulted in the transcription of 64,513 words, which led to an overall transcript of 137 pages in length. In Table 1, the 9 experts are described in more detail in terms of their industry function, job position and years of experience (YoE) in the industry.

Based on the transcripts of the interviews, a data structure consisting of first-order concepts, second-order categories and higher-level aggregate dimensions was derived. A complete overview of the structure can be given upon request. In detail, the open-coding process examined 512 quotations from the transcripts of the interviews. These quotations were condensed into 251 initial concepts based on overlaps in meaning. Then, the 251 open-coding concepts were abstracted to 69 second-order categories during axial coding. To improve the accessibility of results, the 69 second-order categories were aggregated into 36 second-order category clusters which present an aggregated form of the second-order categories. Despite this being a deviation from the usual data structure of grounded theory models, this approach allowed for better understandability of the models without sacrificing explanatory value. A final increase in abstraction was achieved by identifying 10 aggregated dimensions from which all second-order categories and first-order concepts can be derived. The coding results were then assigned to key areas of enablers, barriers and implementation strategies.

4 Results

Figure 2 summarizes the findings of the grounded theory model. The left side comprises the identified implementation enablers, which positively affect the implementation strategies via the dashed lines. The right side visualizes the implementation barriers, whose constraining effects on the implementation strategies are represented by dotted arrows. In the middle of the model, the three aggregate strategy dimensions are displayed along with their second-order category clusters. The solid arrows illustrate the causal relationships between the individual elements. As for many GT models, a complex understanding of the mechanisms in the respective research field emerges. For this model, this leads to the fact that all the information gained about the process can hardly be presented in the context of a brief conference paper. Therefore, the core findings are presented here instead of discussing all dimensions in detail. However, upon request, all information can be provided in more detail.

Looking into the barriers it can be seen that the pharma industry is very slow in adopting new technologies which is also because the industry is very reactive to legislative requirements. At the European level, the DataMatrix-code increased the security against counterfeit medicines to such an extent that illegal drugs hardly matter any-more and the incentive to purchase illegal medicines diminished substantially over recent years. Moreover, initiating a blockchain solution would require several technological standards that are hard to introduce in a very fragmented pharma industry with many stakeholders having different target systems and fearing competitive disadvantage. The non-willingness to share certain information can also be observed in the pharma industry since information asymmetries are often beneficial for some parties in the chain. Wholesalers risk negative consequences from full information transparency. On the one hand, wholesalers are concerned that manufacturers could exploit data transparency to control the flow of products through the logistics network, reducing the wholesalers’ scope for action. On the other hand, wholesalers source their products from different countries to leverage price advantages, thereby decreasing product costs and easing the burden on the healthcare system. Even though this form of parallel importation is legal and allows patients to purchase safe and high-quality pharmaceuticals at attractive prices, tracking efforts are enormous. Parallel imports pass through complex logistics networks and cross multiple jurisdictions before being dispensed.

Fig. 2.
figure 2

Implementation framework for blockchain-based traceability to tackle drug-counterfeiting

Full size image

There are several strategies to cope with a plethora of barriers. However, stakeholders’ incentivization and collaboration emerged as the central strategy that opposes a huge challenge for successful blockchain implementation. Accordingly, most blockchain implementation projects do not fail due to the technology’s immaturity but because of the unwillingness of the required partners. Hence, a holistic approach that provides smooth implementation processes in a resource-saving manner is needed. On the one hand, an incentive environment for using the technology has to be created. On the other hand, blockchain developers should regard streamlined adaptation processes and cost-efficient operation from the beginning. Since most pharmaceutical enterprises act largely independent and profit-oriented, heavy investments need to be justified by beneficial returns of investment. Patient safety, however, is difficult to quantify and therefore hard to measure as a KPI. Thus, the assessment of monetary value propositions becomes more important. Especially in the EU, where supply chains are very secure, improving anti-counterfeiting doesn’t suffice as a sole motivator for large implementation projects. This insight might be the biggest contradiction to researchers’ opinions, who suggest anti-counterfeiting as the primary use case for global blockchain implementation. The justification of financial expenditures is even more challenging for those stakeholders that have to invest the most but benefit the least from new solutions. In the pharma logistics network, this applies in particular to wholesalers. They already contain high visibility over the information flows with the existing end-to-end solution. Moreover, wholesalers need to introduce individual mass scanning for all products to allow full data transparency in the EU. This, however, is heavily disincentivized due to competitive pressure and low-profit margins. Consequently, the likelihood of success for the deployment of blockchain is relatively independent on technological advancements. Instead, incentive mechanisms are needed to justify the adaptation for each stakeholder individually.

5 Limitations and Final Remarks

Although this study provides interesting insights that expand the current state of science, its limitations must also be mentioned. The number of interviews conducted is still low considering the variety of stakeholders in the process. However, most important stakeholders have been covered and all interview participants had expertise about issues along the entire logistics network, leading to a low probability of bias in the data. Furthermore, interviewees came from the field of logistics and supply chain and comprised high blockchain expertise, but with limited experience in concrete technological configuration and implementation as software developers. Still, it is very unlikely that the main insights of this thesis become invalid.

This study addresses the research gap of analyzing blockchain implementation given a specific use case. As a result, the pharma industry’s dynamics and strategies were systematized and related. However, the derived implementation framework does not provide an action guide to select interrelated work packages for project management purposes. Instead, future research is needed in applying and developing the created framework within the pharma industry. This will require intensified networking between developers and management as well as the analyses of industry-specific incentive systems. Furthermore, in-depth assessments of stakeholders’ requirements need to be executed to support developers in designing a fitting, compliant and streamlined blockchain platform. With these efforts being made, blockchain may grow beyond technical feasibility studies and find its way into the commercial use of various industries.