Managing Supplier Risk in Health Care Supply Chain

Leveraging Blockchain and AI

Recently, I had the opportunity to speak at the ConVerge2Xcelerate symposium in Quincy, MA, by Partners In Digital Health . In this article, I summarize the key elements of my talk which focused on how blockchain and AI can help bring efficiencies to the pharma supply-chain.

My journey in healthcare began in 1983 at the Tata Memorial Cancer Center in Mumbai, where I applied basic statistical methods to demographic data to analyze the impact of cancer across various population groups, considering factors such as region, religion, gender, and cultural habits. Later, I contributed to technology optimization, service enablement, consolidation, and integration at Pfizer, Blue Cross, and Medtronic, leading to significant efficiencies and improvements.

In the Research Triangle Area, known for its numerous drug contract research companies, firms are investing in chemistry to build molecular libraries that support new drug formulations. I had the opportunity to be connected with each in some manner. Even before the advent of blockchain and AI, companies like Avior Bio made notable strides by discovering compounds with expired patents that revealed new therapeutic uses. CiVentiChem, a CRO, has amassed an extensive library of compounds to aid in complex drug research. At RTI, discussions have explored using blockchain and AI to safely and discreetly dispense opioid reversal medication Naloxone, addressing the stigma associated with its use. The USPS has conducted proof-of-concepts to investigate how blockchain can help border patrol detect drugs in incoming packages. At Blue Cross, efforts have focused on reducing the number of claims and the cost of processing them. Tranzyme Pharma discovered and developed novel, mechanism-based therapeutics based on its proprietary small molecule macrocyclic template chemistry (MATCH™) technology. MATCH enabled the rapid construct of synthetic libraries of drug-like, macrocyclic compounds in a predictable and efficient manner. These initiatives could have greatly benefited had AI, machine learning, IoT, and blockchain technologies were available to them as tools.

A view of the nature of use-cases that some of the clients I have worked with expected. It is noteworthy that some pharmaceutical companies still have siloed disconnected processes and applications that become roadblocks to taking advantage of advanced technologies..

In the following sections we will discuss two aspects of the Health Care supply chain i.e. pharma and health plans. The Health Care domain is vast and covers many more sub-domains such as hospitals and clinics, clinical trials, contract research and much more.

A Simplified View of the Pharma Supply Chain

A supply chain consists of several processes grouped by upstream and downstream. The initial stages of a pharmaceutical manufacturer supply chain includes sourcing, acquiring, and transporting raw materials to manufacturing facilities.

The downstream processes are complex and include the customer order management and customer order fulfillment. Suppliers play an active role in both streams. Depending on the scale and distribution of the supply chain, supplier risks become exponential leading to inefficiencies, poor quality and lost revenue.

Looking a bit deeper into the manufacturing process, one would discover that there are quality requirements and compliance with local and global regulations, practices and standards.

Typical critical supplier categories in the supply chain

• Raw Material Suppliers
• Contract Manufacturers
• Contract Research Organizations
• Packagers and  Labelers
• Logistics / Transportation Providers
• Warehouse Operators
• Distributors

The drug discovery and development process is often hailed for its breakthroughs,. Behind the scenes, an intricate network of suppliers, manufacturers, and regulators works tirelessly to ensure a consistent and safe drug supply.

Raw Material Sourcing and API Production: A critical aspect of drug manufacturing is the procurement of raw materials. These can be divided into three categories: primary (active pharmaceutical ingredients or APIs), secondary (excipients), and supporting (packaging materials). Many of these materials, particularly APIs, are sourced from countries like India and China, underscoring the global nature of the drug supply chain.

The production of APIs is a highly regulated and complex process. Depending on the drug or vaccine, different manufacturing methods are employed. APIs are the molecular components that target specific diseases. Once produced, they are combined with non-active ingredients (excipients) and formulated into appropriate dosage forms.

Packaging and Quality Control: Packaging plays a crucial role in preserving drug efficacy and safety. Drugs undergo primary packaging, which involves enclosing the product in containers like bottles, vials, blister packs, or syringes. Secondary packaging, such as cartons or boxes, provides additional protection from environmental factors that could compromise the drug's quality.

Throughout the manufacturing process, quality control is paramount. Regulatory standards are strictly adhered to, ensuring that labeling, packaging, and manufacturing practices meet the highest quality standards. Detailed records are maintained for each batch of a drug, providing traceability and facilitating quality investigations if necessary.

In the drug discovery and manufacturing process, each process could be executed within the main pharmaceutical company to fully or partially outsourced across the globe.

A Simplified View of the Health Plan Supply Chain

Health Plan products follow a similarly intricate manufacturing and distribution process. Annually, health insurance companies evaluate product performance by analyzing patient and product data to develop new plans or improve existing ones. This process includes determining the target asset to be insured, coverages, reviewing providers and partners, negotiating costs, pricing models, and underwriting.

Products are then tested for various scenarios before being made available to customers. Post-sale, products must be serviced and monitored continuously. AI and ML can assess customer sentiments, enabling the innovation of better products tailored to individual needs. Blockchain technology can enhance trust, privacy, and transparency for all participants.

A key component across all supply chain scenarios is the supplier. Effective management of the supplier lifecycle—from onboarding and continuous risk monitoring to regulatory compliance and service quality—ensures the supply chain remains resilient, cost-effective, and efficient. The table below illustrates the compliance and regulatory landscape and highlights the numerous risk analysts who provide crucial insights into various risks.

In the pharmaceutical space, companies must adhere to Good Manufacturing Practices (GMP), encompassing every aspect of procurement, drug manufacture, material handling, quality control, regulatory compliance, waste disposal, and transportation. Additionally, HIPAA enforces privacy in the management of personal health and private data, adding another layer of complexity and responsibility.

Generative AI, Predictive AI, Blockchain, and IoT can collaborate to create a safe and secure drug supply chain, compliant with the Drug Supply Chain Safety Act (DSCSA). Retrieval-Augmented Generation (RAG) models can be developed and deployed to address various applications, such as:

• User Services

- User Assistants such as Chat, FAQ

- Insights on Application Usage,  Automated Self-Service

• Insights from

- Corpus of Documents such as Contracts and Research

- Location-wise Regulatory and Compliance

• Customer Services

- Automated customer service requests

- Resolve and/or route service requests

- Auto Translate request/response to appropriate languages

Custom-trained language models, created using techniques like LoRA, qLoRA, and Reflective Tuning, can be tailored to meet specialized needs across various industries. For instance, a legal firm might fine-tune a model to understand complex legal documents, while a healthcare provider could use one to analyze medical records for early disease detection and customized medical plans.

In the supply chain domain, Generative Adversarial Networks (GANs) have proven invaluable for tasks such as detecting fraudulent products, counterfeiting, mishandling, and compliance violations. By analyzing images of products, packaging, and shipping processes, GANs can identify anomalies that may indicate issues. The proliferation of accessible tools and a vast array of language models has significantly lowered the barrier to entry for businesses adopting AI, making it easier than ever to leverage these technologies for competitive advantage.

A Hypothetical Example - Claims Processing

Health plan administrators' are still dealing with legacy systems, and benefits and claims management often cost them tremendously. Data validation, detecting errors in coding and sometimes erroneous or fraudulent transactions are key issues.

A Hypothetical Example that we built for fun to demonstrate the use of GenAI in Claims Processing is depicted. The goal was to test data extraction accurately of claim forms and comparing it against plans and benefits before approving. Traditional data processing systems implement this as master data transactions, but here we are testing Generative AI capabilities using llamaIndex and language models.

In summary, the combination of Generative AI (GenAI) and Blockchain technology offers a powerful solution for managing supplier onboarding and supplier risk within the pharmaceutical supply chain, ensuring a trusted and secure ecosystem. GenAI can streamline the onboarding process by automating tasks like data collection and verification, while Blockchain provides an immutable record of transactions and interactions, enhancing transparency and traceability. This collaborative approach mitigates risks associated with counterfeit drugs, contamination, and improper handling throughout the supply chain. Moreover, emerging technologies like IoT, AR, VR, and XR (extended reality) further bolster these capabilities by providing real-time monitoring, remote inspections, and immersive experiences that enhance quality control and authenticity verification.

I would welcome the opportunity to engage in discussions or partner with companies to share ideas and explore these topics in greater depth. Reach out to me on LinkedIn or mail://bits4bytes@outlook.com or X @vsmohan