Move supply chain resilience into the clinical operating model

Map the dependencies behind every essential product

Resilience begins when leaders can see beyond the contracted supplier to the full chain that produces, moves, stores, and uses a product.

Most health systems have partial visibility. Contracting systems show vendors and prices. enterprise resource planning systems show purchase orders. distributors show allocations and fill rates. clinical systems show procedures and utilization. pharmacy, laboratory, facilities, information technology, and emergency management may maintain separate inventories or continuity plans. The risk lies in the gaps between those systems.

An executive network map should begin with the patient service, not the item master. For each essential service, identify the products and support resources required to deliver minimum safe care. Then trace each product through manufacturer, production location, raw-material dependency, distributor, transportation route, storage requirement, facility location, point of use, substitute, and clinical owner. Include digital dependencies such as ordering platforms, interfaces, scanning, warehouse systems, automated dispensing, and cybersecurity controls.

The map should distinguish verified information from assumptions. Many organizations know their tier-one supplier but not the actual manufacturer, production geography, or common upstream dependencies shared by “different” vendors. Two contracted suppliers may rely on the same factory or component. Dual sourcing is not true diversification when both sources fail for the same reason.

The U.S. Government Accountability Office review of supply-chain resilience describes national efforts to strengthen collaboration and information sharing while also identifying coordination challenges. Health systems face a smaller version of the same problem. No department owns every data element, so governance must define who integrates the evidence and who resolves disagreement.

Leaders should also map time. Document normal lead time, disruption lead time, allocation behavior, shelf life, conversion time, substitute approval time, and the duration for which the system can maintain essential care. A product with seven days of inventory and a two-day replenishment cycle has a different risk profile from a product with seven days of inventory and a six-month manufacturing recovery.

For infrastructure and hazard dependencies that can disrupt a hospital campus and its suppliers at the same time, see Climate Change and Healthcare: Preparing Facilities for Environmental Challenges. Supply resilience and facility resilience should share scenarios, escalation criteria, and executive ownership.

Segment products by patient consequence, not purchase price

The highest-spend item is not necessarily the highest-risk item. Criticality depends on what happens to care when supply is constrained and how quickly the system can adapt.

A useful segmentation model combines five factors: clinical consequence, time to harm, demand volatility, supply concentration, and substitute readiness. Clinical consequence asks whether a shortage delays convenience, reduces efficiency, changes quality, closes a service, or creates immediate risk of serious harm. Time to harm asks how long the organization can operate before the consequence occurs. Demand volatility reflects whether use is stable, seasonal, outbreak-driven, or dependent on case mix. Supply concentration addresses manufacturers, locations, and logistics. Substitute readiness measures whether an alternative is approved, available, interoperable, stocked, and familiar to users.

The tier should determine the control. Tier A products may require executive ownership, verified upstream mapping, multiple qualified sources, higher safety stock, substitute simulation, conservation protocols, and regular board visibility. Tier D products can rely on standard purchasing and inventory controls. Applying the same policy to every item wastes capital and management attention. Applying a low-intensity policy to a life-critical item hides risk.

Medication and device shortages require distinct expertise. FDA notes that medical-device shortages can arise from manufacturing and quality problems, geopolitical conditions, natural disasters, transportation delays, public-health emergencies, and discontinuations. Its medical device supply-chain and shortage resources provide an external signal, but health systems still need local clinical translation. A national shortage notice does not tell a hospital which cases depend on the product, which locations hold stock, or whether staff can safely use an alternative.

Substitution should be treated as a change in care, not a purchasing transaction. The review should include clinical efficacy, safety, compatibility, packaging, concentration, preparation, workflow, equipment, infection prevention, documentation, billing, training, waste, and patient communication. A nominal substitute can create new risk if units of measure differ, connectors are incompatible, instructions are unclear, or staff must change familiar routines under pressure.

Buy optionality, transparency, and recovery capability

A low unit price can be expensive when it concentrates risk, obscures upstream dependencies, or leaves the organization without enforceable information during disruption.

Strategic sourcing should separate routine commercial negotiation from resilience requirements. For clinically critical products, requests for proposal and contract reviews should address manufacturing locations, alternate sites, component dependencies, quality history, business-continuity plans, cyber controls, transportation modes, allocation methods, surge capacity, notification expectations, and recovery objectives. The goal is not to demand confidential details that suppliers cannot share. It is to establish enough transparency to make an informed risk decision.

Supplier diversity must be tested for independence. Two distributors do not equal two manufacturers. Two manufacturers in the same hazard zone may not provide geographic diversification. A domestic source can still depend on imported ingredients or components. An international source may have stronger redundancy than a nominally local source. Leaders should resist slogans and evaluate the actual failure modes.

Domestic sourcing can be one component of a balanced portfolio. HHS efforts to strengthen the public-health industrial base include support for domestic production and scalable capacity. The ASPR industry roundtable on domestic manufacturing emphasizes resilient, scalable, and sustainable production of critical medical components. Hospital leaders should translate that policy direction into product-level analysis rather than assuming every domestic option is available, affordable, or independent.

Contracts should define the information and behaviors required during disruption. Examples include prompt notice of production or quality events, allocation methodology, access to substitute products, reporting on fill rate and lead-time changes, participation in incident calls, data standards, and post-event corrective action. Contract language cannot create supply that does not exist, but it can reduce delay, ambiguity, and unequal access to information.

Inventory strategy should follow product criticality and recovery time. Safety stock is appropriate when it bridges a plausible disruption, storage conditions are controlled, expiration is manageable, and demand is measurable. It is not a substitute for diversified sourcing or operational readiness. Excess inventory can tie up cash, consume space, expire, and create false confidence. The stronger approach combines segmented inventory targets, rotation, demand sensing, supplier intelligence, and predefined conservation.

Total value analysis should include the cost of disruption. A contract that saves two percent on a product but exposes a high-margin service line to repeated cancellations may destroy value. Finance and supply chain should estimate the cost of deferred cases, premium freight, emergency purchases, overtime, training, waste, quality events, patient transfers, and reputational damage. The estimate need not be perfect. It needs to be good enough to compare resilience investments with the consequences they are designed to avoid.

Predefine the decisions that become difficult under pressure

The first hours of a shortage should not be spent debating ownership, locating inventory, or discovering that a proposed substitute changes clinical practice.

Every health system needs a shortage response pathway that distinguishes routine backorders from enterprise threats. The initial signal may come from a manufacturer, distributor, FDA notice, clinician, pharmacy, inventory system, or external risk-monitoring service. Signals should enter one triage process. The triage team validates the issue, identifies affected products and locations, estimates time to impact, and determines whether the event can be managed within supply chain or requires clinical and executive escalation.

The response should use staged thresholds. A watch stage increases monitoring and verifies data. A conservation stage limits nonessential use and redistributes inventory. A substitution stage activates approved alternatives and targeted education. An allocation stage applies ethically and clinically governed priorities when demand exceeds supply. A service-change stage modifies scheduling, location, transfer, or scope of care. A recovery stage returns the organization to standard practice while reconciling inventory, cost, quality, and lessons learned.

Allocation requires explicit ethics and governance. Purchasing history, political influence, or local negotiating strength should not determine access to a scarce product. Criteria should reflect patient need, clinical urgency, benefit, alternatives, and equity. Legal, ethics, quality, and clinical leaders should review high-consequence rules before a crisis and document decisions during the event.

Communication should be role-specific. Executives need consequence, time horizon, decision, owner, and financial exposure. Clinicians need what is changing, why, when, which patients are affected, what alternatives are approved, and where to ask questions. Patients may need clear information when products, scheduling, or location changes. Suppliers need accurate demand and a single enterprise point of contact. Messages should avoid speculation and acknowledge uncertainty.

After-action review must change the operating system. Document whether signals arrived early enough, data were accurate, owners responded, substitutes worked, communications reached users, quality remained stable, and suppliers met expectations. Correct item data, contracts, stocking policies, clinical standards, training, and escalation rules. A retrospective that produces slides but no control changes does not improve resilience.

The FDA’s annual drug-shortage reporting illustrates the value of early manufacturer notification and coordinated mitigation. The FDA supply-chain reports and publications should inform external monitoring. Local leaders should pair those signals with enterprise utilization and patient-consequence data.

Build a decision system, not a decorative dashboard

The control tower earns value when it helps leaders recognize a threat earlier, choose an action faster, coordinate consistently, and learn whether the action protected care.

Technology can connect item-master, contract, procurement, inventory, distribution, clinical utilization, case scheduling, supplier, transportation, regulatory, and external risk data. Predictive analytics can identify demand shifts, abnormal lead times, allocation patterns, or concentration. Automation can accelerate alerts and reconciliation. Artificial intelligence can help synthesize large volumes of unstructured supplier and risk information. None of those capabilities replaces governance, data quality, clinical interpretation, or human accountability.

The foundation is disciplined data. Standardize product identifiers, units of measure, manufacturer information, substitute relationships, contract terms, location, expiration, and clinical ownership. Reconcile inventory across warehouses, nursing units, procedural areas, pharmacies, laboratories, clinics, and emergency caches. Define which system is authoritative for each field. Assign owners for correction. Poor data transmitted faster is still poor data.

Cyber resilience belongs in the supply-chain design. Ordering, warehouse, transportation, dispensing, scanning, and supplier portals can fail during a cyber incident. Leaders should know which transactions can continue manually, how inventory will be located, which suppliers can receive alternate orders, how deliveries will be authenticated, and how the organization will reconcile activity when systems return. For physical and digital continuity principles, see The Future of Hospital Design: Creating Adaptive and Resilient Facilities.

Targets should reflect clinical context. A high inventory level is not automatically good when products expire quickly. A low level is not automatically unsafe when replenishment is reliable and alternatives are ready. Leaders should review measures together and investigate exceptions. The dashboard should display the evidence behind the signal and the owner responsible for action.

Board reporting should remain concise. Show the top enterprise exposures, trend in critical-product coverage, unresolved single points of failure, significant events, capital or contracting decisions, and corrective actions. Avoid overwhelming directors with thousands of items or operational detail. Their role is to test whether management understands material risk, has set tolerances, funds controls, and verifies performance.

Establish the control tower in 90 days

The first 90 days should create one governance model, expose the most consequential dependencies, and prove that the organization can make and execute a coordinated decision.

Questions for the executive team

1. Which essential services have product dependencies that are not fully mapped?
2. Which critical products depend on one manufacturing site, region, component, or logistics route?
3. Where does enterprise inventory visibility differ from what frontline teams can actually access and use?
4. Which substitutes are approved on paper but not stocked, compatible, documented, or rehearsed?
5. How do allocation decisions incorporate clinical urgency, patient benefit, alternatives, and equity?
6. What information must suppliers provide before and during disruption, and is that expectation contractual?
7. Which technology failures would prevent ordering, locating, distributing, or documenting supplies?
8. What supply risk should the board see regularly, and what evidence demonstrates control effectiveness?

Leaders should not wait for perfect data. Begin with a limited portfolio of high-consequence products and improve the system through use. The goal is not a one-time inventory project. It is a repeatable management discipline that expands across categories, facilities, and service lines while preserving focus on patient consequence.