Data Streaming in Healthcare and Pharma – Use Cases and Insights from Cardinal Health

To start now, move from batch reporting to a streaming data fabric around Cardinal Health’s network that ingests signals from suppliers, warehouses, labs, and EHR/LIMS feeds. This presence of live data, which provides immediate visibility, lets a small operations team detect anomalies within seconds, trigger actionable alerts, and keep the right people informed. Those signals help align inventory, manufacturing, and clinical workflows in a single, coherent view.

Over years of operations, those teams have learned that real-time signals reduce delays in responding to stockouts, temperature excursions, or QA issues. They will appreciate the predictable throughput and reduced variance in handling events. The move toward streaming strengthens strategies by standardizing event schemas, adopting a single messaging backbone (for example, Kafka or a managed service), and applying backpressure handling to prevent data loss. This approach improves average processing times and overall performance across sites.

Key use cases to prioritize in the near term include real-time inventory optimization, cold chain monitoring, pharmacovigilance, and lab-result routing to clinicians. Data streaming also supports acquisition-related decision-making by aligning supplier movement with demand signals, shortening time to approvals and improving cross-functional visibility for those projects. They can guide prioritization and resource allocation across regions.

Over the years, Cardinal Health has built a resilient streaming layer with strong security, auditable data lineage, and clear governance. They focus on decoupled producers/consumers, schema evolution with a registry, and end-to-end monitoring to maintain presence and reliability. The team is excited about how this setup will move decision-making closer to the data source, strengthening patient safety and cost optimization for those engaged in care and supply.

Four strategies for creating a consumer-driven supply chain in healthcare and pharma

Begin with establishing a real-time, consumer-driven data fabric using Confluent to unify ordering, inventory, and patient access across pharmacies, retailers, and providers. This approach creates a continuous, single view that supports their decision-making and improves outcomes for their customers.

1. Strategy 1: Build a real-time data fabric that bridges retail and healthcare networks.

   • Ingest ordering events (order_id, product_id, quantity, location, timestamp) from pharmacies, hospitals, and distributors through Confluent, enabling a continuous flow of actionable data.
   • Incorporate equipment telemetry and transit status to reflect true on-shelf availability and expiration risk, so operators can act before stockouts occur.
   • Define a common data model so the information is applicable across use cases like retail, hospital, and home health channels, which streamlines onboarding of new partners.
   • Run a pilot in ohio with mcbride pharmacies to bridge retail and provider networks and validate the model at scale.
   • Expected outcomes: stockouts drop 12-20%, order cycle time cuts 20-35%, and productivity rises 10-25% across the ecosystem.
   • Documentation and dashboards track key metrics; publish learnings on linkedin to extend impact and attract new collaborators.

2. Strategy 2: Standardize data contracts and governance to accelerate integration.

   • Create standardized data contracts for ordering, shipping, receiving, inventory, and patient requests, making them applicable across ERP, WMS, and pharmacy management systems.
   • Apply data quality rules and lineage with Confluent governance features to support audits and reduce reconciliation effort by 40-60% in early deployments.
   • Adopt a clear integration plan to connect EHRs, pharmacy systems, and supplier networks, avoiding silos that slow change.
   • Document runbooks, API specs, and change logs so operators have self-serve access to the latest guidelines.
   • Demonstrable value: faster supplier onboarding and fewer errors, with measurable improvements in spend visibility and control.

3. Strategy 3: Drive replenishment through consumer spending signals and collaborative planning.

   • Use spending signals from patients and providers to adjust stock levels at each location, aligning inventory with demand and reducing waste.
   • Coordinate with mcbride and ohio-based retail partners to align forecasts with promotions, payer timelines, and patient access windows.
   • Implement joint planning with suppliers to improve service levels; target on-shelf availability of 95% for priority meds and rapid fulfillment for urgent needs.
   • Bridge data between retail channels and specialty pharmacies to maintain consistent product availability across touchpoints.
   • Expected productivity gains arise from smarter ordering, faster decision cycles, and better utilization of equipment and space.

4. Strategy 4: Modernize systems and cultivate a culture of continuous learning and collaboration.

   • Execute modernization of legacy systems by migrating to streaming analytics, upgrading interfaces, and enabling real-time visibility through the pipeline.
   • Establish a cross-functional governance board with operators, pharmacists, clinicians, and supply chain leaders to review metrics and adjust priorities monthly.
   • Document best practices and case studies; share milestones on linkedin to amplify innovations across the network.
   • Invest in training and certifications to keep teams excited and proficient, translating into higher productivity and better patient service.
   • Track time-to-insight, change adoption rate, and spend optimization to quantify impact and guide ongoing modernization efforts.

Real-time integration of patient data into care pathways

Deploy dell edge gateways to capture presence, latest vitals, vaccination status, and pharmacy events in real time, and feed them into care pathways through a standardized methods layer. This setup minimizes lag and ensures clinicians see up-to-date alerts in the workflow before critical decisions are made.

Map data to interoperable paths using FHIR resources and cordis-backed reference models, so each data type flows through gateways and into the care pathway logic without manual re-entry. Use a forward-looking design to adapt to new data sources such as wearables and external labs.

Triggers include overdue vaccination, abnormal labs, and dispensing events that indicate adherence gaps. Real-time alerts prompt teams across care settings to intervene. This approach improves outcomes. Leaders monitor a single dashboard to oversee cross-department activity.

In the clinical workflow, the presence of a patient at a touchpoint automatically advances the path: if a vaccination is due, the system surfaces tasks for the nurse and updates the path accordingly. This supporting data strengthens patient experience and reduces unnecessary visits.

Pharmaceuticals teams gain visibility into inventory and dispensing patterns. Real-time data informs stock decisions and coordination with supply chain. This reduces stockouts and enables timely dosing plans, aligning with cordis and guidelines. The approach scales across departments, enabling cross-functional collaboration for paths that adapt to patient needs. Much reduces waste and delays.

Security, privacy, and data quality controls are embedded: role-based access, consent flags for sharing, and data mapping checks at ingestion. This keeps the workflow safe while enabling timely insights for clinicians and pharmacists.

Implementation plan: start with a september pilot in two clinics, then scale to hospital-wide use within six months. Establish a feedback loop with teams and leaders to fine-tune triggers and paths as you gather real-world results.

Streaming telemetry for medication provenance and inventory visibility

Implement a streaming telemetry layer that ties serialization and batch provenance to real-time inventory updates across warehouses, distributors, and pharmacies. This delivers ορατότητα into where each unit stands and helps prevent stockouts or overstock, with updates flowing continuously from source to point of use.

Adopt an integrated platform that ingests data from providers, manufacturers, and sites, then surfaces a unified view for the entire chain. Foster συνεργασία among people απέναντι technical, operations, quality, and compliance teams. Design workflows that are applicable across segments and anchored to a shared data model, according to knauff, goldman, και wavemark. Team alignment happens here.

Define telemetry events with concrete mapping: shipment departure, custody transfer, receipt confirmation, batch/lot, expiration, temperature and humidity readings, GPS location, and pallet-level scans. Use a canonical schema to enable cross-system matching and ensure immutable logs for medication provenance.

Run streaming analytics to detect anomalies within seconds and push updates to dashboards or APIs. Implement role-based access and reconciliation rules to ensure data quality and auditability. Alerts should trigger for temperature excursions, missing milestones, or misaligned lot numbers, and all critical events should be time-stamped and traceable within the platform.

Measure impact with concrete metrics: latency under 15 seconds for critical events, throughput of tens of thousands of events per hour, coverage of end-to-end shipments in at least 90% of the chain segments, and provenance accuracy above 99.5%. Track earnings impact from reduced stockouts, lower write-offs, and improved inventory turns, validating with quarterly ROI analyses. Build a feedback loop from the χαρτοφυλάκιο του programs to continuous improvement cycles.

From a governance standpoint, establish data quality gates, provenance audits, and privacy controls aligned with healthcare regulations. Provide training and governance artifacts for provider partners, wholesalers, and pharmacies, and maintain a living playbook that guides onboarding of new sites and suppliers.

Experts knauff, goldman, και wavemark emphasize the value of a modular, scalable approach. Start with a focused set of high-value products, then expand across the chain as data quality and collaboration mature, ensuring the platform remains integrated and extensible.

Using patient-driven demand signals to optimize procurement and manufacturing

Providing real-time visibility into patient-driven demand signals enables procurement and manufacturing teams to align production and purchasing with what patients need. Begin by establishing a cross-functional governance that includes provider representatives, clinicians, pharmacists, and manufacturing leads to translate those signals into concrete plans for procurement and production. This approach supports meeting patient needs for therapies and supporting their treatment plans.

Combine prescription fill data, patient-reported outcomes, provider notes, and therapy adherence metrics to create a robust signal set. Such data helps find bottlenecks across supply and manufacturing, not just at the warehouse but throughout the chain. It also reveals underbilled items and offers opportunities to strengthen collaboration with suppliers, ensuring pricing and terms support meeting patient needs.

Develop a robust data pipeline that ingests signals from subscribing patients, providers, and care teams. Define what signals to capture (treatment cycles, refill intervals, appointment windows) and how to slide demand into procurement cadences. Establish leading indicators and set targets to reduce stockouts and the pressure on supplier capacity. Provide clear guidance to buyers and plant managers so actions are aligned.

Discussions with providers and customers should be routine, with monthly meeting cycles to discuss forecast accuracy and service levels. The most actionable insights come from cross-functional reviews that occur throughout planning, production, and distribution. Providers who are subscribing to a shared signal set help ensure operations run seamlessly across sites.

Monitor KPIs such as demand forecast accuracy, on-time delivery, service levels, and patient-level fulfillment times. The guidance should specify what signals are used to trigger orders and how to scale procurement during demand surges. With a robust data foundation, underbilled services can be identified and addressed, strengthening provider contracts and offers to ensure patients receive uninterrupted therapy.

By strengthening the link between patient demand and supplier capacity, teams gain a robust forecast that reduces bottlenecks and improves service levels throughout the network. The outcome is seamless procurement and manufacturing alignment that supports providers’ ability to meet most patients’ therapy needs, even as volumes shift. Regular reviews with clinicians, pharmacists, and distributors ensure feedback loops close the gap between what patients experience and what manufacturers plan.

Privacy, governance considerations for streaming health data

Implement end-to-end encryption and zero-trust access for all streaming health data; codify these controls into a policy that teams around provider networks, pharmacies, and developers can follow here.

Define a streaming privacy governance model with a data governance council, data owners, and cross-functional teams. Map data flows across ingestion, processing, and delivery, documenting conditions around consent, purpose limitation, retention, and shared processes.

Adopt shared methods for de-identification and encryption, ensuring data can be analyzed while reducing re-identification risk; this supports automation and accelerates insights for clinicians and researchers, and expands capabilities across the stack.

Implement real-time access controls and continuous auditing; use logging and anomaly detection to detect unusual patterns and enforce least privilege across pharmacies, provider systems, and downstream analytics.

Scale considerations: modernized streaming programs must address accretion and data volume, including major streams across pharmacies and provider networks, and a major solution for managing billions of events to drive outcomes while reducing pressure on teams.

In a webinar, developers and security teams said this approach reduces risk and accelerates outcomes, with insights to help optimize workflows here.

Use a privacy card in the governance toolkit to remind teams about consent, data minimization, and sharing rules.

Apply these controls iteratively, starting with high-risk data streams and expanding to routine sensor feeds across pharmacies and provider networks.

Architectures for scalable data streaming: event streams, pipelines, and dashboards

Recommendation: Move to an industry-leading, event-stream-first architecture that unifies healths devices, patient signals, and supplier data into scalable pipelines and intuitive dashboards. This approach reduces latency, boosts productivity, and lowers cost, enabling automation across clinical and operations workflows for every stakeholder they serve, which accelerates decision cycles.

Event streams form the backbone. Use confluent as the backbone with topics, which include healths.patient.readings, devices.events, orders.alerts, and pharma.signals. Target sub-100 ms latency for critical alerts, retain 7–14 days for fast review and forensics, and apply a schema registry to enforce data contracts across areas. For external sharing, apigee governs APIs and enforces security policies. Those streams move data to real-time analytics, dashboards, and downstream warehouses; never lose fidelity with replayable events and exactly-once semantics. Analysts said a unified, cross-domain view yields the largest productivity gains.

Pipelines and processing: Build streaming ETL/ELT pipelines with idempotent transforms, stateful operators, and automated data quality checks. Use confluent’s ksqlDB or Kafka Streams for on-the-fly enrichment, risk scoring, and inventory forecasting. Design paths for clinical analytics and operations dashboards; enforce gmpd standards for data governance. Follow GMPD standards for data governance. Containerize microservices and automate deployment to reduce manual steps, which lowers cost and increases productivity.

Dashboards and governance: Turn streams into decision-ready visuals. Build comprehensive dashboards for clinicians, lab managers, and executives; ensure comparable views across sites and GMP-compliant privacy. Tie streaming outcomes to earnings by illustrating cost-to-value, service levels, and utilization. Use apigee so partners securely access data through APIs, while keeping data governance intact. Include signals from twitter to identify early market trends and supply disruptions, enabling proactive planning. As a provider, Cardinal Health can standardize across sites.

Review and optimization: Establish a cadence to review latency, throughput, error rates, and data quality; compare dashboards across areas; track cost per event and the return on investment; adjust pipelines to align with the move to streaming. This keeps the platform scalable for the largest volumes and will change how health systems and pharma networks operate.