Healthcare IT Infrastructure

🧬 We talk about “health data” as if it’s one thing, but it’s really hundreds of incompatible languages trying (and failing) to talk to each other. Every layer speaks a different dialect: • EHRs: HL7 v2, CDA, FHIR • Claims: X12 837, UB-04, CMS-1500 • Labs: LOINC, SNOMED CT • Devices: DICOM, IEEE 11073 • Genomics: VCF, FASTQ, BAM Each was built for a single purpose, not interoperability. The result? 🚑 A patient’s data is scattered across 40+ systems, each with its own schema, timestamps, and access controls. But things are shifting. Newer models are moving beyond formats to: • Graph-based data structures • Semantic layers • Federated architectures These approaches preserve context, not just content, across systems. FHIR paved the road. But the next frontier is semantic interoperability. That’s not just data exchange; it’s data understanding. 🧠 The future of healthcare intelligence isn’t in collecting more data, it’s in connecting meaning. #HealthTech #DataInteroperability #FHIR #HealthcareAI #KnowledgeGraphs #SemanticWeb

When I say, ‘See you at 7’, do I mean 7 AM or 7 PM? ⏰ This is what we call, an interoperability problem - the inability for systems to exchange data and understand it in the same way. Why is interoperability in healthcare so hard? Because it’s not just a tech issue. It’s a stack of challenges And the hardest part isn’t connection, it's understanding. Let’s break this down. 1️⃣ Technical Interoperability - can systems connect and exchange data? Sounds simple, until: 🔸One system uses CSV, another wants XML 🔸Dates are DD/MM/YYYY vs MM/DD/YYYY 🔸Fields don’t match or exist Without standard formats, even basic connections break. Challenging - yes. But ironically, the easiest layer to fix. (Most teams stop here. That’s the issue.) 2️⃣ Semantic Interoperability - Can systems understand the data? Take “discharge date” as an example: 🔸One system uses the paperwork date 🔸Another, the bed exit time 🔸A third, the billing date Same label, different meanings. Now try running a report across all three. This is where projects quietly fail. Semantics needs shared meaning, clinical context, and governance. (And that’s just admin data, imagine lab values, diagnoses, or clinical notes. Get it wrong and it’s not just inefficiency, it’s a safety issue!) 3️⃣ Workflow Interoperability - do systems fit real care delivery? 🔸A patient sees a doctor in the morning, does a lab test in the afternoon 🔸Lab results are ready but not visible till the next day 🔸Why? The EHR and lab system don’t sync in real time, and no one flagged it. Digital isn’t fast if the workflow stays broken. 4️⃣ Organizational Interoperability - do institutions even want to collaborate? 🔸Hospitals, clinics, insurers, labs etc. have different systems, incentives, and vendors 🔸Even if tech and semantics align, nothing moves without shared ownership The real question isn’t “Can systems talk?” It’s “Do they understand each other and act together?” And more importantly - who’s responsible for making that happen? Because in healthcare, everyone is in charge, yet no one really is. Let’s stop treating interoperability like a checkbox and start treating it as a system-wide commitment: to shared meaning, coordinated action, and patient-centered design. What’s one interoperability headache you’ve seen that should’ve been solved by now? #Interoperability #SemanticStandards #SystemThinking #HealthData 💡This post is part of 'Rethinking Digital Health Innovation' (RDHI), empowering professionals to transform digital health beyond IT and AI myths. 💡The ongoing series and additional resources are available at http://www.enabler.xyz 💡Repost if this message resonates with you!

FHIR is not a silver bullet that solves all interoperability problems. Throwing money at a big FHIR project that exposes or consumes data via FHIR APIs does not mean you can tick the “Interoperability” checkbox. Here are some examples of how systems using FHIR can look interoperable without truly being interoperable. 1. “FHIR compliant” is not the same as good data - Data can pass FHIR validation and still be incomplete, inconsistent, or misleading. 2. Context can be lost in transmission - Observations recorded without Encounters, intent or circumstances. 3. FHIR validates structure, not plausibility - A perfectly valid 5 year old patient: male, married and pregnant. 4. Coded data can still be bad data - Custom codes when bindings are loose. Text values in place of codes. 5. Bad data can be shared as easily as good data - FHIR enables all data to move efficiently (bad data as well as good data). When two systems speak to each other using FHIR for the first time, there’s often celebration. “Data is flowing. Everything is working!” But a key part of Interoperability is that data should be usable by other systems. If another system can read your FHIR data but is unable to confidently use it, then you are NOT interoperable, no matter how valid and compliant your FHIR resources are. When data flows but meaning doesn’t, interoperability is an illusion. ~~~ Ways to work with me: https://lnkd.in/eWwXNk_U

Imagine a future in which the same Wi-Fi signals that stream our podcasts and deliver our emails quietly safeguard the people we care about. In 2013, researchers demonstrated that ordinary Wi-Fi reflections could operate as a form of sonar, capturing human movement through walls without cameras or wearables; at the time, the concept seemed far-fetched. A decade later, Carnegie Mellon University advanced the field, showing in January 2023 that a standard router, paired with machine-learning models, can reconstruct full human poses in real time, moving the idea from curiosity to commercial roadmap. Attention has since shifted to LoRaWAN, the ultra-low-power network best known for industrial and agricultural telemetry. Recent studies report 93 percent accuracy in detecting falls through walls at distances of up to ten metres. Healthcare technology providers are already offering LoRaWAN-enabled systems that monitor bed presence, mobility, and falls, issuing alerts that help older or infirm individuals remain independent for longer, and easing the financial pressures of residential care. The lesson is clear: transformational breakthroughs often arise not from novel inventions, but from re-imagining the overlooked capabilities of technologies already woven into our environment. As AI continues to percolate through our business and personal lives these novel solutions will continue to emerge. Existing business capabilities will undoubtedly be leveraged in new and innovative ways too.

I did my first Residential Aged Care Medication Round today after working as a doctor for 30 years- not by choice by rather by urgent circumstance. This morning, I hit a wall in patient care. Not a medical wall, but a digital one. A long-term patient, admitted urgently into a Residential Aged Care Facility with influenza and worsening heart failure. I needed to prescribe urgent medication. But I couldn't. 🚫 IT glitch. Couldn't access the aged care facility's electronic health record (EHR). 🚫 Remote system. No immediate IT support. 🚫 Policy barrier. Nurses, rightly following protocol, couldn't accept a verbal order. They could only administer what was on the locked EHR chart. My patient was deteriorating, and the system designed to help was blocking me. The solution? I did the medication round myself and recorded it in my *clinic's* separate, unconnected EHR under the watchful eye of the nurse in charge. I work across 2 hospitals, 2 clinics, and 2 aged care facilities. **Six different, siloed EHR systems.** This isn't efficiency; it's a patient safety risk waiting to happen. This isn't an IT problem. It's a **healthcare system problem**. How many other clinicians are facing this digital fragmentation daily? How do we advocate for connected care that puts the patient, not the platform, at the center? #DigitalHealth #HealthTech #Interoperability #PatientSafety #AgedCare #Healthcare #Medicine #LinkedInMedics #FutureOfHealth

2,000 CMS Auditors and the End of Assumptions On May 21st, CMS dropped a bombshell. They’re expanding RADV audits from 40 auditors to 2,000. A 50x increase. And they’re not just targeting a few plans—they're coming for all of them. This isn't a regulatory blip. It's a full system recalibration. And it forces a mindset shift for anyone working in Medicare Advantage. It increases patient record reviews from 2,100 per year to 110,000 or more. For years, the industry’s rhythm has been: capture diagnoses → justify revenue → repeat. That loop is being broken. CMS is re-centering around defensibility, not documentation volume. It’s not just about having the codes. It’s about proving them. With audit-grade specificity. In every chart. At every visit. The impact downstream is huge: • Providers will feel this in their inboxes and exam rooms. • CDI teams will need new playbooks. • Coders—already stretched—will now be revenue protectors as much as revenue drivers. This also resets the relationship between plans and providers. There’s going to be more scrutiny, more release-of-info requests, more “we need to talk about your documentation.” If you’re a health system, hospital, or medical group leader, you might be asking: • Are we capturing conditions or defending them? • Do our CDI strategies still fit the risk landscape? • Who’s helping our physicians navigate this new terrain? The CMS mandate isn’t just a federal announcement—it’s a workflow challenge, a compliance risk, and a reputational minefield. And most importantly, it’s already here. Let’s talk about how you’re planning for this. Or if you’re not, what’s in the way?

The healthcare data monopoly is about to crack open. After years of watching patients like my daughter Bailey struggle with 26 different portals that don't talk to each other, the policy winds are again shifting in our direction. CMS and ONC just dropped an RFI that reads like our business plan. They're asking the right questions about value-based care, data standards, and true interoperability. More importantly, they're signaling something bigger: we're at a crossroads between allowing natural monopolies to dominate healthcare or building genuine competitive ecosystems. The current administration is bringing back pro-transparency and pro-competition policies with renewed urgency. They understand what we've been saying: you can't unlock AI's potential in healthcare when data sits trapped in proprietary silos. As Amy Gleason rightfully said "We’re not rehashing the interoperability discussion, we’re here to finish it”! In this interview with David Raths, Healthcare Innovation, Jill DeGraff and I break down what it will take:  -      Enforce the Information Blocking and Interoperability mandates. While most providers and payers comply, the list of excuses is long for those who don’t.  -      EHR certification needs to move to the APIs that surround the EHR. Data AND Access APIs are required to create real value. Shoppability should look more like Google Flights than a guessing game. -      We need to cure portalitis. My daughter should be able to use her biometrics to access her complete medical record, just like she can use them to board a plane. Instead, she's locked out of pediatric portals because she aged out of their system. -      Continue the mandate on Bulk FHIR. If we really want to get all beneficiaries into a value based arrangement by 2030, providers must be able to access their OWN data at a population level. If the industry hopes to truly leverage AI to transform care, then data can’t be locked up any more. The infrastructure exists. The technology is ready. The question is whether we'll seize this moment to build the patient-centered, transparent healthcare system we've been promising for decades. Article link in comments ↓

📸 After my recent presentation on cloud adoption for healthcare, one of the most frequently asked questions was about the main roadblocks hospitals are facing in the Middle East as they consider cloud solutions. The key challenges? Legacy systems and internet dependency. Here’s what hospitals are up against: 1. Legacy systems: Many hospitals run older systems that aren’t cloud-compatible, leaving a tough choice: • Upgrading existing systems initially seems like the less expensive option. But it can quickly become a costly “sinkhole” due to hidden issues and expenses. • Adopting a new cloud-based system from the start is recommended. While it requires upfront investment, data transfer and process adjustments, this approach ultimately offers a more scalable and efficient setup. 2. Internet dependency and fiber optic issues: To address these, a hybrid cloud architecture is recommended. By setting up an on-premises data center that functions as a private cloud, hospitals can maintain full control over clinical and sensitive data. This setup allows cloud bursting—using third-party cloud resources for less sensitive systems and scaling up for sensitive EMR as needed. The hybrid approach gives hospitals both security and flexibility, expanding and contracting cloud resources as required without compromising data control. This hybrid model ensures sensitive data stays protected on-premises, while less critical data and overflow can leverage the cloud, balancing security with agility. #CloudAdoption #HealthcareInnovation #DigitalTransformation #HybridCloud #DataSecurity #HealthcareIT #MiddleEastHealthcare #LegacySystems #CloudComputing #HealthTech #EMR #DataPrivacy #Cloudfirst #Vision2030

For more than a decade, FHIR has been positioned as the backbone of modern healthcare interoperability. Yet adoption tells a more complex story. Fourteen years after FHIR’s birth, 79% of countries report national implementation guides, but only 20% report widespread use. This gap has led to a critical question. Are standards sufficient for healthcare’s long tail of specialised high coordination workflows? Conversational Interoperability, or COIN, proposes a different model. Instead of months of pre coordination, agents negotiate data exchange requirements in natural language and then execute the exchange using available tools and standards. At a recent HL7 Connectathon, agents were demonstrated handling referrals, prior authorisations, clinical trial matching and registry reporting in live conversational flows. One demonstration showed agents resolving a failed JSON exchange by switching to plain text and continuing the workflow, a self healing pattern rarely possible in rigid API integrations. Others included bi directional clinical trial matching and guideline driven decision support. However, governance remains unresolved. Identity verification and scope control are open questions. Food for thought. COIN does not replace standards, instead complements them. Standards reduce systemic complexity and COIN takes it a step further and reduces coordination friction. If interoperability is moving from static transaction exchange to dynamic negotiation, then COIN may not be the next replacement layer. It may be the missing coordination layer that sits above existing rails. Watch this space! Check out our blog for more: https://lnkd.in/eBJ7wvs9 #SharedCareRecord #EHR #COIN #Interoperability #HealthcareInnovation #HealthData #DigitalHealth Orion Health HEALWELL AI (TSX: AIDX)

It's time to get real: billions of dollars and 20+ years of health IT regulation has resulted in adoption of EHR systems with limited interoperability and data exchange capabilities, and has made it incredibly difficult to access complete patient records electronically. It's hard to convince providers to share across vendor networks by default, it's hard to break through digital roadblocks and end paper bridges, and it's hard to build new workflows that incorporate outside data and information. In a new article in the Journal of AMIA (American Medical Informatics Association) by Assistant Secretary for Technology Policy's Jordan Everson and Chelsea Richwine assesses the American Hospital Association's 2023 Health Information Technology Supplement survey, and find that most hospitals still experience at least one minor (81%) or major (62%) barrier to exchange, with the most common major barriers relating to different vendors and exchange partners’ capabilities. Rural and lower-resourced hospitals fared worse. Patient matching and cost to exchange were reported as major barriers. What works? Health Information Exchanges (HIEs), Health Information Service Providers (HISPs), and national networks. "...supplemental analysis indicated that use of HIEs was related to substantially lower rates of reporting barriers related to different vendor platforms, exchange partners, the need for customized interfaces, and data formatting. Use of national networks was related to lower rates of 6/8 barriers, with the strongest association with lower rates of barriers related to different vendor platforms, costs to exchange, and a need for customized interfaces."