Sabbir Hossain

The Fragmentation Problem in Healthcare Program Integrity (some thoughts after attending the HPRI Chicago 2025) One thing that struck me at HPRI Chicago was how payment integrity and FWA teams often don't talk to each other. They're frequently reporting to different leaders, working on what are really sub-puzzles of one problem, but approaching them from completely different angles. Part of what keeps these teams apart is all the jargon we've built up around what should be straightforward work. Pre-payment, post-payment, claim edits, prior authorization, utilization management, data-mining - these terms have become more like barriers that facilitate the divide between teams and technologies and point solutions. I think in the end, the task is as simple as: find things that don't look right, and if they're not right, make them right. The level of expertise in this field ranges widely - from people who mostly repeat buzzwords to those who actually understand the underlying problems. It's always enlightening to find people I'd like to learn from and showing lots of their passion, but I think the average level isn't quite there. Correction. Let's say it is not that they are not good at what they do. They know their domain well enough, but it's hard for them to cross over into other areas. The real difficulty of having a comprehensive view of program integrity is that it requires a combination of medical expertise, policy expertise, analytics, informatics, and cybersecurity. It's often difficult to excel in just one vertical, but program integrity demands all of these at a high level. Sometimes this work is run by medical officers who understand data and policy, sometimes by data experts who grasp the other pieces. But finding that combination isn't easy. The industry's immaturity shows in the lack of senior leadership dedicated solely to program integrity at most payers. This absence of executive accountability reflects how the field still treats program integrity as separate functions rather than an integrated discipline requiring unified leadership. While CMS and other regulators are pushing for more integrated approaches to program integrity, the fragmented structure makes it harder for organizations to respond cohesively to regulatory expectations and reporting requirements. This makes me think there may need to be a leadership position called Chief Program Integrity Officer - someone who oversees the correctness and compliance of all transactions and records. It's a bit like an extended concept of Chief Privacy Officer, but broader. They'd need to understand the business of health insurance regarding policy and medical practices and data, and most importantly, they'd need to understand how it all should work together. What are your thoughts? Is it about time to have Chief Program Integrity Officer in payers? Image credit: Gemini

13

5 Comments

88% of patient referrals are still booked manually. Only 35% ever become closed-loop referrals today. This is a massive level of inefficiency & referral leakage that is simply accepted as the norm, today. Not for long… Intake, referrals & scheduling are being permanently disrupted by agentic automation technology. Had such a great time Becker's Healthcare in Chicago, discussing the hottest trends within healthcare tech for providers! #healthcare #referrals #rcm

15

HealthCare CX strategy lives. When UX sits inside AI decisioning and data orchestration, complexity is absorbed by the system — not pushed onto agents to manually connect dots across fragmented tools, rules, and workflows. In healthcare payers, most CX pain shows up downstream. But the causes are upstream. Eligibility. Benefits clarity. Claims. Prior auth. System handoffs. That’s why ~55% of member calls still tie back to benefits and claims across medical, dental, vision, and pharmacy. Here’s the part many teams miss 👇 Small systemic changes — like call-note automation or a first learning agent with smart links — can return 20–35%+ time to service teams. That time isn’t just efficiency. It’s how organizations: • scale faster • support more members • without linear cost growth So the real question isn’t “Do we need better CX?” It’s this: Where is your team losing the most time each week — and is the system addressing it, or are people compensating for it? CX doesn’t scale through heroics. It scales when systems do the work. #HealthcareCX #MemberExperience #HealthTech #AIinHealthcare #UXStrategy #PayerOperations #ServiceDesign #OperationalExcellence #DigitalTransformation

4

The HITECH Act (2009) reshaped digital health in the U.S. Before HITECH, hospital EHR adoption was only ~10–20%. Today, it’s above 96%. What drove this massive shift: • USD 30B in incentives to adopt certified EHRs. • Meaningful Use standards to ensure better care, not just digital paperwork. • Penalties for staying non-digital. • Stronger HIPAA rules for data privacy and breach reporting. • A national push for interoperability through HIEs. • ONC empowered to set standards and certify systems. • Workforce programs to build health IT talent. • Funding that accelerated innovation across the ecosystem.  With India’s hospital EHR adoption still around ~35%, is it time for a similar nationwide push? #DigitalHealth #HealthIT #HITECH #EHR #Interoperability #HealthcareInnovation #HealthTech #MeaningfulUse #ONC #HIPAA #HealthTransformation

11

1 Comment

Most scheduling APIs are built to serve clicks. AI agents need APIs that can negotiate. This week we went deep with a large healthcare network that’s designing a brand-new internal scheduling API specifically for voice automation. Not a PoC. A production-grade interface that will soon handle real patient traffic. What mattered most in that discussion wasn’t JSON structure or field naming. It was this: Where do you place the intelligence - in the API, or in the AI? A few very real, very practical lessons from that session: 1️⃣ Returning a single “best” slot kills conversation quality If the API only ever returns one pre-booked slot: - the AI can’t show alternatives, - can’t reflect patient trade-offs, - can’t negotiate. You end up with robotic flows like: “8:00 is unavailable → transferring to an agent.” When the API instead returns a small pool of available slots (not pre-booked yet), the AI can: - group options (“I see several openings between 10 and 13”), - adapt to rejections in real time, - iterate without bouncing the patient to a human. In production, this difference alone can decide whether a 13-minute complex scheduling call ends in success or failure. 2️⃣ Pre-booking should be surgical, not global Pre-booking everything the API returns feels “safe”… and quietly destroys flexibility. What scales better: - Fetch a wider set of available slots. - Pre-book only the specific slot(s) the AI actually reads out loud. That mirrors how real receptionists work. And it prevents the system from blocking inventory during long negotiations. 3️⃣ Time filters should be anchors, not cages “Morning / afternoon / evening” buckets seem helpful. Until a patient says: “Something around 16:00.” If your filter closes at 15:45, the AI is forced to say: “There’s nothing available.” If your filter is an open anchor (“search from 12:00 forward”), the AI can instead say: “I don’t see anything exactly at 16:00, but the closest is 16:20.” That single sentence is often the difference between: - conversion, and - escalation to a human. 4️⃣ APIs should stay dumb about policy - on purpose Cancellation rules, protection against critical specialist slots, double confirmations via SMS - all of that belongs in: - the AI layer, or - a domain orchestration service. Not hard-coded into the API. Because the moment clinical policy changes, you don’t want to wait for a new backend release to stay safe. ⸻ The pattern that keeps repeating across our deployments is simple: - Rigid API → fragile AI. - Flexible API → adaptable AI. If you’re designing a scheduling API today and expect AI agents to use it tomorrow, the real question isn’t: “Does it support appointment booking?” It’s: “Does it support negotiation?” That’s where real automation either works - or quietly collapses.

24

2 Comments

It was a true privilege to be a panelist on the Inaugural AWS Healthcare Leaders Roundtable — the latest evolution in AWS's ongoing Healthcare and Life Sciences Roundtable series and sponsored by their brilliant partner, Drata. This session brought together an exceptional group of healthcare leaders, entrepreneurs, and thinkers to explore one central question: How do we responsibly and effectively bring the transformative potential of Generative AI to patients — starting with public healthcare systems like the NHS? From system-wide innovation to patient-centred implementation, the conversation was as energising as the food was delicious. A few key themes emerged: 🔹 The NHS needs AIML innovation key in areas like reducing waiting lists and improving workforce productivity — foundational, not flashy. These are high-impact challenges where agile StartUps can make a real difference. 🔹 Healthcare must find its pace. While Life Sciences races ahead, public health systems operate in a complex, regulated, and often fragmented environment. Can we reconcile the need for speed with the imperative for safety and equity? 🔹 “Move fast and break things” doesn't apply here. In healthcare, innovation must be steady, thoughtful, and deeply aligned with clinical realities. 🔹 StartUps are rising to the challenge, navigating the ethical, regulatory, and operational complexities with real ingenuity. 🔹 And above all — the patient voice must remain central. Generative AI should not just enhance efficiency, but genuinely improve the patient journey and outcomes. Huge appreciation to AWS executive speakers, and Healthcare & Life Sciences experts, Boris Bogdan and Daniel Hickmore for their leadership and insight. And to the brilliant StartUp leaders in attendance — thank you for fuelling such a lively, inspiring, and solution-focused discussion. Onwards! Here’s to more dialogue, more collaboration, and above all, more innovation that truly serves patients.

27

📊 What’s the biggest pain point in health tech hiring right now? We asked. You answered. Here’s what the poll told us ⬇️ 🔘 38% – AI, Data & Cybersecurity talent is too specialized to find easily 🔘 29% – DevOps & Health IT roles are stalling infrastructure growth 🔘 18% – PMs and Compliance roles often get overlooked—until it’s too late 🔘 15% – Genomics & Precision roles are expanding faster than hiring plans 🎯 Key Takeaways: ☛ Clients are struggling to attract specialists who can scale ☛ Candidates want transparency, technical clarity, and purpose ☛Everyone feels the pinch on roles that combine tech + compliance 💬 What’s missing from this list? Drop your thoughts or tag someone building in this space. In the coming weeks, we’ll explore candidate red flags and how to design hiring processes that don’t scare away the good ones. #HealthTech #HiringChallenges #DigitalHealthTrends #TalentAcquisition #RecruitmentInsights #LinkedInPoll

13

3 Comments

Why Small Language Models (SLMs) Are Powering High-Precision Prior Authorization in Healthcare Prior authorization remains one of the most expensive and friction-heavy workflows in healthcare. The challenge isn’t lack of data — it’s precision. Large, general-purpose language models are impressive, but prior authorization demands something different: deep domain understanding, deterministic behavior, and consistent outcomes across payer-specific rules. This is where Small Language Models (SLMs) are proving to be far more effective. SLMs trained specifically on prior authorization workflows, payer policies, CPT and ICD mappings, medical necessity rules, and historical authorization outcomes can operate with significantly higher precision. They reduce hallucination risk, handle edge cases more reliably, and align better with compliance and audit requirements. More importantly, SLMs enable intent-driven automation. They can identify authorization requirements early, extract the exact clinical evidence needed, assemble payer-specific submissions, and flag gaps before a request ever reaches the payer. The result is fewer delays, fewer resubmissions, and faster patient access to care. Another advantage is deployability. SLMs are lighter, faster, and easier to operate in private or hybrid environments — a critical requirement for healthcare organizations that care about data control, latency, and regulatory boundaries. At Contiinex, we see prior authorization evolving from a manual, reactive process into a predictive, AI-assisted workflow powered by specialized SLMs — not generic models. #PriorAuthorization #HealthcareAI #SLMs #RevenueCycleManagement #DigitalHealth #Contiinex

11

Here’s your recap of last week’s health IT partnership and product release news. 🗞️ 👇 (Part 1 of 2 - this was a busy week for health IT announcements!) 🤝 𝗣𝗮𝗿𝘁𝗻𝗲𝗿𝘀𝗵𝗶𝗽 𝗻𝗲𝘄𝘀: • UMass Memorial Health is expanding Mednition KATE AI system-wide. • Stanford Health Care is using Atropos Health to surface personalized evidence at the point of care.    • MetroHealth System selected Pieces for inpatient and outpatient settings.    • TriHealth selected Reveal-Dx and announced a clinical study to evaluate real-world impact.    • Cleveland Clinic announced it would be using AKASA's coding AI assistant and CDI tool.    • Allegheny Health announced its first "touchless" prior auth with Humata Health.    • Sentara Health is expanding its use of Andor Health.    • Albany Med selected Honest Health as its ACO enabler. 🚀 𝗣𝗿𝗼𝗱𝘂𝗰𝘁 𝗡𝗲𝘄𝘀: • Videra Health launched AI-powered Group Notes • Cedar released an agentic AI solution to handle patient billing calls. • Relatient launched a conversational AI feature for automating patient scheduling calls. • Zocdoc released Zo, a voice AI agent that handles  inbound scheduling calls. • CliniComp integrated enterprise-wide PACS with its Intrinsic AI solution suite. • Viz.ai announced Viz 3D CTA, which converts computed tomography angiography scans into AI-enhanced 3D images for neurovascular visualization.    • Limbic launched Limbic Intake Agent, a voice AI agent that handles patient intake. • CoreCare launched Pre-Admit to help SNFs assess referrals. • NeuroFlow published new research showing the efficacy of its solution in recognizing suicidal ideation. • Get Well announced an AI patient assistant to support inpatient care navigation and discharge.    • Digital Medicine Society (DiMe), Duke University, University of North Carolina at Chapel Hill, Fitbit (now part of Google), + ŌURA announced a collaborative aimed at helping prevent relapse and reduce deaths from opioid use disorder using wearable technology. 𝗥𝗲𝗺𝗶𝗻𝗱𝗲𝗿: you can always catch the full details of each of these announcements in our weekly Elion Briefing (sign up at Elion [dot] health [slash] newsletter).

39

2 Comments

Delivery models are rarely replaced; they are layered. ​Just as Scrum wrapped around Waterfall cores, we are now seeing a new layer emerge as AI moves from the periphery to the center of product strategy. ​When AI starts influencing outcomes, "delivery speed" loses its status as the primary metric. The new challenge is behavioral integrity. ​Scrum defines the build. N8–GEMS defines the life of the system thereafter. ​N8 shifts the mindset to AI as a living entity. ​GEMS provides the governance and observability required for trust. ​We’re no longer just building software; we’re stewarding systems that learn. It’s complex, high-stakes, and happening mid-flight. ​Which of these feels closest to your personal voice?

33

1 Comment

Is Healthcare ready for its "Cursor" moment? 🏥✨ Y Combinator’s latest Request for Startups (RFS) introduces a fascinating concept: "Cursor for Product Managers." While the last year focused on AI helping engineers write code faster, YC is now calling for tools that help teams figure out what to build in the first place. In no industry is this "discovery gap" more critical—or more difficult—than in Healthcare. The Problem: The "Lost in Translation" Tax In healthtech, the distance between a clinician’s need and a shipped feature is massive. Spending months: - Synthesizing clinical trial data and user feedback. - Writing dense PRDs and HIPAA-compliant requirements. - Manually mapping workflows for overstressed nurses and doctors. The Vision: AI-Native Healthtech Discovery Imagine an AI-native system where you upload patient journey maps, clinician interviews, and EHR usage data. Instead of just a summary, the tool: - Identifies the Friction: "Doctors are abandoning this screen because the triage logic doesn't match their bedside workflow." - Proposes the Solution: It generates a new UI wireframe and data model optimized for speed-of-care. - Ships to Agents: It breaks the design into tasks for coding agents to implement immediately. Why it Matters Building the "wrong thing" in healthcare isn't just a waste of VC capital—it’s a waste of provider time and, ultimately, affects patient outcomes. By automating the loop of discovery → synthesis → proposal, we can finally build software that works with clinicians rather than getting in their way. The frontier of AI isn't just about faster code; it's about deeper empathy for the end user. Healthcare Innovators: Are we ready to start building the "Cursor for PMs" in the healthcare space? I’d love to hear how you’re tackling the complexities of clinical workflows. #HealthTech #YCombinator #ProductManagement #DigitalHealth #AI #Innovation

14

8 Comments