User Experience Design for Healthcare

This is a must read for every HealthTech CEO. The UK Government’s AI Playbook outlines ten principles that ensure AI is used lawfully, ethically, and effectively. 1. Know AI’s Capabilities and Limitations AI is not infallible. Understanding what AI can and cannot do, its risks, and how to mitigate inaccuracies is essential for responsible use. 2. Use AI Lawfully and Ethically Legal compliance and ethical considerations are paramount. AI must be deployed responsibly, with proper data protection, fairness, and risk assessments in place. 3. Ensure Security and Resilience AI systems are vulnerable to cyber threats. Safeguards like security testing and validation checks are necessary to mitigate risks such as data poisoning and adversarial attacks. 4. Maintain Meaningful Human Control AI should not operate unchecked. Human oversight must be embedded in critical decision-making processes to prevent harm and ensure accountability. 5. Manage the Full AI Lifecycle AI systems require continuous monitoring to prevent drift, bias, and inaccuracies. A well-defined lifecycle strategy ensures sustainability and effectiveness. 6. Use the Right Tool for the Job AI is not always the answer. Carefully assess whether AI is the best solution or if traditional methods would be more effective and efficient. 7. Promote Openness and Collaboration Engaging with cross-government communities, civil society, and the public fosters transparency and trust in AI deployments. 8. Work with Commercial Experts Collaboration with commercial and procurement teams ensures AI solutions align with regulatory and ethical standards, whether developed in-house or procured externally. 9. Develop AI Skills and Expertise Upskilling teams on AI’s technical and ethical dimensions is crucial. Decision-makers must understand AI’s impact on governance and strategy. 10. Align AI Use with Organisational Policies AI implementation should adhere to existing governance frameworks, with clear assurance and escalation processes in place. AI in healthcare can be revolutionary if it’s done right. My key (well some) takeaways: - Any AI solution aimed at the NHS must comply with UK AI regulations, GDPR, and NHS-specific security policies. - AI models should be explainable to clinicians and patients to build trust. - AI in healthcare must be clinically validated and continuously monitored. - Having internal AI ethics committees and compliance frameworks will be key to NHS adoption. Is your AI truly NHS ready?

Why great processes still fail in digital health? On paper, everything worked, 🔸The patient showed up 🔸The doctor joined on time 🔸The consultation was completed The system logged it as a success! But here’s what really happened. 🔹The patient joined the call from a noisy food court 🔹He asked questions based on a TikTok video 🔹His phone died before the doctor could explain the prescription 🔹He took the wrong dose 🔹He ended up in the emergency department the next day The process (on paper) did not reflect reality (workflow)! 1️⃣ Processes are clean, logical, and rule-based 2️⃣ Workflows are messy, human, and context-dependent How we optimize them differs too: 1️⃣ Processes are often improved using methods like Lean or Six Sigma - focusing on efficiency, consistency, and waste reduction 2️⃣ Workflows, however, are best shaped through Design Thinking or Service Design - methods that center on behavior, experience, and context Why does this matter? Most digital health systems are designed around processes. But real-world care happens in workflows - where distractions, tech friction, and human behavior collide. Designing for real-world success? Start here: 🔸Ask: “What’s the patient actually doing during the consult?” 🔸Consider: Environment, attention span, device quality 🔸Anticipate: Interruptions, confusion, low health literacy 🔸Optimize using service-oriented methods, not just system logic If your digital health intervention only works when everyone behaves perfectly - it probably won’t work at all! What’s one real-world behavior or barrier that completely changed how you thought about “good” design in healthcare?  #HealthInnovation #WorkflowDesign #HumanCenteredDesign #SystemThinking 💡This post is part of 'Rethinking Digital Health Innovation' (RDHI), empowering professionals to transform digital health beyond IT and AI myths. 💡Find the ongoing series and resources on our companion website (URL in comments). 💡 Repost if this message resonates with you!

Following up on my earlier post about the AI medical app reportedly developed by a high school student Clarisse Poon in Hong Kong, I must raise urgent legal and ethical concerns that demand immediate clarification from all involved parties. Recent media reports indicate the app has tested with over 1,000 real patients, claiming zero errors and 100% accuracy. Such claims for a clinical tool built on AI require rigorous scrutiny, especially if it leverage Azure OpenAI Service and real patient data. This raises profound questions: - Were these 1,000+ patients explicitly informed that their highly sensitive health data might be uploaded to, stored by, or used to influence AI models hosted by third-party cloud providers like Microsoft? - How is a patient's fundamental "right to withdraw consent" or "right to be forgotten" handled when their data has potentially been used to train or influence an AI model? - Given potential international data transfer and the diverse citizenship of Hong Kong patients, what legal safeguards were in place to ensure compliance with frameworks like HIPAA, GDPR, and Hong Kong’s PDPO? In Hong Kong, accessing identifiable clinical data at this scale typically mandates IRB/Ethics Committee approval, formal hospital data agreements, and explicit, informed consent from every patient, fully complying with the PDPO. This leads to critical inquiries: - What rigorous processes allowed a secondary school student to gain such access to sensitive patient data? - Was the model subjected to independent audit, regulatory review, or clinical validation before its deployment with real patients and real prescriptions? Screenshots previously circulated by the developer reportedly showed real patient names, diagnoses, and ID numbers being input into AI models without clear evidence of prior anonymization. Such practices, if true, raise serious red flags under HIPAA, GDPR, and Hong Kong’s PDPO, frameworks that are non-negotiable safeguards for patient rights. According to the AI Health Studio website, the app's client is the Hepatobiliary-Pancreatic and Colorectal Surgery Centre. We need to understand: - Was this the direct source of patient data? - Was an independent ethics board consulted and did it approve this data's use? Once a system influences real prescriptions for real patients, it transitions from a school project to a clinical tool. As such, it must adhere to the highest clinical, legal, and ethical standards. I urge them to publicly clarify: - What comprehensive legal and ethical reviews were conducted for this project? - Precisely how patient data was collected, used, and stored throughout the app's development and trial phases? - Who bears ultimate responsibility for medical outcomes influenced by the Medisafe system? Medical AI is not a playground. Patient data is not a private asset. Trust must be earned through transparency and accountability, not through publicity alone.

A few deep thoughts on the announcement of ChatGPT Health's launch. First off, this felt inevitable. Not just as a product moment, but as the first tangible expression of something those of us working at the intersection of health and AI have felt coming for years. This isn’t incremental...it’s foundational. I remember sitting in a room over a year ago at Spring Health talking about the massive scales of fragmented health and behavioral data being created every day. It was so obvious that once we could securely connect medical records and bring connected health data together in AI, we would fundamentally change how people understand, navigate, and manage their health. This announcement feels like the start of that moment. ✨ So let's talk about why this matters so damn much: 💡 Health is already one of the most common use cases people bring to AI, with hundreds of millions of weekly health and wellness queries on ChatGPT alone. 💡 Healthcare as a system is widely experienced as broken, with 62% of Americans saying it isn’t working, and clinicians are burned out under current workflows. 💡 We now have a clear path to bridge scattered portals, apps, wearables, records, and conversations into something that can support better decision-making and continuity of care. But, y'all with great potential come serious responsibility. What we must take seriously: 👉 Accuracy & Safety. AI systems are pattern predictors, not clinical judgment engines. Without rigorous validation and guardrails, they can produce plausible but incorrect or misleading health information...a risk with real consequences. 👉 Ethics of Data & Vulnerability. Health data is among the most sensitive there is. As we connect records and personal wellness signals, we must lead with privacy, consent, clarity on use, and accountability. We should also be mindful, especially in the realm of mental health, of how AI can unintentionally reinforce harmful patterns or emotional dependencies without proper human oversight. The hard work starts now - in product, in policy, and in partnerships with clinicians and the people whose lives this tech touches. However, the opportunity to bring connected health intelligence into mainstream care in a way that truly serves is real, and it’s here. #healthtech #AI #ethics

A consultant asked me to write down "how a customer-first health insurance company would operate?" Here's my limited perspective, based on observation and interaction with customers over decades👇 1. Start with clarity. Every company should publish an open document that explains every complicated word and condition in simple language. No fine print. No room for interpretation. They should also list real-life scenarios and clearly show how each will be treated. No surprises at the time of claim. 2. Every policy should come with a short, personalized video explaining what’s covered, what’s not, and what to remember. 3. Only certified, trained people should be allowed to sell policies. They must pass an exam, follow a code of conduct, and face strict action if they don’t. 4. Health insurance should come with an option for a thorough medical check-up. If you take it, the company should guarantee your claim won’t be rejected for non-disclosure. Charge for the test if you want. 5. The proposal form should be Doctor AI voice led, to capture nuances, intricate details in the medical history. Make it easy for customers to be honest. 6. Pricing should be fully transparent. Customers should know how premiums are calculated, why they increase, and what to expect in the future. There should be clear guidance for senior citizens, with flexible and empathetic options to manage payments. 7. Proactive, preventive healthcare to ensure hospitalizations are avoided, creating a win-win situation for everyone - from customer to the insurer. 8. The company should be clear, open and social-media-first.Every question, complaint, or doubt deserves a human, clear response, not a template reply. 9. The claims process should be simple, transparent, and fair. Customers should know what is happening and what to expect at every step. 10. Grievances should not be hidden behind process. Every complaint must be answered, tracked, and used to fix the root problem. 11. The grievance team should fight for the customer, not the company. There should be an independent customer advocacy board or ombudsman inside the company, someone who can call out unfair treatment. 12. Once a year, the company should publish a public report showing claims, grievances, and how they were resolved. Let data, build credibility. Leaders should be accountable not just for profit, but for customer outcomes. Because in insurance, trust is the real product. An insurer built like this won’t need advertising. An insurer built like this won’t need to pay fat commissions. People will stay not out of compulsion, but because they feel safe.

LinkedIn 𝐢𝐬 𝐪𝐮𝐢𝐞𝐭𝐥𝐲 𝐫𝐞𝐝𝐞𝐬𝐢𝐠𝐧𝐢𝐧𝐠 𝐰𝐡𝐚𝐭 𝐯𝐢𝐬𝐢𝐛𝐢𝐥𝐢𝐭𝐲 𝐦𝐞𝐚𝐧𝐬 𝐨𝐧 𝐭𝐡𝐞 𝐩𝐥𝐚𝐭𝐟𝐨𝐫𝐦. For years, many creators believed reach was mainly driven by engagement signals. More comments, more reactions, more activity. In a recent update, LinkedIn feed leader Tim Jurka explained that the platform is evolving with ranking systems powered by generative recommenders and large language models. At first glance, this sounds like a technical update. The deeper shift is about what kind of content professionals actually see. These models analyze signals such as a member’s skills, experience, interests, and long-term engagement patterns to understand what professionals genuinely want to learn from. This means the Feed is gradually moving beyond popularity signals alone and moving closer to relevance, expertise, and authentic professional insight. LinkedIn has also made it clear that several behaviors that once artificially boosted reach are being reduced. Automated comments, engagement pods, recycled thought-leadership posts, and engagement bait such as “Comment YES if you agree” are becoming far less effective signals. Instead, the platform is prioritizing posts that reflect real experience, useful insight, and meaningful professional perspective. I have started noticing this shift in my own journey as well. When I began focusing my content more clearly around my niche, LinkedIn strategy and professional authority building, the nature of the conversations began to change. The discussions became more thoughtful, 𝐩𝐞𝐨𝐩𝐥𝐞 𝐬𝐭𝐚𝐫𝐭𝐞𝐝 𝐚𝐬𝐤𝐢𝐧𝐠 𝐝𝐞𝐞𝐩𝐞𝐫 𝐪𝐮𝐞𝐬𝐭𝐢𝐨𝐧𝐬, 𝐚𝐧𝐝 𝐦𝐚𝐧𝐲 𝐨𝐟 𝐭𝐡𝐞 𝐜𝐥𝐢𝐞𝐧𝐭𝐬 𝐰𝐡𝐨 𝐫𝐞𝐚𝐜𝐡 𝐨𝐮𝐭 𝐧𝐨𝐰 𝐦𝐞𝐧𝐭𝐢𝐨𝐧 𝐭𝐡𝐚𝐭 𝐭𝐡𝐞𝐲 𝐝𝐢𝐬𝐜𝐨𝐯𝐞𝐫𝐞𝐝 𝐦𝐲 𝐰𝐨𝐫𝐤 𝐭𝐡𝐫𝐨𝐮𝐠𝐡 𝐩𝐨𝐬𝐭𝐬 where I shared practical insights from real experience. That reinforced something important for me. Trust on LinkedIn rarely grows from viral tactics. It grows when professionals consistently 𝐬𝐡𝐚𝐫𝐞 𝐩𝐞𝐫𝐬𝐩𝐞𝐜𝐭𝐢𝐯𝐞𝐬 𝐭𝐡𝐚𝐭 𝐡𝐞𝐥𝐩 𝐨𝐭𝐡𝐞𝐫𝐬 𝐮𝐧𝐝𝐞𝐫𝐬𝐭𝐚𝐧𝐝 𝐬𝐨𝐦𝐞𝐭𝐡𝐢𝐧𝐠 𝐦𝐨𝐫𝐞 𝐜𝐥𝐞𝐚𝐫𝐥𝐲. Which aligns closely with the direction LinkedIn appears to be moving. Professionals come here to learn from other professionals, not from automated conversations. And the Feed is increasingly being designed to reinforce that. My biggest takeaway from this update is simple. 𝐕𝐢𝐬𝐢𝐛𝐢𝐥𝐢𝐭𝐲 𝐨𝐧 𝐋𝐢𝐧𝐤𝐞𝐝𝐈𝐧 𝐢𝐬 𝐠𝐫𝐚𝐝𝐮𝐚𝐥𝐥𝐲 𝐬𝐡𝐢𝐟𝐭𝐢𝐧𝐠 𝐟𝐫𝐨𝐦 𝐚𝐜𝐭𝐢𝐯𝐢𝐭𝐲 𝐬𝐢𝐠𝐧𝐚𝐥𝐬 𝐭𝐨 𝐥𝐞𝐚𝐫𝐧𝐢𝐧𝐠 𝐬𝐢𝐠𝐧𝐚𝐥𝐬. As LinkedIn continues evolving its Feed around authentic professional conversations, the real question becomes: Are we creating content to trigger engagement, or sharing insights that professionals genuinely want to learn from? Source: LinkedIn Feed Update – Tim Jurka LinkedIn News India LinkedIn News #LinkedIn #ThoughtLeadership #FutureOfWork #PersonalBranding

AI scribes didn’t just make notes faster- they cut burnout. New study in JAMA Network Open (Oct 2, 2025): after just 30 days with an ambient AI scribe across 6 health systems and 263 ambulatory clinicians, burnout fell from 51.9% to 38.8% (adjusted OR 0.26). What changed: • Cognitive load dropped (−2.64 on a 10-point scale) • After-hours charting shrank by ~0.9 hours/week • More undivided attention for patients • Slightly clearer care plans for patients reading notes • Easier to add urgent slots when needed Ambulatory medicine is cognitive work. For me, analyzing a patient's history, deeply listening to them to find clues on diagnosis and to hear whats not being said, can be more mentally taxing then a full day in the OR. Anything that decreases cognitive load, helps free up bandwidth to focus on getting the patient better Every minute reclaimed from the EHR is a minute returned to clinical reasoning, patient connection, and recovery time the real levers against burnout. Caveats (and still promising): It’s self-reported, no control group, and only 30 days, but the signal is strong and consistent with smaller pilots. The next step is measuring. Track EHR time, after-hours work, inbox load, note quality, and patient access before/after deployment. My take as a practicing GI: The win isn’t “AI writes the note.” It’s that clinicians think more, click less, and patients get clearer plans. That’s the bridge from potential to practice. #AIinHealthcare #DigitalHealh #PhysicianBurnout #HealthInnovation

It doesn't matter how amazing your benefits package if your team doesn't use it. I've learned that what I value might not be the same as what my team values. As I shared on Episode 136 of "Build to Enough," at Little Fish, I've implemented unique benefits that make my employees feel valued while also recognizing that they are human. For example, I offer "Sick and Sad Days"—time off that isn't counted against anyone if they're sick or just can't do it that day. I wanted to ensure they have room to take time off when they aren't at their best. We also close for five weeks out of the year: one week during spring break for tax season, one week at the end of summer, and two weeks at the end of the year. These breaks are automatically built in and fully paid for everyone. We offer flexible work hours with some overlapping core hours, but they can work at a time that suits them best. Plus, we have an annual all-expenses-paid company retreat, a 401k match, and internet reimbursement. Now, I didn't start with all of this. Bit by bit, I figured out what made the most sense for the business and what the team actually wanted. If you're looking to develop a benefits package that truly supports your team, here are some steps to consider: 1. Assess your team's wants and needs - Ask them what they value and what perks would make a difference in their lives. 2. Prioritize core benefits - Focus on essentials like PTO, health benefits, and retirement plans, but don't forget to explore other perks. 3. Research your options - There are many health and retirement plans available for small teams. Do your homework to see what will work best for your team (and your budget 😉 ). 4. Consider supplemental benefits - Look for inexpensive perks that have a significant impact, like flexible hours or remote work options. 5. Maximize your budget - Allocate a specific amount for benefits and make the most of it. Seek group buying opportunities and tiered benefits to offer more without overspending. 6. Review and adjust regularly - Benefits aren't a set-it-and-forget-it deal. As your team evolves, so should your benefits package. Creating a benefits offering that truly supports your team not only helps retain your current employees but also makes your company a place where people want to work.

What if AI could give clinicians back 8 hours a week? Administrative work consumes nearly a third of healthcare professionals' time. Documentation, Scheduling, Revenue cycle management, Tasks that pull clinicians away from what matters most: patient care. AI changes this equation dramatically. Imagine walking into your practice and finding your notes already drafted from patient conversations. Picture calendar conflicts resolving themselves automatically. Envision billing cycles completing with minimal human intervention. This shift does more than save time. It transforms healthcare delivery at its core. Clinicians reconnect with their original calling when freed from paperwork. Patient interactions become more meaningful. Treatment plans receive proper attention. Medical decisions improve with reduced cognitive load. Healthcare organizations benefit too. Resources flow to direct care instead of administrative overheads. Operational costs decrease while quality metrics rise. Staff retention improves as job satisfaction grows. The math becomes compelling. Eight reclaimed hours weekly translates to hundreds of additional patient interactions monthly. Those interactions build stronger therapeutic relationships and drive better health outcomes. Burnout rates fall when administrative burdens lift. Clinicians report renewed passion for medicine. Teams collaborate more effectively without documentation demands draining their mental bandwidth. AI handles the routine. Humans handle the human. The technology exists today. Forward-thinking healthcare organizations already implement these solutions. Early adopters report significant improvements in both clinician wellbeing and patient satisfaction scores. The question becomes less about if we should embrace AI for administrative tasks and more about how quickly we can responsibly implement these transformative tools. Your patients deserve your best. Your practice deserves efficiency. You deserve to practice medicine rather than manage paperwork. What would you do with those eight extra hours?

Digital health promises transformation but it also raises deep ethical questions. A new perspective article argues that the principle of justice must guide how we design and deploy digital health. The authors remind us that equality, equity and justice are not the same. Equality gives everyone the same resources, equity adapts resources to individual needs, and justice goes further by addressing structural barriers that exclude people in the first place. Key insights from the paper: 1. Digital determinants of health matter: Access to connectivity, digital literacy, algorithmic bias, and trust are as important as traditional social determinants of health. 2. Justice requires more than access: Providing devices or portals is not enough. Structural issues like inaccessible design, digital deserts, and biased algorithms can perpetuate exclusion unless actively corrected. 3. Vulnerable groups must be included: Older adults, people with disabilities, language minorities and those with low digital literacy are among the heaviest users of health systems yet the most at risk of exclusion. Co-creation and participatory design are essential. 4. Policy and practice must integrate ethics: Justice in digital health requires equity assessments, digital facilitators to support patients, literacy programs, and collaboration across sectors such as health, education and technology. Digital health is not just a technical or clinical transformation, it is an ethical one. Justice must be the guiding value to ensure that digital innovation closes gaps rather than widening them. #DigitalHealth #HealthEquity #Bioethics #PatientEngagement #HealthInnovation #JusticeInHealth #HealthIT #DigitalInclusion #Techquity #HealthcareTransformation https://lnkd.in/d6TxRU2F