Healthcare Analytics Tools

Working with an interdisciplinary team, we have developed a website to communicate how the White House's proposed cuts to health research would cause losses of $16B and 68,500 jobs. Find out how your community may be impacted at SCIMaP: https://scienceimpacts.org As context, on Feb. 7th, 2025, the White House ordered across-the-board cuts to NIH funded research. The order drastically reduces the amount that universities/hospitals/institutes receive for essential facilities, services, and staff required for health research. Nearly two dozen states and allied institutions sued leading to a temporary injunction to across-the-board cuts nationwide. The NIH distributes approximately $37B in external grants/awards in FY 2024. These grants/awards have a force-magnifying effect, generating $2.56 of economic activity for each $1 supported, i.e., more than $94B in activity and more than 400K jobs (source: United for Medical Research). But this impact is hard to see and interpret. You might wonder: perhaps the impacts are focused only on a few, potentially 'elite' institutions? The answer is far different. Soon after the executive order was released, it became apparent that these across the board cuts would have damaging & consequential effects in communities across the United States, in places like State College, PA, Birmingham, AL, and across the medical research infrastructure of Texas. Led by the efforts of Allie Sinclair joint with Emily Falk, Clio Andris and more, we have developed an interactive visualization of the impact of federal cuts to health research in communities nationally. In practice, we take anticipated reductions in NIH supported grants and then leverage US census data on commuting to project the impact of these cuts across and within communities. Through interactive, data-driven visualizations, we aim to help Americans explore how research fuels the economy, supports jobs, and improves health outcomes. This website and interactive visualization is a step in that direction, with more to come joint with Alyssa (Allie) Sinclair (now at UPenn), Emily Falk (UPenn), Clio Andris (GT) + others in The Science and Community Impacts Mapping Project: https://scienceimpacts.org

🚑 Heart Failure Analysis using Power BI In this data-driven project, I built an interactive Power BI dashboard to uncover critical health insights from patient data. 🔍 Key Insights Unlocked: KPI Overview: Tracked key metrics like Survival Rate, Average Age of Survivors, Total Survivals & Deaths to understand patient outcomes. Age-Based Trends: 📊 Survival Count & Avg Serum Creatinine by Age Group: Revealed how kidney function markers correlate with survival. 📉 Survival Count & Avg Ejection Fraction by Age: Showed how heart function varies with age among survivors. 📈 Survival Rate by Age Group: Identified age-related risk patterns in patient outcomes. Ribbon Chart Analysis: Assessed the influence of smoking, high blood pressure, diabetes, and anaemia on different age groups using visual ribbons for intuitive risk comparison. 🧠 Dynamic Filtering: Used slicers to explore how survival trends differ by gender. 🔧 Tools Used: Power BI, DAX, data modeling, and advanced visualizations. ✅ The project not only enhanced my skills in healthcare analytics but also taught me how to communicate health risks effectively through interactive visuals. #powerbiproject #dataanalytics #visualizations #powerbi #datacleaning #dataanalysis

The rural health care ecosystem is fragile, and we are starting to better understand the devastating impacts the One Big Beautiful Bill Act would have on rural communities, specifically. We know that Medicaid plays an outsized role in rural communities, covering a larger share of children and adults than in urban communities. In fact, nearly half of all children and 1 in 5 adults living in rural communities rely on Medicaid or CHIP for their health insurance. According to a recent analysis conducted by Manatt, Phelps & Phillips, LLP, over 1.5 million rural residents (1 in 10 of currently covered individuals) would lose their health care coverage through proposed changes to Medicaid. Proposed changes to the Marketplace plans would only add to the coverage losses for rural communities. Combined, these proposed changes would have a negative impact on millions of people in states like Kansas, Missouri, Oklahoma, Maine, North Carolina and others.  What hasn't been well understood until today, is what are the economic impacts of physicians leaving rural communities. New analysis from the Robert Graham Center finds that the loss of a rural primary care physician results in an average $1,348 increase in health care expenditures for each rural patient. Meaning that the health care costs for a family of four will be $5,000 higher than they would be if a family physician was in the community. Preserving rural hospitals is important, but PRESERVING RURAL FAMILY PHYSICIANS IS A HEALTH CARE AND ECONOMIC IMPERATIVE.

What happens to hospitals when millions lose Medicaid coverage? The financial fallout is already beginning. --- As federal work requirements for #Medicaid expansion populations move toward implementation, most headlines focus on potential coverage losses. A new analysis highlights another angle: The financial impact on hospitals. --- If work requirements are fully implemented in all expansion states by 2027, the analysis estimates #hospitals could see their operating margins drop by 11.7% - 13.3% on average. For safety-net hospitals, the financial hit could be 25.9% - 29.6%. Rural safety-net hospitals (already under strain even without work requirements) are especially vulnerable since many already have negative margins. --- According to the analysis, 5.1-5.8 million people are expected to become uninsured, increasing uncompensated care by up to $8 billion. Many more will also be shifted to commercial coverage. Safety-net hospitals could lose up to $2.8 billion in net operating income. Reduced revenues and higher costs may force hospitals to cut services or staff, impacting access to care for entire communities, not just Medicaid enrollees. The Affordable Care Act’s Medicaid expansion helped stabilize hospital finances. The Medicaid work requirements #policy could reverse that progress in many states, particularly where hospitals rely heavily on Medicaid revenue. --- Many states are already deep in the weeds of implementing these work requirements to make them minimally burdensome while still meeting the policy mandates. How will health systems adapt to the financial repercussions? Some have already given up and plan to close... No matter what your opinion is on work requirements, it’s already impacting hospitals.

People don’t “fail” healthcare. They’re often just blocked by social conditions that healthcare hasn’t accounted for. This isn’t about noncompliance. It’s about no transport, no childcare, no stable housing, no trust, no safe time to be sick. These are social problems. And they’re the real reasons people can’t engage with care—not ignorance, not laziness, not cost alone. 📉 Yet clinical systems still design pathways assuming that if we just offer more services or spend more money, outcomes will improve. But health doesn’t happen in the clinic. It happens in the neighborhood, long before (or if) someone ever walks through the door. So what needs to change in boardrooms? Too often, power and control sit far from the lived realities of patients. We need: - Investment in community-rooted care models, not just institutional ones - Inclusion of neighborhood voices in decision-making - Metrics that reflect social impact, not just throughput or compliance - Systems that accommodate, not punish, social complexity If healthcare wants different results, it must design for the world people actually live in—not the one we assume they do. It’s not about “fixing patients.” It’s about dismantling the barriers to what health can offer people #SocialDeterminants #NeighborhoodHealth #HealthEquity #SystemicChange #ClinicalPathways #PublicHealth #BoardroomToStreet #PowerShift #PatientCentredCare

🔔 My latest visualisation on Coronary Heart Disease (CHD) - Heart Disease Intelligence. 🫀 As a Tableau/Power BI Developer and BI Consultant, I’m always looking for ways to turn complex datasets into dashboards that are not only visually engaging, but genuinely useful for decision-making. CHD is a great example of a condition where risk builds gradually over time — influenced by factors like age, blood pressure, cholesterol, smoking behaviour, BMI, diabetes, and glucose — often long before symptoms appear. In this project, I built a multi-dashboard Tableau report, based on synthetic data from Kaggle that includes: ➡️ A high-level population risk overview (age bands, gender split, risk distribution) ➡️ A risk factor impact panel using a dumbbell comparison chart (to highlight risk gaps clearly) ➡️Clinical deep dives using boxplots and risk matrices ➡️ A hotspot view using an Age × Systolic BP risk heatmap for early risk identification The focus throughout was on building an experience that feels clean, fast, and insight-led - the kind of report that a stakeholder can actually use, not just admire. This report is designed to raise awareness of common CHD risk patterns and support data-driven discussion — it is not a clinical diagnostic tool and should not replace professional medical assessment. Link to viz: https://lnkd.in/eZdziRe7 #Tableau, #BusinessIntelligence #DashboardDesign, #DataVisualization, #HealthcareAnalytics, #BIConsulting, #Analytics, #DataStorytelling, #datafam , Tableau , #tableaupublic , #hearthealth , #heartdisease , #CVD , #CHD , #heartattack , #cardiology , #fitness , #stroke , #cardio , #heartfailure , #wearredday , #health , #uiux , #uidesign , #infographic

Scrolling through Tableau Public like I usually do, I came across this killer dashboard by Waqar Ahmed Shaikh — and I couldn’t help but share because it’s loaded with smart design choices that actually make the data easier to digest. 🧠💥 A lot of people think a good dashboard is about cramming in as many charts and metrics as possible, but it’s really about guiding the viewer’s attention to what matters most. So, here’s why this one stood out: 1️⃣ Card Layout: Breaking data up into card-style visuals isn’t just about aesthetics. It’s about creating mental compartments for your audience. It segments complex information into digestible bites, making it easier for anyone — even someone completely new to the data — to follow along. If your dashboard looks like a tangled mess, it is a tangled mess. 2️⃣ Heat Map for Day & Hour Analysis: This is pure brilliance for time-based insights. Heat maps visually show the frequency of events over time, making it easy to spot trends and outliers. In this case, it highlights hot spots in patient treatment patterns. Imagine trying to sift through hundreds of rows in a spreadsheet to find these patterns—good luck with that. Instead, the visual tells you everything in seconds. 3️⃣ High-Low Dots in KPI Spark Lines: This is what I call a "shortcut to insights." It’s not just about showing trends but immediately pointing out where the highs and lows occurred. This way, you don’t waste time digging into what’s normal and what’s not. High-Low dots say: “Here’s where you should be paying attention.” This is exactly the kind of detail that separates a good dashboard from a great one. Moral of the story? Design for impact, not just for the sake of being flashy. Waqar’s dashboard does this well, and it’s a reminder that visuals are tools for clarity, not confusion. If you want to elevate your dashboards, check out his work. It’s a solid benchmark to measure your own designs against. 🔥 #DataVisualization #Tableau #DataDesign #DashboardDesign

I spent years creating "beautiful" dashboards that executives ignored. Then I discovered 4 strategies that turn complex charts into decision drivers. Here's how to make your data impossible to ignore: It all started with an insight from Storytelling with Data by Cole Nussbaumer Knaflic. Your tools don't know your story. You must bring it to life. 𝗕𝗲𝗳𝗼𝗿𝗲: Hours creating fancy charts with gradients and random colors. 𝗔𝗳𝘁𝗲𝗿: Simple visuals that stakeholders actually use. 4 Core Visualization Principles: 1. Strip Chart Junk ↳ Remove unnecessary gridlines ↳ Delete pointless labels 2. Focus Single Message ↳ One insight per chart ↳ Everything else creates noise 3. Strategic Color Usage ↳ Highlight only critical data ↳ Gray out supporting information 4. Clear Takeaways ↳ State conclusions upfront ↳ Make messages obvious The transformation results in improved attention, understanding, and taking action. Your Implementation Plan: 1. Delete pointless gridlines 2. Remove unnecessary labels 3. Choose one color for key highlights 4. Write titles that state your conclusion Small adjustments create a massive impact. Which visualization principle will you implement first? Share your approach below! 📚 Resource: Storytelling with Data: https://amzn.to/4fHenmA ♻️ Repost to help others create impactful data stories

Trend analysis isn’t just math. It’s your early warning system for patient safety.   What most quality teams get wrong? They default to familiar methods.   Across hundreds of device failures and safety signals in MedTech, the pattern is clear when teams pick the wrong method:   Real problems stay hidden.   While everyone else drowns in false positives, you could be catching:   → Gradual quality degradation before a recall → True compliance shifts, not random noise → Critical safety patterns your competitors miss   But each method has blind spots. Some are major.   Without matching method to purpose, you’ll face:   • Wasted investigations on false alarms • Missed signals in complex data • Flawed forecasts driving bad decisions • Patient risks hiding in plain sight   Here’s your method selection playbook: 1️⃣ P-Charts – For Compliance Monitoring  → Ideal for tracking incident rates over time  → Visual spikes = instant red flags  → ⚠️ Needs stable baselines to be accurate 2️⃣ Linear Regression – For Reliability Forecasting  → Models long-term degradation patterns  → Quantifies relationships between variables  → ⚠️ Assumes linear trends, even when they’re not 3️⃣ CUSUM – For Early Detection  → Catches subtle, sustained shifts early  → Triggers alerts before problems escalate  → ⚠️ Prone to false alarms without careful setup 4️⃣ Moving Averages – For Trend Clarity  → Smooths out noisy complaint data  → Reveals direction of performance over time  → ⚠️ Window size dramatically affects results 5️⃣ Poisson Models – For Rare Events  → Built for low-frequency, high-impact incidents  → Handles varying sample sizes with ease  → ⚠️ Less intuitive for teams to interpret The difference between teams that struggle and teams that prevent harm?   They don’t rely on habit. They match their method to their mission. But knowing the right tool isn’t enough. Smart teams build decision trees so they never have to guess. You know exactly which method fits your data, your risk, your goal. ✅ No wasted investigations ✅ No missed signals ✅ Just the right method, every time Your next safety analysis doesn’t have to be a statistical gamble. It can be predictable. Even effortless. ⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡⬡ MedTech regulatory challenges can be complex, but smart strategies, cutting-edge tools, and expert insights can make all the difference. I'm Tibor, passionate about leveraging AI to transform how regulatory processes are automated and managed. Let's connect and collaborate to streamline regulatory work for everyone! #automation #regulatoryaffairs #medicaldevices