Health Risks and Impacts

🎉 Pleased to share our paper published in Nature Portfolio digital medicine. 🥳 We’ve developed a comprehensive framework called CREOLA (short for Clinical Review Of Large Language Models (LLMs) and AI). This framework is pioneered at TORTUS, taking a safety-first, science approach to LLMs in healthcare. 🔹 Key Components of the CREOLA Framework -Error Taxonomy -Clinical Safety Assessment -Iterative Experimental Structure 🔹 Error Taxonomy Hallucinations: instances of text in clinical documents unsupported by the transcript of the clinical encounter Omissions: Clinically important text in the encounter that was not included in the clinical documentation 🔹 Clinical Safety Assessment: Our innovation incorporates accepted clinical hazard identification principles (based on NHS DCB0129 standards) to evaluate the potential harm of errors: We categorise errors as either ‘major’ or ‘minor’, where major errors can have downstream impact on the diagnosis or the management of the patient if left uncorrected.  This is further assessed as a risk matrix comprising of: Risk severity (1 (minor) to 5 (catastrophic)) compared with Likelihood assessment (very low to very high) 🔹 Iterative Experimental Structure We share a methodical approach to compare different prompts, models, and workflows. Label errors, consolidate review, evaluate clinical safety (and then make further adjustments and re-evaluate if necessary). ----------Method-------------- To demonstrate how to apply CREOLA to any LLM / AVT, we used GPT-4 (early 2024) as a case study here. 🔹 We conduct one of the largest manual evaluations of LLM-generated clinical notes to date, analyzing 49,590 transcript sentences and 12,999 clinical note sentences across 18 experimental configurations. 🔹 Transcripts-clinical note pairs are broken down to a sentence level and annotated for errors by clinicians. ----------Results-------------- 🔹 Of 12,999 sentences in 450 clinical notes, 191 sentences had hallucinations (1.47%), of which 84 sentences (44%) were major. Of the 49,590 sentences from our consultation transcripts, 1712 sentences were omitted (3.45%), of which 286 (16.7%) of which were classified as major and 1426 (83.3%) as minor. 🔹 Hallucination types Fabrication (43%) - completely invented information Negation (30%) - contradicting clinical facts Contextual (17%) - mixing unrelated topics Causality (10%) - speculating on causes without evidence 🔹 Hallucinations, while less common than omissions, carry significantly more clinical risk. Negation hallucinations were the most concerning 🔹 we CAN reduce or even abolish hallucinations and omissions by making prompt or model changes. In one experiment with GPT4 - We reduced incidence of major hallucinations by 75%, major omissions by 58%, and minor omissions by 35% through prompt iteration Links in comments Ellie Asgari Nina Montaña Brown Magda Dubois Saleh Khalil Jasmine Balloch Dr Dom Pimenta M.D.

Something is shifting in healthcare—but most people don’t have a name for it yet. It’s not just prior auth headaches, claim denials, or doctors drowning in paperwork. Those have always existed. What’s new is how quietly and systematically patients are being sorted, deprioritized, or excluded—often before they even seek care. This is Actuarial Medicine. Healthcare decisions are no longer just about clinical need. They’re being reshaped—not through outright denials, but through system design, automation, and interface constraints that subtly dictate the flow of care at scale. 🧮 Your medical record doesn’t just track your health—it generates a risk score that determines which treatments, referrals, or authorizations are frictionless and which face silent resistance. 🔁 Claims don’t just get denied—they get algorithmically stalled, redirected, or deprioritized before a human ever reviews them. 👨💻 Physicians aren’t just encouraged to consider cost—their tools nudge them toward pre-approved, lower-cost pathways, making other options harder to access. The Impact ❌ A test that was routine last year is now “not medically necessary”—but the denial appears as a pre-checked system recommendation, not an explicit rejection. ⏳ A referral that used to take weeks now takes months—not because of medical necessity, but because of administrative triage embedded in scheduling algorithms. 💊 A patient’s medication is switched to a cheaper alternative—not through direct denial, but because the better option is buried behind extra documentation and system hurdles. For patients with chronic, rare, or complex conditions, these aren’t just bureaucratic slowdowns—they are engineered barriers to care. And when those patients stop seeking treatment? The system records it as cost savings, not harm. This isn’t just about efficiency. It’s a fundamental shift in how care is controlled. And the longer it stays hidden, the harder it is to challenge. So let’s call it what it is: Actuarial Medicine. Are you seeing this in your work or care journey? How do we push back? Let’s talk. 👇 #ActuarialMedicine #Healthcare #AIinMedicine #HealthPolicy #MedicalEthics #HealthTech #DigitalHealth #PatientAccess

The Embedded Bias series by STAT sheds light on hidden biases — especially #racial and #gender biases — woven into the technologies and algorithms within our healthcare system. STAT, which is a media company that focuses on health, medicine and life sciences, dives deep into how these #biases affect patient care and outcomes. These critical insights show that bias in technology often reinforces the very health disparities we’re working to eliminate. The first part of this series exposes how racial biases in diagnostic tools can lead to inaccurate assessments for Black patients. The second reveals how gender biases in clinical trials have left women’s health concerns under-researched and under-treated. In later parts of the series, we see how bias in AI-driven healthcare solutions risks worsening disparities if not carefully checked. This investigation is a powerful reminder that transformative technology can still reflect and exacerbate existing societal inequities. If we’re not intentional about rooting out these biases, we risk further marginalizing communities already struggling to access quality care. https://bit.ly/3ZW14uD

Understanding Personal Noise Exposure in the Workplace Managing personal noise exposure is paramount to safeguarding employee health and well-being. Let's delve into key aspects surrounding personal noise sampling and its implications. Personal Noise Sampling: Personal noise sampling involves the measurement of an individual's exposure to noise during their working hours. This method provides a detailed understanding of the specific noise levels encountered by a worker, enabling tailored interventions to mitigate potential health risks. Sampling Process: Sampling is typically conducted using personal noise dosimeters or sound level meters attached to a worker's clothing. These devices continuously measure and record noise levels, offering a comprehensive overview of the individual's exposure throughout the day. Health Effects of Noise Overexposure: Excessive noise exposure can lead to various health issues, including hearing impairment, tinnitus, and increased stress levels. Prolonged exposure may contribute to more severe conditions such as cardiovascular problems, sleep disturbances, and decreased productivity. Hierarchy of Control for Mitigating Noise Exposure: Elimination: Identify and eliminate sources of noise where possible. This may involve modifying processes or equipment to remove or reduce noise generation. Substitution: Explore alternatives with lower noise levels. Consider replacing loud machinery or equipment with quieter options to mitigate overall workplace noise. Engineering Controls: Implement engineering solutions to attenuate noise at its source. This may include installing noise barriers, enclosures, or vibration isolators to reduce the transmission of noise. Administrative Controls: Introduce administrative measures to manage noise exposure. This can involve scheduling breaks in quieter areas, rotating workers to minimize prolonged exposure, and providing education on the risks of noise. Personal Protective Equipment (PPE): When other controls are not feasible, provide and ensure the proper use of personal protective equipment, such as earplugs or earmuffs, to minimize the impact of noise on individual workers. Adhering to this structured hierarchy of control measures empowers organisations to systematically address and reduce personal noise exposure in the workplace, promoting a safer and healthier environment for all employees. In conclusion, understanding and effectively managing personal noise exposure is pivotal for fostering a safe and healthy work environment. Compliance with legal limits, coupled with a strategic hierarchy of control, ensures a comprehensive approach to mitigating the adverse effects of noise in the workplace.

15% of all diagnoses are wrong, delayed, or missed and it’s costing healthcare 17.5% of its budget. According to the OECD, financial burdens from misdiagnosis are estimated to be 1,8% of the country's GDP. Not only do medical errors undermine trust in the healthcare system, it also leads to increased resources used for unnecessary tests, treatments, and hospital readmissions. Reducing errors would not only reduce costs, but also patient safety, improving patient treatment, treatment success, patients’ quality of life. What if AI in the future can be used to reduce diagnostic errors? Here are 12 ways how AI could potentially support reducing diagnostic errors. Information and Insights: Using AI to support comprehensive, accurate, and timely patient data collection, integration, and effective communication. 1) Multimodal Data Integration Combining diverse patient data types —such as medical imaging, lab results, vital signs, clinical notes, and genetic information, into a unified, comprehensive view. 2) Real-Time Patient Monitoring Continuous collection and analysis of patient physiological data through wearables and bedside monitors to detect early signs of deterioration or abnormal patterns. 3) Risk Stratification AI to analyze historical and real-time patient data to predict the likelihood of specific diseases or adverse outcomes. Diagnostic Imaging Accuracy 4) Error Detection AI algorithms analyze imaging data to detect abnormalities with higher accuracy and fewer false positives/negatives than human readers alone. 5) Lesion Detection AI systems highlight suspicious lesions or nodules that may be overlooked by radiologists due to fatigue or cognitive biases. 6) Imaging Reports Standardizing imaging reports to generate consistent and clear reports, reducing miscommunication. Clinical Decision Support 7) Pattern Recognition AI analyzes patient data to recognize disease patterns and suggest possible (differential) diagnoses ranked by likelihood. 8) Reducing Bias AI offers objective analysis that counters human cognitive biases, which can skew clinical judgment. 9) Real-Time Alerts AI systems can notify clinicians about critical findings, potential drug interactions, or overlooked symptoms during patient encounters. Error Detection 10) Data Quality and Consistency Checks AI tools can continuously monitor healthcare data for duplicates, missing values, or conflicting information. 11) Symptom-Disease Pair Analysis AI links symptoms from earlier visits with later diagnoses to detect possible delayed or missed diagnoses. 12) Real-Time Error Detection AI systems can analyze clinical notes as they are being written to flag potential errors or inconsistencies immediately. What potential do you see AI having in reducing diagnostic errors and improving patient treatments?

Most of us are back in the office, but are we really considering the impact of fluctuating workplace occupancy on indoor air quality (IAQ)? It's more than just a matter of comfort; it's about employee health and productivity. Studies have shown a direct correlation between poor IAQ and reduced cognitive function, increased sick leave, and lower overall well-being. We're talking about real impacts on your team's performance and health. So, what can we do? 1. Implement occupancy-based ventilation systems: These systems adjust airflow based on real-time occupancy, ensuring adequate ventilation regardless of the number of people present. 2. Educate employees: Make them aware of the impact of IAQ and encourage them to report any concerns. 3. Regularly maintain HVAC systems: Ensure filters are clean and systems are functioning optimally. 4. Consider biophilic design: Incorporate plants and natural elements to improve air quality and create a healthier environment. 5. Invest in real-time IAQ monitoring: Devices like the uHoo air quality monitor can provide continuous data on CO2, particulate matter, VOCs, and other crucial metrics, allowing for proactive adjustments. We need to move beyond simply filling office spaces and start designing for the health and well-being of our employees. It's not just a nice-to-have; it's a fundamental aspect of a productive and sustainable workplace. What strategies are you implementing to ensure optimal IAQ in your workplace? Share your insights and experiences in the comments below. #IAQ #WorkplaceWellness #Occupancy #HVAC #EmployeeHealth #Sustainability #FacilitiesManagement

Mastering System (HIS / EMR) Validation in Hospitals: Why Uncommon and Later Processes Deserve Equal Attention As healthcare administrators, we often dive into system validation projects with intense focus on early and common processes such as admissions, patient registration, and routine ordering workflows. These processes are critical to initiating patient care, which is why they receive such extensive attention during testing. However, the challenge is twofold: as the validation process progresses, less attention is given to both the later phases of workflows (like discharges, follow-ups, and reconciliation) and the uncommon or subspecialty processes (such as specific oncology workflows, complex surgical preparations, or specialized lab test ordering). This lack of focus can lead to overlooked gaps, impacting patient care and creating inefficiencies in areas where precision is just as critical. Examples of Overlooked Areas: Discharge Processes: Proper discharge instructions and medication reconciliation might not be tested as thoroughly as patient admission workflows. Subspecialty Workflows: Processes for rare diagnostics or complex treatments in fields like oncology or interventional radiology often fall through the cracks. Follow-Up Coordination: System functionalities for post-discharge follow-ups or appointment scheduling for subspecialists may be deprioritized. Data Transfers: Transitions between departments, like ICU to general wards or lab results from specialized tests, are sometimes under-tested. How Can We Address This? Equal Focus Across Phases: Ensure that both early and late-stage processes, as well as the common and uncommon workflows, are mapped and validated thoroughly. Stakeholder Representation: Include team members from subspecialties and later-phase process areas in validation discussions. Comprehensive Validation Planning: Allocate sufficient time and resources for uncommon scenarios and later phases to avoid last-minute fatigue or neglect. Systematic Reviews: Regularly revisit less-common processes to ensure they are functioning optimally even after go-live. The lesson? It’s not just about mastering the trunk 🌳 of healthcare workflows. Success lies in ensuring every single branch 🌿—no matter how small, specialized, or late in the process—is validated through process mapping 🗺️ and detailed workflow diagrams 📊. These tools help visualize the current state, identify gaps, and ensure no workflow is overlooked, enabling a comprehensive and efficient validation process. For a deeper dive into these challenges and strategies to overcome them, I recommend this excellent article: "Methods and Lessons Learned from a Current State Workflow Assessment following Transition to a New Electronic Health Record System" https://lnkd.in/dxDtUC9E #HealthcareAdministration #SystemValidation #PatientCare #EHRImplementation #ClinicalProcesses #HospitalManagement #WorkflowOptimization

Case Study: Health Hazards of Night Shifts – The Impact of Consistent Night Shifts on Worker Health Working the night shift, especially on a regular basis, can lead to significant health challenges. In one real-world example, consider an employee who worked 20 consecutive night shifts. While the individual was dedicated and focused, the demanding schedule began to take a toll on their physical and mental well-being. Here’s what they experienced: Health Impacts Observed: Sleep Disorders: The worker struggled to maintain a consistent sleep pattern. Sleeping during the day is challenging, and the disruption to their circadian rhythm led to insomnia, fatigue, and restlessness. Cognitive Decline: The lack of restorative sleep impacted the employee's concentration and decision-making skills, creating potential safety risks both at work and during their commute home. Digestive Issues: Irregular eating patterns, a common side effect of night shifts, affected the worker’s digestive health. Many reported stomach issues, irregular appetite, and fluctuations in weight. Increased Stress: The imbalance in work-life harmony was evident. Family life, social interaction, and personal relaxation time were compromised, resulting in higher levels of stress and irritability. Long-term Health Risks: Studies have linked chronic night shift work to long-term risks such as cardiovascular diseases, type 2 diabetes, and even an elevated risk of certain cancers. Prolonged exposure to these health risks can have severe impacts on quality of life. Lessons Learned: Organizations should recognize the potential health hazards associated with night shifts. Implementing strategies to support workers, like rotating shifts, providing health monitoring, and offering wellness resources, can help reduce the adverse effects. Individuals should also prioritize sleep hygiene, balanced nutrition, and regular health check-ups if they frequently work night shifts. This case highlights the need for both employer support and personal health awareness for those who work unconventional hours. Balancing productivity with health is essential for long-term well-being. 🌙💼 #NightShift #WorkplaceHealth #EmployeeWellbeing #SleepHealth #LinkedIn

In the world of law enforcement, where duty calls at all hours, the allure of overtime pay can be a significant motivator. But like many things in life, this financial benefit comes with its own set of challenges. Here's why #PoliceOvertime is more than just extra cash: The Financial Pull: Police officers often face the temptation of #ExtraIncome through overtime. With departments sometimes offering time-and-a-half, the opportunity to boost one's salary is tempting. For instance, in Austin, officers earned nearly $50 million in overtime in a year, showcasing how departments struggle with staffing but also how lucrative the extra hours can be for officers. The Hidden Costs: Stress: The job is inherently stressful, but extended hours exacerbate this, potentially leading to chronic stress or PTSD. Research indicates that long hours can impair decision-making, putting both officers and the community at risk. Health: Fatigue from extended work hours can have dire health implications. Studies, like those from the National Institute of Justice, link sustained fatigue to chronic fatigue syndrome and other health issues, impacting an officer's ability to perform their duties effectively. Family Time: The sacrifice of family time is substantial. Officers often miss significant life events, leading to strained relationships. The work-life balance in policing is notoriously challenging, as highlighted by insights from law enforcement experts and officers themselves. Balancing the Badge: While the financial incentive of overtime is undeniable, departments and officers must consider: Health Initiatives: Encouraging physical activity, mental health support, and adequate rest periods. Policy Adjustments: Some departments are implementing limits on overtime or exploring schedule changes like 10-hour shifts to combat fatigue and improve quality of life. Community and Personal Life: Focusing on community policing and personal well-being can help officers see beyond the financials to the broader impact of their work. #PoliceLife #LawEnforcement #OfficerHealth #WorkLifeBalance #MentalHealthMatters #PublicSafety #FamilyFirst #OvertimeDilemma #CopsOfInstagram Call to Action:If you're in law enforcement or know someone who is, let's discuss how we can better support our officers in managing the demands of overtime. Share your thoughts or experiences in the comments below. Remember, while the badge is a symbol of service, the health and well-being of those who wear it should never be compromised. Remember, every extra hour worked is not just time on the clock but a piece of life spent away from the ones they protect and serve. Let's advocate for a balanced approach to policing that honors both duty and personal well-being.

🩺 When advocacy meets lived experience This week, I found myself in hospital for a check-up. As the clinician clipped on the pulse oximeter to check my oxygen levels, I sat there with an uncomfortable awareness: 👉🏾 Pulse oximeters are known to be less accurate on darker skin tones. This isn’t new information. Harvard research has shown how these devices can misread oxygen saturation for Black and Brown patients, often overestimating levels. That means potential misdiagnoses, delayed treatment, and unequal care. Read more here : https://lnkd.in/eTsM_weq Even though I train NHS staff on this very issue, being on the receiving end was different. I had to advocate for myself—asking whether the equipment was calibrated appropriately, whether alternative checks could be made, and whether the results truly reflected my health. ⚖️ But here’s the dilemma: How do you advocate confidently for your safety without worrying it might affect the care you receive? This is what medical racism looks like. It’s not just about outdated language or offensive behaviours—it’s about systems, tools, and practices that still fail to protect Black lives. If you want to learn more about this space, I highly recommend following Joel Bervell (Instagram/ TikTok)—he consistently sheds light on the hidden biases in medical research and practice. Medical racism isn’t abstract—it’s lived. And sometimes, it’s sitting right there on your fingertip. #MedicalRacism #HealthEquity #RepresentationMatters #JoelBervell Photo: Getty Images