Improving clinical workflows requires intentional organizational change management, built on a solid foundation of trust. Lencioni's model demonstrates how the results you seek are inextricably tied to trust, vulnerability in healthy conflict, and honesty. www.bitesizelearning.co.uk/resources/fi...
5 Dysfunctions of a Team: Summary, pyramid & manager tactics — BiteSize Learning
Fumbling around the internet looking for reliable sources for stats for TAM/SAM/SOM or pitch decks in health tech? This guide will get you to the right urls quickly. Includes a JSON file for your LLM. 50OFFNEWITEM for 50% off for the next two weeks. donicaconsulting.myshopify.com/products/mar...
Market Sizing Data: Directory of 300+ Free Web-based Data Sources
Instead of labeling a patient as non-compliant, we should consider what's getting in the way - transportation needs, child care, cost, etc. The same goes for clinicians. Instead of seeing non-compliance as a policy enforcement or education problem, consider what's getting in the way. #digitalhealth
As life-and-death as everyday clinical workflows are, things happen that make caring for patients exponentially harder. A fire or farmer severing network and VOIP communication. Disease outbreaks. Evacuations. Build for everyday processes, but plan for inevitable, emergent deviations. #digitalhealth
Consent is such a critical prerequisite to what we do, particularly as finely parsed sensitive data makes it more complex for patient care providers to track. Substance abuse, genetic data, etc. I'm glad to see it getting this attention. #digitalhealth www.healthcareitnews.com/news/new-int...
New interoperability resources for adopting computable consent
The potential for combining a patient's genetic data with symptomatic data to identify differential diagnoses with AI, particularly for rare diseases, is immense. We're just starting to see the possibilities in retrospective data sets. #PrecisionMedicine #HealthAI ai.nejm.org/doi/full/10....
LLM-Assisted Reanalysis of Unsolved Rare Disease Genomes Increases Diagnostic Yield
There is immense potential for earlier diagnosis of rare diseases if we use generative AI to find trends that point to a diagnosis, hand those findings to clinical research to vet, and then utilize the confirmed findings to provide targeted clinical decision support. #digitalhealth #healthai #medsky
Much of the industry is still treating AI governance as though it is a separate function, different from an org's usual governance process. But "AI-enabled" is now a baseline expectation, and AI requirements should be additive to a consolidated review process. #digitalhealth #healthai
In emergencies, checklists free up pilots' mental bandwidth to handle unique situations. Likewise, EHRs should streamline the routine, reducing cognitive load so that physicians can creatively focus on individual patient needs. To learn more about doing it well: atulgawande.com/book/the-che...
The Checklist Manifesto
Human-in-the-loop is not a lever to pull to abdicate responsibility for AI model outcomes, or to reduce vendor liability. The oversight person must have the time, knowledge, and autonomy to override, and the model must provide transparency and explainability. Otherwise, it's all for show. #HealthAI
Although a step in the right direction, having a link available in the EHR to launch an AI tool is not EHR integration, and certainly not clinical workflow integration. #digitalhealth #CDS #ClinicalAI www.mobihealthnews.com/news/openevi...
OpenEvidence, Boston Children's partner to study AI's impact on clinical care
Even an accurate and explainable clinical decision support alert can be counterproductive, unless it is provided to the right individual, at the right time, within the context for that person to take action. Without all three, it's just more noise for physicians. #digitalhealth #CDS
Under HIPAA, one method for de-identifying PHI is determining that "a person with appropriate knowledge of and experience with generally accepted statistical and scientific principles and methods" couldn't re-identify the patients. With ready access to AI, we must assume capabilities have changed.
It’s important to get agreement on success metrics prior to starting implementation, otherwise you run the risk of a mismatch on expectations. Look carefully at whether the selected metric(s) will be impacted at all by the changes, and whether the metric is aligned with the central objective.
Your organization's AI governance maturity is an extension of your existing governance framework. If governance has historically been complex or incomplete, you'll need to fix that foundation quickly so that you can adapt to the new pace and risks. Without a firm foundation, governance fails.
Startups commonly measure how well their models perform based on retroactive data. But we can’t afford to see that data as infallible. Differing processes or perceived meanings, bad judgement, and biases all live in the data before we start. How do you normalize your data for model training?
One reason physicians struggle with alert fatigue is that hospitals default to more inclusive alerting - just in case - rather than paring them to a level where they are all meaningful. I understand the fear of missing an alert, but more-is-better results in frustration and unread critical messages.
Healthcare is complex. To be successful in healthcare tech, you need to lean into the complexity.