Creative Health Labs

This is really interesting. Elon Musk has another company called OpenAI, an artificial intelligence company. OpenAI built a neural network that was designed to teach *itself* how to play a very difficult video game and then the AI played against the best players in the world on stage in front of millions of people.  What’s really interesting is that the AI used strategies that the humans had never considered and could not even initially understand.  The implication for the ER is really interesting. Optimizing ED workflow is really a “game” in the same sense - you have two teams (providers/nurses/techs vs patients), with limited resources and functions, and each side has specific goals (for patients it is to get out quickly and have their healthcare issue resolved, for hospitals/doctors it is to keep the patient alive and get them out as quickly as possible by using the fewest resources).  Imagine if somebody created a video game that mimicked the peopl...

This is pretty cool. Medicare/CMS posts a lot of their data in two main websites for public consumption: Data.Medicare.gov Data.CMS.gov I don't know why these two are separated, but it does seem like they host two separate datasets of healthcare data. One can connect Windows Excel to an OData feed from Medicare/CMS, but I can't figure out how to do it with Mac Excel '11 (I don't have a Windows computer). I prefer using Google Sheets anyways, so I wanted to find an easy way to connect Sheets to Medicare/CMS data. Here's the technique I have found, and please let me know if there is a better way: 1. First, access the table of data you want. For me it was this one: https://data.cms.gov/Public-Use-Files/Medicare-Provider-Utilization-and-Payment-Data-Phy/ee7f-sh97 2. Because these tables are usually massive, you will probably need to login and create a saved view of a subset of the data . Here's the subset of data I created, where I did a search for ...

I am happy to announce that my detailed post on how I built CMElog.org using Knack , Zapier, Dropbox, Mailchimp, Mandrill, and Google Drive has been published on the Knack blog: In this post, I dive into details of how to use Zapier to build automated services, like CMElog.org, in which medical providers' continuing medical education credits can be automatically extracted, identified, copied for back up, organized, and accessed.

There is a wonderful TED talk by Shawn Achor in which he explains that most people have the success and happiness equation backwards. Whereas most people believe achieving greater success will result in more happiness, he argues that being a happier person will make you much more successful. He goes on to say that happier and less stressed doctors are 19% faster and more accurate in diagnosing patients.

Medicare just recently published additional public data about what physicians received in fee-for-service payments from Medicare in 2014. Being an Emergency Medicine physician, I was naturally curious about this data and what information I could glean about our specialty. The first thing I did was to download the data from Medicare , and I extracted only those providers who are in Emergency Medicine. For EM providers, Medicare publishes for how many cases of levels 3, 4, 5 and critical care ("CC") each provider was reimbursed. These are represented by 99283, 99284, 99285, and 99291. Critical care, 99291, is a very interesting one because it is up to the doctor's judgement (more so than the other codes which have very strict criteria) as to whether a case qualifies for critical care. There are, of course, guidelines, but still, it's mostly a judgement call and the physician is expected to write a brief explanation as to why the case qualifies. Critical care, of co...

In my recent article in Emergency Medicine News, titled " The Most Solvable Cause of Medical Error ," I explain how a "Slack for Hospitals" would greatly reduce interruptions in the hospital environment, and thus (hopefully) reduce medical errors. In the article I explain the reason that such a high number of interruptions occur in the hospital workflow (because there is almost no asynchronous communication system), and how frequent interruptions have been directly linked to a high rate of medical errors. Add this fact to the recent news that medical errors now make up third leading cause of death in the U.S. Tech entrepreneurs who don't know or understand the real issues in healthcare often make the naive assumption that  diagnosis  is the major challenge in healthcare.

I am excited to announce that we have relaunched Jobmap.io - healthcare jobs on an interactive map with additional useful resources like median income for the zipcode and school rankings, essentially like a "Redfin for healthcare jobs." We will be launching some cool new services within jobmap soon, including 1-click employment contract review for physicians, APPs, and other healthcare providers.

Much has already been said about physicians' frustration at becoming data entry robots, and here's a great graphic to communicate some of the true costs of ICD-10: Click to Enlarge Image

The number and characteristics of people in California who have been certified eligible for Medi-Cal is changing dramatically over the last several years. Here's some data I extracted from the following dataset: Statewide Medi-Cal Certified Eligible Individuals, by Aid Code, 2013 - 2015 . Where M1 (the light blue line) = " Adult 19 to 64 at or below 138% FPL Citiz/Lawful" and 3N = " AFDC-MN-1931(b) Non CalWORKS" (more info here:  http://healthconsumer.org/cs0151931b.pdf). It would be interesting to dig deeper to understand the basis of this change (is this a true change, or just a change in definitions?).

As an Emergency Medicine physician in the year 2015, it is really astounding to think that we landed a man on the moon more than 40 years ago, yet I have patients walking in the door who had tests done at a facility across the street yesterday and I can't access those records. Thankfully some states have taken this situation into their own hands and developed systems for sharing medical records, such as the CRISP system in Maryland . Alas, most states have not yet built health information exchanges, and even those that have (like Maryland) don't necessarily have a patient portal. There are many different models for how medical records should be stored and shared in the future, and I'm not sure which model will win out in the end. What seems most compelling is for records to revolve around the patient (as opposed to the hospital), and two products out there that do this well (to my knowledge) are: PicnicHealth and Gliimpse . Both PicnicHealth and Glimpse will help retriev...

Much has been made of using Google Search data to predict health events, especially flu outbreaks. Here's how this flu season is proceeding compared to prior years: The blue line represents the influenza trend for 2015, and so far it is the second highest year of the last 5 years. Note that the highest year was the 2012-2013 season which actually started off quite slow.