I’ve started reading a book by William Baumol called The Cost Disease, which raises an interesting question. Why in 2014 can I buy a laptop computer that is smaller, more powerful, and most of all, much cheaper than one I could have bought just ten years ago, but healthcare costs have risen from ten years ago? Why are some industries able to become more efficient, and produce more of their goods or services, faster and cheaper, while other industries are stuck in a spiral or are continually raising costs with stagnant efficiency?
I won’t give a full, in-depth review of the book, but in short, the analysis lays out the premise that in some industries, such as with computers or automobiles, manufacturers are able to continually improve on both their manufacturing processes and the quality of the goods they are manufacturing. This enables these manufacturers to produce better goods at lower costs. These lower costs then enable them to pass some of these savings on to their customers, as well as to pay their employees more money. Continue reading
As more and more hospitals and healthcare organizations convert more and more of their paper medical records to electronic health records (EHRs), an interesting dynamic has begun to emerge, as well as an interesting challenge.
The dynamic is that while the conversion from paper to electronic records was promised to provide time and cost savings for healthcare, the adoption of EHR systems by physicians has led to a number of perhaps unforeseen consequences. One of the chief consequences, which could also be considered a chief complaint, is that physicians who document on their patients electronically make less eye contact with their patients and have lower patient satisfaction ratings, vs. physicians who document on paper. Continue reading
In his book Blink, Malcolm Gladwell writes about Dr. Brendan Reilly’s work at Cook County Hospital in Chicago from back in the late 1990s. At that time, the hospital was stretched thin, running low on resources and struggling to deal with roughly 250,000 patients coming through the Emergency Department every year. Patients routinely waited hours to be seen. One of the hospital’s key struggles was determining which patients coming into the Emergency Department with complaints of chest pain were actually having a heart attack and thus required expensive, resource-intensive care.
It’s an interesting case study if you get a chance to read it, but I’ll just give a brief summary here. Dr. Reilly used work that had been done from back in the 1970s by a cardiologist named Lee Goldman. Goldman took the data from hundreds of cases and ran it all through a computer program to identify what kinds of symptoms and clinical findings actually predicted a heart attack. Continue reading
The healthcare industry, and specifically the healthcare provider segment, is moving towards a rather interesting and potentially dangerous intersection in the near future. As electronic health records expand and allow for more digitized patient data to be analyzed by an ever-increasing array of analytical processing power, we’re going to see a huge growth in the amount of “information” that can be returned to healthcare providers. At the same time, we’re seeing the well-known-but-often-ignored issue of “alert fatigue” and EHR workflow frustration becoming a major problem for healthcare providers.
Just browsing through any healthcare industry websites or taking a quick look at the PR coming from healthcare software companies, you can easily see what the McKinsey Company outlines in a recent article, “The big data revolution in healthcare: Accelerating value and innovation”: big data is on its way. Continue reading
A book came out several years ago that has continued to resonate with me and that I still reference in conversations and discussions. The book, What Would Google Do?, by Jeff Jarvis, delves into how the Internet has evolved, and continues to evolve, and why some companies like Yahoo and AOL became has-beens while Google has flourished.
If you think about how Google operates, they focus on giving their users control, while other internet companies focus on trying to control users. The example that illustrates Google’s strategy most clearly is to just go to www.google.com and look at what you see, and then go to www.yahoo.com or www.aol.com and compare these two to Google.
Go ahead, check it out. Continue reading
Who has recently logged in to Facebook to see what your friends are up to? (Anyone still have a MySpace account?) Anyone purchased something online at Amazon this holiday season? When’s the last time you searched for something online and didn’t “Google” it?
It’s interesting that there are a select group of companies who have come to dominate our online, internet lives, and as a recent December issue of The Economist (1) points out in “Survival of the Biggest,” this comes with some positives and negatives. Facebook owns the world of social media, aside from the Twittersphere, and is looking to capitalize on all that personal information and preferences it knows about us. Amazon is dominant in online retail, with a third of online buyers starting their product search on Amazon, and two-thirds of all e-books downloaded coming through Amazon as well. Apple dominates online music sales and the growing tablet market. Google has long been the far and away leader in search and online advertising, though it is now under scrutiny having been accused of favoring search results to promote its own products and services. Continue reading
In the August 1st issue of the Journal of the American Medical Association, Robert E. Hirschtick, MD, wrote an article entitled “John Lennon’s Elbow.” In the article, Dr. Hirschtick writes about how the quality of progress notes created within EMRs during patients’ hospital stays has consistently declined as these notes have become longer and longer from carry-forward and copy/paste usage. Progress notes have lost most of their value in updating other providers about the current status of the patient, and no one seems overly concerned about this. Continue reading
A couple of years ago, I read a news article about an unusual kind of chess competition. This competition allowed teams of up to two competitors to use any chess computer software they chose in order to play against other teams of two with their own chess software. This may sound like an unusual competition, and it was, but the goal was interesting: to see which team of humans could best utilize technology to their advantage.
Most teams bought off-the-shelf software and just plugged their opponents’ moves into their chess program and then made the moves that their own program churned out. The winning team, however, actually developed their own software, leveraging machine learning and feeding their program the moves and results of thousands of games in order to “teach” it different strategies and moves. The really interesting part was that in some crucial situations this team actually made moves that went against their program’s suggested moves. They let the machine guide their overall strategy but were not afraid to make their own moves when they felt it was to their advantage. In the end, they walked away champions, with man-plus-machine being superior to just machine alone. Continue reading
Now that it looks like ICD-10 implementation will be postponed only one year, the already hot-topic of computer-assisted coding will get even more attention. Hospitals obviously need to ensure they have a strategy in place to deal with ICD-10 and the highly anticipated, but often misunderstood, ramifications of ensuring appropriate reimbursement under the new coding standard. Most hospitals I’ve come across have an “ICD-10 Committee” working overtime to determine what the right strategies, solutions and processes are for their organization.
With all of this focus on how to add intelligence and automation to the coding workflow, and mainly the output of that workflow, there has been a lack of attention given to inputs that feed the coding workflow: physician clinical documentation. ICD-10 at its root is a documentation problem. The best computer-assisted coding solution in the world is still dependent on the content of the documentation that it analyzes for the quality of codes that it produces. If key information is missing, you’ll never achieve appropriate reimbursement or accurate profiling and reporting. Continue reading
Today, 3M HIS announced it has acquired CodeRyte. As division scientist, this is exciting news from a technology perspective. Here’s my take on what it means for NLP:
The distinctive feature of CodeRyte’s technology is its strong statistical NLP capability. As I discussed in the 3M white paper, Auto-coding and Natural Language Processing, statistical machine learning systems offer the possibility of significant accuracy improvements in data-rich environments, compared to traditional rules-based approaches. A good example is in the outpatient coding environment, where data volumes are large, and CodeRyte has a history of performing very well compared to other systems.
There are multiple ways to boost the accuracy of statistical NLP, and the combination of 3M and CodeRyte will allow us to pursue several paths to the direct benefit of end users. Continue reading