Going to ask again about HealWell - they are on an acquisition tear and seem to be very AI-focused. Has…
Curbside Consult with Dr. Jayne 2/18/19
I’m still recovering from HIMSS. What started as a slightly runny nose on Thursday has turned into a full-blown head cold with plenty of sneezing. At least it’s not influenza, however, so I’m grateful.
Any time you get that many people together shaking hands in flu season, it’s always a risk, but hopefully vaccination and lots of hand sanitizer has worked to keep the bad pathogens away. Unfortunately I’m spending the weekend in the clinic, making up for time out of office, and there isn’t much room for rest and recovery.
I mentioned the Arcadia-sponsored lunch I attended that featured John Halamka. I have been thinking about some of the things I learned during that presentation.
He talked about several countries and some of the challenges they are facing with healthcare. There was a lot to process. First, he talked about his experiences with China’s healthcare system and its approach to primary care, or lack thereof. The concept of coordinated care is just starting to take hold as China works to develop its primary care workforce. Until then, patients might self-select to a subspecialist physician based on what they think might be happening to cause their symptoms. He gave the example of someone waking with a headache and choosing to queue to see an academic neurosurgeon rather than seeing a primary care physician first. He is involved in a pilot project looking at care coordination that has one million patients in the cohort. That’s quite a bit larger than the pilot projects we’re used to in the US.
Next, he talked about healthcare in India, where a workforce of 600,000 physicians care for approximately 1.4 billion people. (Google shows the number at a bit closer to 800,000, but either way it’s a relatively small number for the care that needs to be delivered). Halamka has traveled to India multiple times and shared photos of his experience looking at x-rays that patients are carrying around with them as part of their personal medical record. There are significant healthcare disparities by region and the informatics community is starting to look at using clinical decision support to address issues where certain diseases such as tuberculosis might be more common in certain regions of the country.
He shared a story about a post-partum patient with abdominal pain and polled the audience (there were a handful of doctors in the crowd) about their thoughts. Nearly everyone thought about post-partum infection or complications like a retained placenta, but the real answer was abdominal tuberculosis, which isn’t even on the radar for most of us in the US, but is apparently fairly common in some parts of India. It really makes one think about how different healthcare experiences can be around the world.
The third country he talked about was Norway, where patients are expected to contribute their personal health information to the overall dataset of public health records. There’s not an “opt out” per se, which results in much more information being able for research and for investigation of health practices. I did some additional digging on this and found that although the Norwegian national Summary Care Record is in place for all citizens, the tool isn’t used as routinely in practice as it might be. Halamka also touched on the drinking habits in Scotland, which are among the largest per capita across the globe.
He also discussed interoperability in Australia, which is chiefly by PDF. This means there is no drug-drug or drug-interaction checking, no clinical decision support, and no common medication lists. There are, however, probably a great deal of duplicate data entry and wasted resources. He touched on Japan with its aging population and low birth date, where fewer clinicians are tasked with delivering care to a growing population, often with significant quality variation. Apparently in Japan it’s also illegal to host electronic health records on cloud-based services, which is going to need to change for optimal use of digital health resources.
Halamka paused from his round the world tour to address the role of artificial intelligence in healthcare, particularly the need to make sure that training sets used to build AI technologies aren’t full of biased information based on historical challenges such as delivering care to patients whose primary language is something other than English. We also need to take care with the use case for AI in healthcare — he used the lack of progress with Watson as an example. He doubted that the big EHR companies would really be doing much innovation, going on to say that the real innovation will be in a cloud of apps around those legacy systems and that the “24 year olds in the garage are the real innovators.”
He noted some of the challenges with interoperability in the US, namely with data sharing for adolescent patients, which must be controlled by policies and age ranges to meet the changing needs of children who are transitioning through adolescence and into the adult world. He mused on Apple’s partnership with Aetna, where patients will be given Apple watches as tools for health monitoring. It will be challenging for providers to figure out how best to use patient-generated data and for systems to figure out what data is important enough to act on.
He used his own example of a rapid heart beat as a test case. When it happens, it usually goes away in a minute or two. How will the system know not to act on it vs. to act on someone who is having a new onset of rapid heart rate or a more dangerous type? It’s one thing to have data, but another thing entirely to transform that data into information, knowledge, and wisdom.
There were good questions from the audience and Halamka went on to explain how he sees his role as “care traffic controller” for three generations of people in his family. I hadn’t heard that phrase before and I have to say I like it and will be stealing it for future conversations. Overall it was a solid session and I am even more curious about how other countries are handling various healthcare delivery crises and whether we’re really as bad off as we think we are. I’m going to have the opportunity to practice medicine with people from around the world this summer and I hope it will be a great learning experience.
What’s the best session you saw at HIMSS? Leave a comment or email me.
Email Dr. Jayne.
It would be great to know about healthcare costs and outcomes in China, India and Norway, to learn how the various care models impact both. As we continue to innovate and inroduce technology, there needs to be accountability. .
Re the Kaiser Permanente medical school: I think it makes a lot of sense for them to train physicians in how Permanente medicine works. (The Permanente Medical Groups are the for-profit physician partnerships that pair with the Kaiser Health Plans in various regions.) The KP group practice model is a different beast. They also have a mature and well-supported Epic implementation that hits most of the notes you mentioned from the Judy Faulkner interview. Epic was first installed in KP’s Northwest Region (Portland, and southern Washington State) many years ago so KP had a chance to learn the right way to approach their large scale Epic implementation. They did have physicians training physicians and it was hard to find the right people to do that work. A number of older physicians retired rather than learn the new way of doing things. They do standardize order sets on a facility and regional basis just like they control their formulary and the kinds of equipment and surgical supplies and implants they buy. They are able to do high volume medicine with good quality because of the ‘systemness’ of how they approach things and the fact that most of their physicians don’t fight it. As they grow their own physicians they reinforce that process. Dr. Jayne, you’d find it an interesting place to work if you had a chance to consult with them. Keep up the good work.