Well that's a bad look as the Senators contemplate filling in the House gaps in the VA Bill
EPtalk by Dr. Jayne 7/8/21
It’s always good to hear about true interoperability in action. The Surescripts Clinical Direct Messaging platform has sent over 7 million COVID-19 immunization notifications from retail pharmacies to primary care providers. Now if only we could get health systems to share amongst themselves so that patients could have one cohesive record, that would be great.
I have multiple Epic charts in practices that are literally across the road from each other, but because they belong to competing health systems, they don’t recognize each other’s data. I know that Epic is capable of sharing, but the systems aren’t ready for that. Information blocking, anyone?
The World Health Organization issues its first global report on the use of AI in healthcare. Titled “Ethics and governance of artificial intelligence for health,” it includes six guiding principles for the regulation and governance of AI that are fairly straightforward and frankly are in line with what we should be doing in all facets of healthcare IT:
- Protect human autonomy.
- Promote human well-being and safety and the public interest.
- Ensure transparency, explainability, and intelligibility.
- Foster responsibility and accountability.
- Ensure inclusiveness and equity.
- Promote responsive, sustainable AI.
The report does note that we need to be cautious about overestimating the benefits that AI can provide, particularly if resources are diverted from core investments needed to achieve universal health coverage. I thought it was a nice way of saying, “watch out for shiny object syndrome.” When you’ve got people in the world who lack basic hygiene and sanitation, clean water, and immunizations, it’s sometimes difficult to think about spending millions of dollars on advances like AI.
During the last few weeks, I’ve seen multiple articles looking at the impact of the COVID-19 pandemic on various preventive screenings. One article looked specifically at screening test volumes through the National Breast and Cervical Cancer early detection program. In analyzing data from January to June 2020, the authors found that the pandemic reduced screening rates among low-income women covered by the program. This is not at all surprising to those of us who have been in primary care. When push comes to shove and women are under stresses, they tend to put themselves last because they’re busy caring for their family members. The pandemic added extra layers of stress, including economic burdens, distance learning, and greater care responsibilities for elderly relatives or those at high risk for complications due to COVID-19.
Several of my clients have asked me to assist them with campaigns to reach out to patients for preventive screenings. The more sophisticated clients can trigger scheduling of the services through text messages, but some still require patients to call in or access a patient portal to schedule.
Although they’re excited about the capabilities of their patient engagement platforms, I have to keep reminding them that getting the patients engaged and scheduled is only part of the battle. They need to be making operational changes to make it easy to actually have the tests performed. This means leveraging technology investments to streamline in-person registration processes and history updates. The facility where I had been getting my mammograms is one of my clients and my last experience was so unfortunate that I transferred care elsewhere.
What could they do to better serve their patients? First, leverage the EHR. Use the system’s capability to generate pre-populated patient information forms so patients merely have to update their history rather than filling out a bunch of redundant information, including name and date of birth on every page. Use the data already in the system regarding primary care physician, ordering physician, and date of last exam to make it clear that you already know a good chunk of what’s going on with the patient.
Second, streamline the “COVID hygiene theater” processes that are still going on in many medical facilities, including excessive distancing and unwarranted surface cleaning that slow patient flow or create unneeded levels of concern regarding infection control.
Third, figure out how to schedule so that you can run on time. Use the data from your systems to fully understand your throughput so people can have timely testing and get back to their other responsibilities. Getting a mammogram or a pap test shouldn’t be an all-day affair, but in many places, it is, which adds additional barriers for patients in hourly jobs or patients who might not have protected time off.
Props to Steve Edwards, president and CEO of CoxHealth in Springfield, MO. He tells those who are spreading vaccine misinformation to “shut up.” Even better is the thread where his mother, a 90-year old retired operating nurse, says “I have always told you not to tell people to shut up, but this it is okay.” Ready to rumble, indeed.
I recently heard the phrase “innovation through imitation” used and kind of chuckled at it, but the more I think about it, the more it applies to entirely too many initiatives. The most recent example I’ve seen is the recent announcement that Dollar General plans to jump into the healthcare fray with a push to expand health offerings across rural communities in the US. The press release summarizes the company’s plan to “establish itself as a health destination” by stocking “an increased assortment of cough and cold, dental, nutritional, medical, health aids and feminine hygiene products” in stores. To further this effort, they’ve hired a chief medical officer, Albert Wu, MD, formerly of McKinsey & Company.
I hope one of the first thing Dr. Wu does is to consider bringing the company’s press release writers into the world of inclusive language by using modern terminology such as “menstrual care products” to describe some of the offerings they plan to stock. News flash: transgender men and nonbinary people may menstruate, and the continued use of “hygiene” around menstrual products perpetuates myths that menstruation is somehow unclean. According to the press release, Dr. Wu went straight from his anesthesiology residency to being a consultant at McKinsey, so I’m betting his missed out on the subtleties that many of us learn to appreciate through decades in practice. I’m a little embarrassed on his behalf about the way it was worded, as well as about some of the things in his LinkedIn profile, but I wish him the best in his efforts.
What do you think would be the most helpful strategy for building greater healthcare infrastructure in rural communities? Leave a comment or email me.
Email Dr. Jayne.
Re: Information sharing.
Epic has a system called Care Everywhere that has as it’s specific mandate, sharing between HIS systems. Those systems include both Epic and non-Epic systems too. Care Everywhere is compatible with the Carequality HIE network.
My suspicion is that Dr. Jayne already knows this.
Why mention this? Well, I think it’s relevant that the vendor appears to have done their homework. The internal structures and mechanisms have been built. There’s a whole support structure to enable information sharing.
Therefore when 2 compatible Epic HIS systems aren’t sharing data, it’s entirely a customer side issue. Maybe they aren’t mature enough to share data (after all, I’d consider external data sharing to be an ‘advanced’ HIS function, and less of a priority than internal needs and priorities). Or maybe, someone at the customer has specifically decided they don’t want to share data.
After all, if setting up Care Everywhere is relatively easy and is fully vendor supported? One has to start to question what the hold-up is.
CareEverywhere is a decent intra-platform solution, although I have experienced zombieism, reflection, and distortion between the various instances of my Epic charts.
Had a pretty bad accident and ended up in the ED, the ED Chief was perturbed that I hadn’t given him my complete medical history and active meds, my blood type, and allergies. I finally had to list out my meds with dosage, my NKDA, and reiterate my blood type — with the followup of, “I just reconciled my med and problem list last week”, I can’t recall, (print) what he said as it was a bit colorful, but it is a problem with all data aggregators.
Zombieism: when data that should be closed/resolved/completed is resurrected as active
Reflection: Data from one system is sent to another system where it is then sent back to the original system
Distortion: When that “reflected” data comes back to the originating system unrecognizable from its original form.
From a technical perspective, a Care Everywhere implementation that revolves around the standard CCDA exchange with other Epic clients should take…maybe a dozen billable hours to do the technical build including some minimal data mappings? Toss in a 0..* amount of hours for getting appropriate signoffs from HIM & Compliance, maybe a few millennia if you want to connect to a backwater homegrown regional HIE…
Anyhow – any two Epic-using groups that don’t communicate well with each other are suffering at the hands of their own decision-makers.
Epic has another system too, called Share Everywhere. Care Everywhere is a classic Big Tech solution with deep integration. But what if there is a hiccup or hold-up of some sort? What if either the sender or receiver has an issue?
That’s where Share Everywhere comes in. Share Everywhere is a minimalist data sharing solution, designed only to be read by the receiving clinician and not a computer. The patient obtains a one-time code, then gives the receiving clinician that code and the Share Everywhere website. Then the patient’s own data will be shared with the receiving clinician.
The only things the receiving clinician needs are a web-connected computer, the Share Everywhere web link, and the access code.
Despite it’s minimalist design, the receiving clinician can add notes, and those notes will be sent back to the hosting Epic instance. Nice!
Note that I don’t have any experience with either Share Everywhere or Care Everywhere. Thus the comments here showing hands-on use, are much appreciated.
Glad to hear that Surescripts is able to send immunization records to the primary patient record. Given that the immunization domain is so easy it is surprising how many times it is corrupted. Essentially, you need the who (patient), the what (CVX code and name), and When (date to at least the year specificity) and you have a valid historical immunization record. Anything else is gravy.
Now, that having been said, there are plenty of solutions and implementations of solutions that can’t seem to get that right. Immunization “L Arm” given on “Tuesday”, for patient “JohnVPMD smith” <– completely useless and in many systems completely valid. The other one I like is "School Vaccines" — I would hate to be an immunization registry trying to make sense of that class of data.
About immunization distribution efficiency. I have been closely following the Japanese vaccination program, and their complaint that there aren't enough physicians to administer the vaccine. I remember doing Vaccine duty in the Air Force at Lackland, two or three medics administering the battery of vaccines for multiple flights (50ish airmen per flight) a day. If you really need an MD to be there for the immunizations then set up 10 medics, 2 for intake, 2 for the monitor area, and 6 to administer vaccines — with a doctor supervising. Easily process 1000 a day. The papers in Tokyo were excited that they were doing 15000 a day — (which is less than 1 tenth of a percent of the Tokyo population) with 100 docs and 1000 medics, you could easily push through 100k per day.
Of course, we need more people like Steve Edwards being direct about shutting down disinformation.
Thanks for your post…
My brother in law is a veterinarian. Really early in the pandemic the MA government asked him if he would be willing to give people shots if it came down to it. They didn’t end up needing it. When I got my vaccine, it was from the local fire department rotating through one of the mass sites.