Well now if you know that Epic is paying KLAS, do tell, and give evidence! Or is this another Oracle…
Curbside Consult with Dr. Jayne 12/14/20
I’ve written a couple of times recently about some healthcare misadventures. The first story involved a delay in care due to COVID-related office shutdowns and the subsequent fallout when a close friend of mine ended up with an abnormal cervical cancer screening test. A couple of weeks later, I mentioned a glitchy pathology results interface that failed to send out results in a timely fashion, leading to delays in notifying patients of their results and subsequent delays in follow up for abnormal findings. I had the physician’s perspective for that one, feeling my colleague’s angst as she had to notify her patients of their delayed results. I also felt her frustration at her healthcare IT team’s poor management of the situation and the potential impact on the patients whose results were delayed.
I had hoped to not encounter any further stories like these in 2020, because heaven knows we’ve had enough stress, challenges, and outright disasters this year. Unfortunately for my friend the patient, her story has continued, and not in a good way. Although we’ve been talking about it on a daily basis in text messages, emails, and screenshots of her MyChart account, she has given me permission to write about the episode as a cohesive story.
Although I’m glad that I’ve been able to provide support and help her navigate the system, the reality of this is that none of it should have happened, and there should have been better systems in place to protect the patient.
Following the initial abnormal test, she went through an office-based biopsy procedure and waited more than two weeks for her results. The delay which was attributed to a continuing backlog in the hospital’s pathology department following some COVID-related furloughs. Although her surgeon expected some degree of abnormality, the result revealed a significantly worse problem that required further surgery and additional biopsies.
The scheduling process was frustrating due to ongoing games of phone tag and the perceived need for real-time communication rather than using email or text to schedule. My friend voiced her frustration: “I don’t know why they can’t just text me the open times. I’ve given them permission and it’s not like they’re talking about anything medical. It’s about as confidential as scheduling to rotate my tires.” Throughout this, her sense of humor has stayed largely intact, but reading statements like that gave me a hollow feeling in my gut as I had a ringside seat to watch the brokenness of the US healthcare system play out.
Instead of sending pre-procedure instructions and patient education via MyChart, the office sent them via postal mail. They were delayed by issues with the postal service that were likely related to the run-up to the US general election. When they finally turned up two days before the procedure, she was glad to confirm what she had already read via Google, but was still baffled at why the health system can send her bill through MyChart but not other items: “I mean seriously, they send me a bill in MyChart two days after the visit before insurance has even paid, but they can’t send me the important stuff like patient ed?” I agreed with her assessment, but didn’t know how much this statement would come into play later.
Finally, procedure day rolled around. Although I was able to drive her, due to COVID, she had to go by herself with no support person. Many in healthcare overlook the impact this has on our patients. She was definitely a little off when she got back in the car, and apparently she’d had an atypical reaction to one of the medications and was having some lingering effects from that. Although I understood as a physician why the rules are the way they are, as a friend, it didn’t seem to be the best thing for the patient to let her leave the facility the way she did. I went through a similar solo experience at my recent colonoscopy and they at least called my driver to pick me up at the door and made sure I got into the vehicle in one piece.
Her surgeon told her that pathology was still backed up and to expect a two-week turnaround for the results. She describes those two weeks as “eternal” from the patient side and joked that they should have tracking on the pathology specimens like online shopping: “Imagine it! You could get an email that said we’ve received your order, then we’ve received your biopsy jars, and it could just go through the whole process. We’ve prepared your slides! You’re next under the microscope! We’re writing your results! We’ve sent your results to your doctor!” It would be funny if it wasn’t so true that patients have more understanding of the flow of goods through Amazon then they do with regards to critical tests affecting their health. Putting on my vendor / consulting hat, I’d love the opportunity to write those messages, if anyone is interested.
She also lamented the fact that during those two weeks, she received numerous messages in MyChart about “complete nonsense that has no bearing on whether I have cancer or not.” Apparently, there were two more billing messages, along with several broadcast messages regarding COVID screening and the availability of telehealth visits at the university medical center. She had a good point: “Why can’t they set this up with a different notification if it’s something important versus something generic? All the emails just say there’s a new message in the chart. It could still be HIPAA friendly – ‘you have a new message in your patient chart’ means it’s about you, and ‘you have a general message from Big Health System’ means your heart can stop racing.”
We joked about taking it further to modify the notification sound on your phone, to a gentle “whoosh” for regular messages, an alert siren for the big stuff, and a cash register noise for the billing messages. We also thought about the ability to snooze general messages for 30 days just like we can snooze annoying Facebook friends.
Almost two weeks to the day, she received a text from her doc: “Path back, margins clear, all good, will send in MyChart” which she forwarded to me, followed by a copy of her doctor’s note and the report. I saw something odd on my first reading of the report, and after re-reading it half a dozen times, I knew I had to call her. It wasn’t in fact all clear, but there was a mismatch between the summary portion of the report and the actual microscopic description of the slides. The translated version is “all the cancer is gone, but cancer is present.” She messaged her doc, who agreed to call the pathology lab to have the slides and report reviewed.
Cue more waiting time, as the lab was vague on how long it would take for a repeat reading. As a good friend, and without naming any names, I reached out to a pathologist who works in the same health system to ask about the process. She was shocked that the report went out, since there was supposed to be a human review of all reports to make sure nothing like that was released. She also told me that re-reads are supposed to be done the same day.
Three days had passed and finally it took another phone call from the surgeon to get things moving. According to my friend, “that lab is in shambles due to COVID cases and quarantines.” She also noted that errors have been on the rise across her system since they replaced the human transcriptionists with voice recognition software, which is why they instituted human review. Although I sympathized with the quarantine and technology issues, I was still shocked at what a mess was going on within the healthcare organizations that we rely on.
After the surgeon’s second phone call, an amended report showed up at 11 p.m. Fortunately, my friend’s doctor was still up working her inbox when it hit, and my friend was still up as well. I was up too, and she texted it over. We chatted the next morning, and although the amended report was reassuring, she still didn’t know if she should trust the lab “since after all, this is just my life, right? They can’t even proofread?”
It’s a good point, and I encouraged her to ask for the samples to be sent to an outside pathologist for a second read, which she did. The overread took another two weeks. Personally, I think the first lab should absorb the cost of the overread since it was their error that triggered it, but I’m sure she’ll be getting a bill. At least though she has peace of mind that the reading is accurate.
Although she has received a reprieve for the next several months, there is still more testing to come. Even if all is good, she will be looking at ongoing followup for a few years. I hope that by the time she goes through this again that the systems have recovered to the point where she doesn’t have to endure anything near what she went through last time.
Many of the factors influencing these outcomes can be laid at the feet of the relentless cost-cutting that is everywhere in healthcare. As a clinical informaticist, this story is going to haunt me for a long time, because some of the work we’ve been collectively involved in is designed to prevent incidents like this debacle. Although we push hard to try to make things more patient-centric, money always gets in the way, not to mention the impact of a global pandemic. There are lessons to be learned here:
- Patient self-scheduling can be an OK thing as long as there are guardrails around it, even for procedures. I’ve helped clients do it, and they’ve not only survived, but thrived.
- It is unclear why a practice would send paper patient instructions to a patient with an active MyChart account, especially during a time of elevated concern about the US Postal Service.
- COVID is going to be with us a long time, and we need to think out of the box on how to get patients the support they need when they’re engaging with healthcare institutions. Asking them to go it alone isn’t the answer, even if they appear young and healthy.
- Hospitals are penny-wise and pound-foolish, and that’s not likely to change.
- Amazon does a better job communicating about the status of a laundry detergent purchase than hospitals do regarding life and death matters.
- Voice recognition software may be cheaper than human transcriptionists, but cost isn’t everything.
- No matter what the technology is, human error still occurs, and multiple people missed the mismatch in the report including her surgeon. Not everyone has a friend who is a doctor to provide an additional line of defense.
- Communication is EVERYTHING.
No matter where you are in this industry, please think about this story and think about how it impacts you professionally and also personally. If this patient were your mother, your sister, your daughter, or anyone you care about, is this how you would want it to play out? Regardless of who we work for, our ultimate customer is the patient, and doing what’s right by them should always be in front of us.
How does your organization tackle some of these missed opportunities? Leave a message or email me.
Email Dr. Jayne.
Thank your friend for allowing you to share her story, and for you for writing up the long, way too long process. Sometimes in HIT we forget the patient lives and emotions are tied up in what we do.
What an unbelievably accurate story of MyChart and its problems. I have completed a chemo/radiation regimen in a facility with MyChart and I could laugh along and shake my head with your commentary. From the patient’s point of view, it is an unmitigated disaster. As I said, I laughed along with your commentary, but at the end you could just cry at how badly the technology is implemented.
To be fair, it’s not just MyChart, it’s any patient portal. It seems they all require so much customization from the organization itself, and at the end of the day you’re reliant on the individual providers putting in the appropriate data. (No different from EHRs I suppose…)
One of the design problems is that the “programmers” are looking at the functionality from their point of view which is “how do I get this particular function coded as fast as I can” rather than having an end user such as Jayne lay out the function from the point of view of the end user with the attitude of “how should this work to make it easy for me to understand and use”.
Until the end users get more of a say in the functionality the problem will never go away. It has been this way for many generations of EHR or whatever you wanted to call it back in the day.
Naw, laying this on at the feet of the organization. MyChart has a ton of dials, sure. But if you as the organization don’t take the time to understand what it if offers and meaningfully set it up it can end up like this story. I constantly ask the analysts I work with to think about things from a patient perspective and we normally end up tweaking the business’s asks for the better at the end of the day.
Well. Maybe, sometimes.
Look, I’m a programmer with 30+ years of experience, around 25 of those in healthcare. I have ALWAYS sought user input into systems design. This ranged from large systems to “simple reports”.
Do you know what my most common responses were? Puzzled looks. “I don’t know, you’re the computer guy.” Total silence and no comment. Or, the arm waving initial concept stuff, which doesn’t translate into individual feature implementations. “I’m super swamped right now, I wish I could help, maybe we could sit down together in, let’s see, October…”
Every once in a while you’d get someone who’d seriously engage with you. They’d take the time, buy in to the concept, learn what they had to learn, give helpful feedback. Those people were gold, both in value and in rarity.
What users are GREAT at, is telling you that you did it wrong. Long after the job was done, and after the project stage that would allow you make it right.
Mostly, the programmers are left to imagine the solution on their own. Some organizations employ systems architects, designers, and even plain managers who will do mock-ups, role-playing, workflow analysis. Not once have I ever had the opportunity with those people. They cost money and take a great deal of organizational willpower and discipline.
Steve Jobs was famous for envisioning a finished system. He’d then follow the implementation to make sure the design & implementation followed the plan. He’d keep after managers and programmers, keep them on track. He’d even kill products that didn’t live up to his expectations.
Steve Jobs was famous for being an exception, an outlier.
In the real world, most systems go through a laborious process, slowly evolving from something terrible into something better.
Agreed! In my opinion, workflow design (and of course system design) questions are often difficult conceptual problems, and they take a good deal of focused thought to come up with a sensible answer. Confounding the problem is the fact that a good design, once clearly laid out, is deceptively easy to follow and understand, leading to the mistaken idea that it was equally easy to come up with. Thus people don’t generally have an adequate appreciation for the task of design. Thus they sometimes aren’t prepared to devote the time and focus actually required.
As my high-school English teacher used to tell us: “Anyone can take a simple problem and make it complicated. It takes a genius to take a complicated problem and make it… anyone? anyone?”
I wanted to reply to Clustered, but there’s no Reply link.
Clustered makes an excellent point. Making it simple, is hard. There is a famous book on web design, called “Don’t Make Me Think!” The author is Steve Krug. Very worthwhile if you care about this stuff.
Here’s another barrier. The actual tools that most programmers use, are extremely detail-oriented. That tends to make their use slow and (frankly) tedious. It’s a huge problem trying to bring users into the design process when the developer tool ensures they will spend most of their time twiddling their thumbs.
That has led to a product segment of “Mock-Up Tools”. These produce a Wizard Of Oz type of result, where the final product looks as the user will see it. However there’s little or no actual functionality! But at least it focuses on the user interface, workflows, inputs and outputs. And normally, the Mock-Up has to be completely rebuilt in the actual dev tool, so there’s some path retracing that has to take place.
Again though, these tools cost money, and they take time and discipline to use. All of my employers could never get past the cost and time barriers. The thinking was (always), “Why spend on the Mock-Up, when we can just build the real thing? I mean, it’s a No-Brainer! Now quit fooling around and get to work!”
And this is the heart of the problem. User input into the development cycle doesn’t get much value or respect. No one wants to invest in that. For a very limited range of systems, yeah, sure. Consumer stuff like smartphone apps. But Enterprise HIS? ERP systems? Finance & Materials? Logistics? Scheduling? Manufacturing?
“We’ll maybe do something about that in Version 2. Right now we’ve got to focus and get this stuff out the door! Why are you still in my office?”
I think we also need to look at what is a vendor specific issue, what is a broad patient portal issue, and what is caused by organizational decision(s).
In this case, it sounds like an organization decision to not use appointment booking and communication but to enable billing.
An issue I see here is where is the incentive for the health system to improve? Yes, when I worked in HIT for 25 years we frequently reminded ourselves that it could be our loved ones being affected by what we do. But that never seemed to stick at some of the organizations we worked with. What other incentive is there? The biggest incentive is to keep the revenue coming in, which explains why the bills came through electronically just fine.
Some good old competition in this space would go a long way. The question is how to effectively do that without compromising essential services (such as ED services and indigent care) that can’t be improved through a competitive process. When I got quotes on an MRI a couple years ago, the local academic center quoted me $6,000. I got it at a retail site for $600. Service was wonderful.
I would disagree that this is a cost problem. For example, electronic notification is cheaper than postage. I looked at each of those bullet points and tlsee a poorly managed company. These problems aren’t solvable by the workers or the shareholders. It’s the people between those groups that are dropping the ball. They would be the one to pressure the lab on quality, to manage personnel effectively to support self scheduling, or to prioritize electronic over paper processes.
The truth is healthcare has exceptionally bad management. Part of that is cost in that healthcare companies aren’t willing to pay for good management. That isn’t cost cutting though, no matter how much professional organizations use cost cutting as the Boogeyman. No one looks at US healthcare and says “Gee, the problem is the costs are too low. We’ve got to stop cutting those costs!”
Healthcare organizations have some structural problems to work through. There is a worker caste system in healthcare. An exceptional manager who happens to be an MA cannot manage mid level providers or doctors. Also, healthcare organizations have only recently grown to require management; much of healthcare used to be delivered by the owner-operator model and a major portion of the workforce still has not matured out of that mindset.
This type of thing won’t change for at least another decade or two. The system as it exists is inefficient but can’t be changed. It will consolidate to avoid pressure until it can’t. Then competition and innovation will be able to start.
Some healthcare groups are terrified of the regulatory fines and attorneys circling around the facilities. Though the “Amazon does a better job communicating about the status of a laundry detergent purchase than hospitals do regarding life and death matters.” is a vivid thought, Amazon is not subject to intense and prolonged litigation for HIPAA violations. Just a thought when folks speak of reform, the Trial Lawyer lobbyists prevent any reform for a piece of the puzzle that costs us all 17% of every healthcare dollar spent.
There aren’t really trial lawyers circling hospitals to drum up lawsuits over HIPAA since patients can’t actually sue over it without a judge quickly throwing the case out.OCR pursues HIPAA violations and I think they are generally pretty fair to business. IMO hospitals use the fear of lawsuits as another talking point to excuse their incompetence.
Not to mention, hospitals already have a HIPAA-approved, secure way to communicate to patients via patient portals such as MyChart. They can give a patient her final results but can’t give her a status update about where her lab is in the process for fear of a HIPAA violation? I call baloney on that.
Exactly. I worked with an org recently that has their own application, completely over MyChart. Nothing that you see is Epic-designed and it’s missing lots of standard Epic features, but plenty of people still refer to it as MyChart. Drives the Epic team nuts, especially since it’s meant a ton of work on COVID features that could have been plugged in straight from Epic.
I feel empathy towards your friend for having endured this poor level of concern by the healthcare organization that is supposed to take care of her needs. It’s easy for healthcare organizations to blame the crisis of the day (in this case, COVID) for their poor performance. Where were the quality monitors and guardrails to catch unreported results within a reasonable timeframe? And incomplete and/or incorrect reports? I am very familiar with Epic configurations, so I know that the tools exist to provide this information internally to lab management from receipt of specimens through final reporting. No patient should have to rely on the repeated intervention of friends and/or surgeons to move the pathology process along (I experienced this myself many years ago, when a biopsy was not read for more than 2 weeks time, and it turned out to be positive for cancer; a physician friend had to intervene to get the final report released). I hope that a copy of your narrative finds its way to the chief operating officer and chief quality officer of the organization in question, so that they can see the effects of poor management on the individual patient, and make appropriate adjustments to operations.
“ our ultimate customer is the patient”! Follow the money. Hospital customers in order of importance:
1) Employers – the pay the lion share of the cost
2) Doctors – they bring in the business
3) Insurance companies – they decide how much to pay
4) Patient
This is spot on! Despite all the talk about upcoming consumerism in health care, the reason for such dismal patient experience is precisely this. Patients don’t control the purse string (even though it is their forking money). All they can do is fill up a toothless HCAHPS paper survey 3 months after their encounter with the health system.
And this is of course true not only for hospitals but for every single aspect of our medical care “system”.