Should Hospitals Charge for MyChart Messages?

What a virtual billing dilemma can tell us about the state of healthcare delivery

It was reported that the Cleveland Clinic is charging for MyChart messages. UCSF has adopted a similar policy. Not everybody is happy. But is it reasonable to charge for questions?

For those not in medicine, MyChart is the patient portal for the the electronic health record system, Epic. MyChart messages are the way patients communicate with their clinics and providers.

I think it's safe to say that the general sentiment around billing for clinic messages is no. Unrestrained access to your doctor has always been an assumed patient privilege. And this is just another way to access your doctor. In fact, health systems have even promoted the MyChart feature as a value-add.

This question of fee-for-message care is interesting because it expose three things about healthcare and tech.

1 | We never know how an end-user really use a tool

As we think about this it’s important to start with a universal truth about technology: Whenever a tool is released we never know what the end user is going to do with it. This truth has played itself out from mobile apps to social media platforms.

In this case, health systems never really anticipated the way MyChart messaging would be used by patients.

So when given the unrestrained opportunity to touch base with their clinics, some patients definitely seize the day. In pediatrics, this can take the shape of 3,000 word messages — sometimes several times a day. These may be follow-up concerns from the day’s visit, or questions. Other times it represents a new, acute issue. As a pediatric gastroenterologist, my inbox is like Instagram for dirty diapers. Which is fine.

So with even a small percentage of a 5,000 person patient population send regular chunky messages, you’ve got yourself a serious inbox. But we never saw this coming when we turned on the feature.

Bob Wachter from UCSF had this to say on Twitter after the pandemic:

“We’re seeing huge uptick in in-box messages for MDs during Covid – now seems like biggest driver of MD burnout. The fundamental problem: we turned on 24/7/365 access for patients (who of course like it) with no operational or business model to handle it. Crucial that we fix this.”

2 | Healthcare is moving from episodic to nearly real-time care

The question of compensation for MyChart messaging lays bare the fact that medicine is becoming more real-time. It used to be that you went to the doctor once a year for your physical. And maybe you visited when you had a sore throat. Beyond the occasional call for refills, this was the burden to the system from the typical patient.

But this has changed:

1. Patients have new and expanding points of access. Patients now have more frictionless ways to connect with providers and healthcare systems. This is a consequence of new technology, portals, etc.

2. Consumer expectations. We are all accustomed to near constant connection. Patients translate these expectations to their healthcare experience.

3. Cures Rule. With the 21st century Cures Rule, the disclosure of results has became a real-time thing. As tests get released, patients want to engage at that moment, or shortly after.

4. Provider reinforcement. Under the assumption that we believed we had to respond in near real-time, we have normalized an unsustainable anytime anywhere communication model.

These changes have facilitated a new clinical backchannel between doctors and patients. This is the ongoing chatter that happens in the space between traditional encounters like office and e-visits.

3 | improved access doesn’t mean improved communication

With the blossoming backchannel, there’s a widening asymmetry between patient demand and provider bandwidth. While tech-mediated ways of reaching a clinic may increase, the attention of responding providers doesn’t. Doctors work in what I call a zero-sum medical day — there are only so many hours to handle inputs. It’s the physics of physician capacity: You can’t add something without taking something away.

Many clinicians handle the clinical backchannel by hustling after hours. What’s important here is that compensation for MyChart touches won’t fix the bandwidth issue unless revenue can be used to offset the salary of virtualist. (In 2017 I wrote that the virtualist had no future — maybe I was wrong.).

I asked GPT-3 to create an oil painting of a doctor at the Epic headquarters. This is what I got.

So what do we do?

Here are some thoughts.

Healthcare delivery has to evolve to be platform agnostic. Okay. so this is a bit of magical thinking: The tools we use to address a patient’s need should be irrelevant. Complex problem solving and care planning needs to be compensated independent of the medium used to connect provider and patient.

It’s magical thinking because this currently isn’t the case. Reality dictates that more complicated issues will be kicked to transaction types that are reimbursable. It’s just a function of where we’re at.

If you look at the Cleveland Clinic criteria, I think they get close to parsing out simple questions from complex problems, which is where the compensation solution is found.

Health tech needs strong governance. Some call the MyChart message mushroom an Epic problem. Actually this is a human problem. This is a failure to create parameters or guidance for clinical users and patients. How we use (or don’t use) a certain technology has to be a new priority of medical leadership. This includes boundaries and clear standards of message response time.

A new relationship with expectations. Once we agree on how a tool will be used, patients need guidance. Patients need to know which tool to use to reach their health provider with and when. For example, these scenarios are great examples of how to use MyChart. And these scenarios will need an office or virtual visit with the doctor. Helping patients understand how to manage a system’s communication wardrobe will go a long way in improving the satisfaction of their encounters.

While practice styles vary, patient expectations need to be managed by the consistent use by practices. Ideally, this should be reflected across an institution although variability by specialty is a reality.

Smart systems might help. Algorithms, machine learning and bots may ultimately be able to handle, or at least triage, some of these issues.

---

There’s an attitude in medicine that technology is deterministic — it is what it is and we simply take what we’re given. But physicians, in collaboration with patients, need to be part of the conversation that shapes when and how we use new technologies.

Andrew McAfee and Erik Brynjolfsson had it right in Machine, Platform, Crowd – Harnessing Our Digital Future:

So we should ask not “What will technology do to us?” but rather “What do we want to do with technology?” More than ever before, what matters is thinking deeply about what we want. Having more power and more choices means that our values are more important than ever.

Thanks again for reading. I would love your comments on fee-for-message care or the GPT-3 generated image of the doctor at the Epic headquarters.