Q&A: UW Prof on Whether Patient Portals Are Working as Designed

The percentage of U.S. hospitals that rely entirely on paper records to document information on their patients now stands at 1 percent, down from 69 percent in 2003, according to a study published recently in the American Journal of Health-System Pharmacy.

For more than two decades, hospitals and clinics have been working to digitize their medical records in tandem with companies that develop electronic health records (EHR) software. These records systems allow users at healthcare organizations to do everything from scheduling appointments to documenting information on patients in clinical settings to tracking the charges that end up on hospital bills. Several of the firms that develop medical records software have even introduced products designed to be used by patients and improve their healthcare experience.

As Xconomy reported last week, many in the healthcare industry believe that tools in the latter category, often referred to as “patient portals,” can improve patient-physician communication, and help both parties save time.

For example, some patient portals allow patients to securely exchange messages with their primary care providers and other clinicians. On one hand, many doctors view the ability to send electronic messages to patients as an improvement over communicating by telephone. However, it’s not clear whether this form of written communication—either via patient portals or e-mail—makes it more likely that patients will be able to get the information and advice they need without having to come in for a doctor’s visit.

A study published in the journal Management Science in July looked at some of the effects of allowing patients to exchange messages with providers at a large U.S. health system (the name of the organization wasn’t disclosed). The study, co-authored by researchers at the University of Wisconsin-Madison and University of Pennsylvania, found that allowing these exchanges, or “e-visits,” did not reduce the volume of office visits at the health system or allow doctors to see more new patients.

One of the study’s authors is Hessam Bavafa, an assistant professor of operations and information management at the Wisconsin School of Business. Bavafa recently spoke with Xconomy about some of the findings in the study, and what implications it might have for patient-physician communication. This transcript of our conversation has been lightly edited for clarity.

Xconomy: What made you and your co-authors decide to do this study?

Hessam Bavafa: One of the motivations of this study is that it’s always interesting to see, as we introduce new channels of service or product delivery, what happens to the customer experience.

I think about myself as a teacher at the university. Students e-mail me questions and I can respond to their emails. I think about an e-mail sent to me as an e-visit, basically an e-consultation. Many of the problems that [lead] students to send e-mails to me, they don’t have to come to my office hours. I can just send them a two-line answer and it’s over.

What about healthcare? Does this trend—this substitute-ability of e-visits and office visits that I see in my teaching—translate to healthcare?

X: Tell me about the study design. The dataset you all built comprises more than 140,000 patients and 90 doctors, correct?

HB: Yes. This is one health system, which runs multiple primary care practices, [studied] over five years and two months. We have data from nine primary care practices.

X: Were the “e-visits” you all examined as a part of this study limited to text-based communication, or did some of them also take place using videoconference tools?

HB: In this study, we focused particularly on text and telephone interactions. In our data, we don’t observe any video interactions. I think those are very exciting and definitely the next step [after] this study.

X: If you go back in time, say, a decade ago, when many health systems were installing new EHR software and introducing patient portals for the first time, do you think the belief at the time was that patient portals would free up doctors’ time and make their jobs easier?

HB: I think that was the premise. [Some reports estimated health systems could] expect to save millions of dollars in terms of fewer office visits, doctors being able to handle more patients, and similar things. So I think that was definitely one of the goals and expectations.

X: One of the key (and to some, surprising) findings in this study was that as the number of e-visits increased, so did the number of office visits, reducing doctors’ capacity to take on new patients. And you all didn’t observe any additional health benefits to patients following the increase in e-visits. Do you think those findings might lead physicians and health systems to rethink their use of patient portals?

HB: I really hope so. The goal of our study was to basically paint a picture: what is happening? We advertise these [portals]. Many people champion these in the health system, to be using these more and more. What we show is that they don’t seem to be doing some of the things that we were expecting, in terms of improving patient health.

I think the best thing that can come out of this study is more research into figuring out how to incorporate e-visits in healthcare, to achieve the real goals: reducing costs, improving care quality, and improving the work of doctors. We need more research into seeing how we can leverage technology to improve patient health, reduce costs, and improve the doctor’s life. Just throwing technology at something doesn’t necessarily help.

X: What about the companies that develop electronic health records software, like Athenahealth (NASDAQ: ATHN), Cerner (NASDAQ: CERN), and Epic Systems? Do you think they ought to rethink the design of their patient portals?

HB: I think that’s highly possible. In the long run, [records software like] Epic is going to be the operating system of hospitals—the Microsoft (NASDAQ: MSFT) of a hospital.

Microsoft was designing its operating systems: Windows 95, Windows 98. Over time, it learned from Apple (NASDAQ: AAPL). It learned from Linux. [Microsoft] basically found the nice features in all of them and started incorporating it into Windows.

It’s quite possible you’ll see a similar pattern from Epic. Epic is observing all these companies that are experimenting with different forms of e-visits. I think it’s entirely possible that Epic [may be in a position to] re-design e-visits as this unfolds, as health systems and Epic itself learns more about how to make these efficient.

X: What type of feedback have you and your co-authors received following the study’s publication? 

HB: We have been in constant touch with the health system that we have been working with. They have been very excited to know more [about] what’s going on. They are providing us with more recent data to do more research on the topic.

A few startups e-mailed me [asking], “What about video?” “What about payments?” I think that’s exactly the goal. I wanted this research to start a conversation and try to push health systems across the country toward learning from each other. Who does what that is right and makes this work? What is the right model?

Right now, in this health system and many others, doctors get zero payments for e-mailing patients. I have another study coming up. [It’s] basically a modeling exercise [to determine] what the best way of paying doctors for these visits [is].

X: What do you think the findings of the study might suggest in terms of the larger picture with e-visits and telemedicine? 

HB: Think about Amazon (NASDAQ: AMZN). People weren’t talking about online shopping until Amazon really materialized this concept in such a nice way. People weren’t talking about hailing cabs [online] until [companies] like Uber and Lyft materialized. They kind of figured out how to [do it].

I think of e-visits as the same. I’ve been working on this for a while with multiple health systems. There’s so many ways of interacting with your doctor remotely. There’s video interaction [and] real-time text interaction. There are remote monitoring devices. I think the big question is, “How can we figure this [telemedicine] thing out in a way that reduces healthcare costs, improves patient engagement and quality of care, and frees up doctors’ time?”

Trending on Xconomy