PROVIDENCE – It was the afternoon before Christmas, Dec. 24, 2018, and all through the Rhode Island Hospital cafeteria, everyone was stirring. A steady stream of nurses, doctors, orderlies and visitors entered and exited the large window-filled room, creating a constant murmur of sound, much like the distortion from an unfiltered hearing aid.

Call it the hospital noisy zone; there was an ever-present hum from face-to-face, in-person conversations: pockets of laughter and cheer from impromptu staff holiday celebrations; the hushed, contemplative exchanges between relatives of patients; and business-like consults between doctors about patients.

Some were there just to grab a cup of coffee or a quick bite to eat, in a rush to finish up work and rejoin their family before the onset of the holiday. On the menu for the hot lunch was a “healthy” variation of American chop suey, featuring spinach or kale as an ingredient, along with garlic green beans, pre-holiday fare.

Underneath the orchestra of spoken voices, if you listened closely, during the pauses in the natural ebb and flow of conversations, you could detect an occasional faint if not familiar tone of beeps and musical rings of smart phones, laptops and pagers.

ConvergenceRI was there to interview Dr. Rebekah L. Gardner, to talk about the recent study she co-authored, “Physician stress and burnout: the impact of health information technology,” published on Wednesday, Dec. 5, in the Journal of the American Medical Informatics Association. [Her co-authors included: Emily Cooper, Jacqueline Haskell and Daniel Harris of Healthcentric Advisors, and Sara Poplau, Dr. Philip Kroth and Dr. Mark Linzer.]

In her ongoing research, Gardner has been tracking and measuring the unintended consequences – stress and burnout – as physicians have become tethered to the 21st century stethoscope: electronic health records.

The hospital cafeteria, ConvergenceRI realized a few days later, had perhaps evolved into a stress-free zone from EHRs for health care practitioners, where non-electronic, in-person conversations took dominion.

Measuring physician stress and burnout
The new study was a direct outgrowth of earlier research conducted by Gardner, who is senior medical scientist at Healthcentric Advisors, an associate professor of Medicine at Brown University, and an internal medicine physician with Brown Physicians, Inc., at Rhode Island Hospital. That work had also been published by the Journal of the American Medical Informatics Association in June of 2017. [See link below to ConvergenceRI story, “New kind of pain scale: what doctors rally think about EHRs.”]

That first study had published the results of an analysis based largely on the qualitative narrative data captured in the 2014 Health Information Technology survey conducted by the R.I. Department of Health. It ran under the provocative headline: “‘It is like texting at the dinner table’: A qualitative analysis of the impact of electronic health records on patient-physician interaction.”

“We’re either left fumbling through data entry with our patient in the exam room, missing out on an opportunity to truly connect, or we’re left with hours of documentation and computer work after a long day of seeing patients,” Gardner had told Reuters reporter Ronnie Cohen in an interview after the first study was published, for a story that ran on July 28, 2017, entitled: “Doctors frustrated that electronic records steal time from patients.”

Stress test
In response to the findings of the first study that Gardner had co-authored, which had analyzed written “optional narratives” provided by physicians, the R.I. Department of Health added specific questions in its 2017 health information technology survey conducted of some 4,200 practicing physicians in Rhode Island, asking about burnout and stress from the use of EHRs.

[The HIT survey, mandated by the R.I. General Assembly and is conducted every two years under the auspices of the R.I. Department of Health; it currently surveys only practicing physicians and not nurses.]

Of the 1,792 physicians who responded the 2017 HIT survey [a 43 percent response rate], 26 percent reported burnout; among those who used EHRs [91 percent], 70 percent reported EHR-related stress, with the highest prevalence in primary care-oriented specialties, according to the study.

The study concluded: “Given the toll of burnout on clinicians, patients, and the health care system, measuring and addressing HIT-related stress is an important step in reducing workforce burden and improving the care of our patients.”

About half of the physicians who responded reported “insufficient time for documentation,” according to the study. Prevalence of stress was found to be highest among physicians in the traditional primary care specialties: general internal medicine, family medicine, and pediatrics.

Further, the study posited that such time pressure “poses a direct challenge to connecting with patients, one of the more sustaining aspects of primary care practice.” Another factor the study mentioned was that documentation time was generally not reimbursed.

Translated, “Physicians may feel that they are spending a large portion of their time on complex and time-consuming work that does not benefit their patients,” according to the study.

Here is the ConvergenceRI interview with Dr. Rebekah Gardner, discussing her latest research and findings about the ways that the advent of health information technology in the practice of medicine is contributing to physician burnout and stress.

ConvergenceRI: Last year’s study was not from any direct survey questions. It was from 2014 data that was qualitative analysis of optional comments by physicians surveyed. Is that correct?
GARDNER: We did ask them a survey question for that study which asked them to reflect on how using EHRs impacted their interaction with their patients. So, there was a survey question prompt that led to the qualitative analysis piece.

ConvergenceRI: This time, there was clearly a series of questions that was built into the survey, in response to the first study. Is that an accurate description?
GARDNER: Absolutely. I would say that it definitely contributed to the reasons why we included more specific questions. We heard physicians’ [concerns] loud and clear; we heard about their unhappiness.

The comments we analyzed – and the themes that we discussed [in the first study were topics] that have been discussed by many [other] researchers in this field.

This is a very well studied field. There are many researchers who are looking at wellbeing, stress and burnout among physicians.

I certainly would not want to discount anything else that’s going on by researchers across the country. This is a very well explored area.

ConvergenceRI: Was Rhode Island’s relative small size as a state but “large” sample size an advantage?
GARDNER: What our study adds is that we have a large sample size, we are able to ask these questions to a very diverse array of physicians who are in different work settings, who have different specialties.

A lot of the work that has been done before in this area has been done among general medicine specialties, primary care specialties, and it had often been done with focus groups.

It was nice that we were able to take a more quantitative approach with [such] a broad cross-section.

ConvergenceRI: Within that framework, one of the things that your study was able to look at in detail was whether or not the addition of scribes was a possible “panacea,” [a cure-all], as a way to reduce stress.
GARDNER: I don’t think that there is anything that’s a panacea.

ConvergenceRI: It has been suggested by some that adding scribes was a way to lessen stress and improve physician-patient interaction. The findings of your study seemed to contradict that, by suggesting one of the stressors not being addressed by scribes was the high volume of inbox messaging.
GARDNER: Let me comment a little bit on the scribe analysis. When we did the unadjusted analysis, we did find that folks who had scribes were less likely to have burnout symptoms.

But, when we put everything into our model, having a scribe was no longer associated with reduced burnout.

[Our findings] sort of reflect some mixed literature about this. There have been some studies that show that scribes do improve physicians’ work experiences, and there are others studies that haven’t shown that.

I think that’s really a function of what’s causing physicians to be stressed in their work environment: what [is it] about the technology that is particularly stressful, for which type of physician?

Physicians have very different jobs, based on their settings, specialty and patient population.

So, scribes may solve a problem for a certain subset of physicians for whom the documentation of cases will be the major piece [of their job].

But it may not for other folks who have trouble with [the volume of] inbox texts, as you suggested.

[Pausing]. I guess that is kind of a complex answer.

Studies have shown that doctors may get between 100 and 120 inbox texts coming to them everyday, depending on the type of medicine they practice and their patient panel. So, it is not insignificant.

ConvergenceRI: Between 100 and 120 a day? That’s large, particularly if that is the work you take home at night.
GARDNER: Absolutely. It is very large.

ConvergenceRI: Another question I wondered about, which is not within the current confines of the HIT study, is: how the role of nurses has changed the delivery of health care by physicians in the last few years? Nurses have often become the first line of care that patients interact with in many practices, and some do a lot of the actual work with managing health IT, as I understand it.
GARDNER: I don’t think you can make that generalization. That is really practice-dependent, and I don’t think it is fair to say that.

I think there are some practices that have really embraced the skills and expertise of nurses. For example, nurses might be the first ones to look at the inbox messages and triage a lot of that work, because it is in their scope of practice. And, they feel comfortable, and the practice feels comfortable, with their expertise.

But, for many practices, they can’t afford to hire a nurse; it’s quite an expense to hire an RN, and so they don’t have additional help.

ConvergenceRI: Would that be something to consider putting in the next survey of physicians, regarding the role that nurses can play in reducing stress and burnout?
GARDNER: Yes, absolutely. Our next survey is going to be released in 2019. We are working on the questions now. We’ve included more questions about the inbox, because we’ve heard from physicians that we didn’t capture their experience about the inbox in the previous [2017] survey.

We will be asking questions about the inbox regarding how much time they feel have for those tasks and whether there is anyone in the practice to [help handle] those tasks, whether it is an RN or someone else.

ConvergenceRI: Is there any consideration to expanding the survey to ask patients about their interactions with physicians? It seems that this is the elephant in the room, particularly because of the way that patients are starting to control their own data with personal wear devices. Would that be a line of questioning to be considered in future surveys?
GARDNER: For the 2019 survey, we are hoping to include questions to physicians, asking them how useful information is that might come through on a FitBit or Apple Watch would be to them. Hopefully we will have some more data on those types of topics.

ConvergenceRI: Last January, the president of Epic, Carl Dvorak, gave a talk at the Warren Alpert Medical School. [See link below to ConvergenceRI, “Is it all about the data, the data, the data?”]

In the question-and-answer period that followed, there was a provocative dialogue that occurred between Dvorak and Dr. Elizabeth Toll, you may know her, a pediatrician and an internal medicine specialist at Lifespan.
GARDNER: Yes, I do.

ConvergenceRI: She challenged Dvorak about his belief about how technology was going to reshape the future world of health care delivery. Dvorak had described how Epic had developed motion sensor technology that they could use to tell them how the doctor was interacting with patients.

Toll’s response was to say: “Medicine is also a human undertaking.” The most important part of medicine, she continued, is what she described as: “The need for human contact, the sense of being with another person and helping them feel better by the simple act of creating a relationship with them.”

Dvorak appeared to dismiss what Toll said, saying that the future of medicine was going to be in data, and she just needed to learn how to adapt and adopt.

Is there a way that the next survey can include questions to measure the physicians’ frustration in developing and sustaining a relationship with a patient, and the way that health IT may get in the way of that?
GARDNER: I think that is such an important issue. I think that policy work that we did in the previous paper that we talked about begins to address some of those concerns. I think with an issue that is so complex, this is a place for focus groups. And, more in-depth qualitative approach would probably get you more of the answers you needed than a scale on a survey, which is something that is not easily captured responding to an A, B, or C prompt.

ConvergenceRI: Is that something that could be considered, as the next offshoot of the qualitative analysis of the survey, to look at? I’m asking, because patients, with whom I talk to on a regular basis, are really frustrated by the lack of interaction.
GARDNER: I will echo that I think this is a really important issue. It is one that we, as a profession, are really struggling with. As to what the HIT survey will do, after 2018, I can’t say.

But, I can see why anyone could make a good case [arguing] for [the need to conduct further analysis].

ConvergenceRI: In terms of takeaways, what do you think the most important takeaways are from the survey and your paper about where you think the focus should be, moving forward?
GARDNER: Well, from the paper, I think some of the most important takeaways are that stress related to health information technology is quite prevalent.

It’s measurable, organizations can measure it, and they should measure it. They should look to see within their own workforce where the stresses are.

It’s something that varies by specialty and by setting. Organizations that are looking to put solutions in place or that are looking to improve their situation, they need to work with their physicians and staff to come up with solutions that address the actual stresses in their particular workforce, knowing that it is going to be quite different from another location or another specialty.

I think that someone who comes in and thinks that there is a one-size-fits-all solution for HIT-related stress is going to be disappointed with the results of their intervention.

I think it’s important to keep measuring this.

ConvergenceRI: I found it fascinating to see that from the results of your study, primary care physicians were found to be high up on the stress level.
GARDNER: Correct

ConvergenceRI: Why is that, do you think?
GARDNER: I think there are probably a couple of reasons, based not just on our survey results but also on my own experience.

For primary care physicians, the reason why most of the folks are in that business is because they truly value a longitudinal, close relationship with patients that they see over and over and over again.

I think that anything that can get between themselves and their patients feels particularly stressful.

I think that primary care physicians are often the ones that have an inbox that’s overloaded, so they are feeling stress from that direction.

They are also the ones for whom a lot of these value-based reimbursements are disproportionately affected.

ConvergenceRI: Because they have a greater risk burden placed on them.
GARDNER: Yes. They are also asked to document and to assess many of these preventive, health quality measures. That’s one more additional thing that they are [being asked] to do, on top of everything else. Which affects their documentation, and how they write their notes.

They are feeling a lot of pressure from many, many different directions. It makes sense that their stress level would be high.

[At the same time], there are primary care doctors in Rhode Island who like their EHRs, who are using them effectively, who believe that their EHRs help them provide better care. We also know that there are folks who are thriving with their system. And, I think we can learn something from that, too.

ConvergenceRI: There are a number of innovative approaches to health care delivery in Rhode Island underway. I was wondering how they might factor into how the HIT survey measures the data and how that data is being used.
GARDNER: I’m not sure I understand your question.

ConvergenceRI: There are two, three initiatives. One is health equity zones, which are trying to change the basis of how we come up with health solutions, based upon communities coming up with their own solutions, because 80-90 percent of health outcomes are not determined by what happens in a doctor’s or nurse’s office.

There is an apparent lack of awareness about what health equity zones are. My supposition is that most politicians, most elected officials, most reporters, and most physicians, could not name the nine communities where there are health equity zones in Rhode Island.

The second part is the work that is being done in Central Falls at Blackstone Valley Community Health Center and its Neighborhood Health Station, with a community where three quarters of the residents will be receiving primary care and urgent care from one facility, with a very sophisticated, population health management IT tool.

My question is: how do these initiatives get measured as part of the data analysis as part of the HIT study of practicing physicians? Is there a need to look at how these innovations may be impacting the data results? As you were saying, some physicians are thriving while others are not.
GARDNER: That’s a really interesting question. The HIT survey specifically speaks to individual physicians about their experiences and perceptions. There is something called an inventory survey, which our state does, that sounds like, given that it is more on a macro level, might be a more appropriate place for such questions.

ConvergenceRI: Will the future survey identify whether physicians are members of an accountable entity or an accountable care organization? And whether they see themselves operating as part of a team? And how that changes the dimensions of their experiences as a physician?
GARDNER: In the 2019 survey, we will be asking about that piece.

ConvergenceRI: What kind of feedback mechanism exists for the study and the data? How does it become part of a broader conversation? You’ve done the survey, you’ve completed the study, but is there a need for a broader conversation, for instance, about the growing suicide rate among physicians, related to stress and burnout?
GARDNER: I think you bring up an important point. Any time the state does a survey or collects data we need to share that [information].

This is something that Dr. Nicole Alexander-Scott has emphasized from the first day she started as director of the R.I. Department of Health. It’s very important to her that this data [are shared] so that we can act on them, and that they can influence the conversation we are having across the state.

Over the past year and a half, we have spoken and presented our results in multiple, multiple venues, to many different types of Rhode Islanders, to share what we found, and to make sure that it can influence the work their organizations or groups are doing.

I couldn’t agree more [about the importance of letting] decision-makers and consumer groups and policy makers become more aware of what we found.