This month, Sanjay Basu, Waymark’s Head of Clinical, took the opportunity to speak with Dr. Urmimala Sarkar MD MPH, a Waymark advisor, Professor of Medicine at the University of California San Francisco, and Associate Director of the UCSF Center for Vulnerable Populations. Urmimala describes her experience as a primary care physician at San Francisco General Hospital, and her latest research on improving primary care in safety-net settings.
Sanjay: We know each other from work in safety-net primary care practices in San Francisco, which often take care of patients who receive Medicaid. Can you tell us a little bit more about your history caring for patients in safety net primary care practices?
Urmimala: When I started my internal medicine training in 2002 at UCSF, I was privileged to work at the General Medicine Clinic at San Francisco General Hospital, which is the primary care site for low-income, chronically-ill patients cared for in our integrated healthcare system. I have now practiced there for 20 years. The clinic and the physicians there are what led me into my career in primary care. I didn't come into my internal medicine training knowing that I was going to be a primary care physician. I knew that I was going to focus my career on health equity, but the experience of developing relationships with my patients over time influenced me towards a primary care career.
Sanjay: Some of your recent work on the topic of primary care in the safety-net focuses on the often-asked question: under what circumstances should we provide tele-support for patients, and under what circumstances should we focus on getting them into in-person visits? You recently studied some of the experiences that professionals in health safety-net settings have had with telemedicine, especially with patients of limited English proficiency. Can you tell us about what prompted the work and the take-home messages for your colleagues?
Urmimala: Over the pandemic we've obviously flipped to telemedicine very quickly. These types of disruptions surface and amplify existing weaknesses or inequities in the system. I think the rapid shift to telemedicine amplified and surfaced a lot of the inequities in how innovations are delivered–both structurally and interpersonally. This is not at all unexpected, but if you look at the extent to which patients are getting audio-only visits versus video visits (which allow for greater contextual information and probably provide better care), low income patients are predominantly getting audio-only visits. Part of what needs to be understood is that this is not just a patient issue. It’s true that there are issues with broadband access, issues with having data plans, and issues with having devices on the patient side.
But it’s also true of where people seek care. We don't have webcams in our clinic. We don't have microphones. When we did a physician survey about telehealth needs in the early part of the pandemic, we found that a lot of safety-net physicians are using their personal devices to do telehealth in the safety net. There are many FQHCs, for example, that don't have good broadband.
So a lot of my recent work has been about rethinking how to get to equity by addressing a breadth of barriers, rather than just trying to slice off one piece at a time. There’s definitely a role for digital literacy training for our patients, but I also think there's a role for properly equipping sites of care and the physicians who predominantly care for these patients.
Sanjay: Related to that, one of the challenges Waymark is hoping to address is the deployment of technologies that help our teams interact with patients. We want to efficiently relay back to the primary care provider whichever subset of that information is most useful and relevant, but not overwhelming. One of your recent studies on electronic patient-generated data for chronic disease management focused on the barriers to effective use of those data in safety-net clinics. Can you talk more about that research?
Urmimala: I think this is a case where technology is useful for us. I don't think technical solutions are what is needed much of the time. Usually what’s needed are workflows and people-power rather than technology. But I do think a major technology gap in patient-generated data is in the processing of that data for clinicians. We hear so much talk about AI and medicine, and I would like some AI processing to effectively synthesize the large amount of patient generated data in a way that makes it useful to me for clinical decision making.
For example, let’s take a pedometer. A pedometer is such a useful device because if people are having a sudden decrease in the number of steps they're taking, it's often a signal that they're having a new health issue. If my patient who was active and had a pedometer, and I could understand what a decrease in activity meant in an automated fashion, that would be useful to me. Say I had a patient with heart failure who normally walked 4,000 steps a day, then one day their steps were under 1,000. I would not want a notification right away, or I would be drowning in notifications. I would want a chatbot sending them an SMS message and saying, “Hey you haven't walked. What's going on with you?” And they might say “Oh, my pedometer doesn't have any batteries,” or “I put it to charge and then I forgot about it.” But every once in a while they might say “Actually my legs are swollen and I’m short of breath.” Then instead of translating into a message in my inbox, if that could translate to an urgent scheduling mechanism or a message to a nurse who could then call the patient, it would help me make a decision.
Sanjay: You are alluding to my next question, which is on the topic of filtering information and workloads that reach the primary care provider. We are seeking to work with primary care providers who are, at the very least, overwhelmed and who might even meet the standards for burnout. We've noticed that you've done a lot of work on the topic of burnout and I'm curious what you would have us relay to our partners about supporting providers with the goal of promoting more healthy and more resilient work?
Urmimala: The thing that I have noticed about burnout is that there is a really clear association between control over your work environment and wellbeing at work. Providers who have the ability to give input into their work environment are less likely to be burned out. The burnout interventions that are suggested, for example, by the American Medical Association assumes two things. One is that there is adequate staffing to have truly team based care, which is important. The second assumption is that technical tools can be adopted for increased efficiency.
What I see in safety net primary care practices is that there is not a lot of staff to truly accomplish team-based care in the safety net. The absence of a functioning team means that responsibilities default to the primary care provider.
In addition, to increase provider satisfaction, you have to get rid of stupid stuff. By stupid stuff, I mean time spent on useless paperwork or processes. In the safety net, people are often using the least expensive electronic health record that they can find. And they're using it out of the box. They're not able to spend money customizing in a way that would make sense for the local context and would streamline the provider’s workflows.
A structural change that is really important is to resource the people who deal with the technical tools to be able to fix things that cause burnout. Part of that is advocating for more usable systems. And part of that is investing locally, which is a hard sell when people are trying to keep the doors open.
Sanjay: As a primary care provider, I can certainly sympathize with that. Are there other topics that we haven't touched on that you would want our audience to know about in particular?
Urmimala: One theme that runs throughout my work is the potential of the primary care medical home. Even in under-resourced safety net settings, there’s really good data that we can do a lot for our patients, whether it's getting them to quit smoking or reducing their blood pressure. One of the things that I really like about Waymark, and why I agreed to advise you, is because there is a shared understanding that these programs need to emanate from primary care, as opposed to being carve-out programs. I really think that is an important evidence-based point to emphasize.