Much like the HIV care cascade, one exists for the hepatitis C (HCV) -- that is basically a map of where a patient is on their journey, whether they've just been diagnosed, linked to care, on treatment, or successfully completed treatment. One study used electronic medical records to help make it easier and quicker to place individuals living with HCV in the care cascade. Terri Wilder, M.S.W., spoke with Jason Zucker, M.D., about the study, which was presented at CROI 2017, in Seattle.
Terri Wilder: Please tell us about your study.
Jason Zucker, M.D.: Sure. With the release of direct acting antivirals, we can now treat hepatitis C. With that, there's been more of a push to get patients diagnosed, linked to care and then, eventually, through treatment. Similar to the model that's been used for years with HIV, we're planning to hire clinical patient-care navigators to help bring patients through the process: help them from the moment they're identified to get linked to care, to help them get insurance approval and treated.
In order to do that, we wanted to develop a tool that our navigators could use to figure out who was hep C positive, and who was diagnosed, and where those patients were in the cascade of care, without having to go through it manually and update it frequently and spend a lot of time and resources on that aspect of it. So, we first went through and identified all of the antibody tests that could be run at our institution, [and] we found nine different antibody tests; [then] the different viral loads that could be run at our institution, [and] we found multiple of those, as well; and then methods for identifying which patients were linked to care, prescribed treatment and then reached sustained virologic response -- all pulled from the electronic medical record [EMR].
Then, we developed an algorithm that was able to place each individual patient into the correct location in the cascade of care. And then, we used a person who was unfamiliar with what we were doing to manually validate. So, they went through all the charts by hand and were able to validate the accuracy of the algorithm.
Essentially, in the end, we had 1,225 patients with new institutional diagnoses of hepatitis C. Only a little under 60% had hep C RNA sent. Only about 27%, or about a quarter, were linked to care. And then only 8% of patients were even prescribed treatment.
The algorithm was able to correctly categorize 90% of our patients. For all the ones that were not correctly categorized, those were all under-called, in that they were later in the care cascade than we expected them to be. Some of them had been linked to care at outside institutions that you could determine from the notes, or had received treatment from another provider. That's something that our clinical navigators would be able to work through as they went through the process.
TW: Is your EMR algorithm able to help prompt the actual offer of the hep C test? Because you're in New York state and you have a hepatitis C testing law.
JZ: We do. So, that's actually something we're working on right now. That was not part of this project, but we are working on developing a prompt to increase the rates of hep C testing institution-wide, because they have been pretty poor. We don't have the denominator of the people who probably should have been tested but haven't been, yet. But we are actively working on that. Hopefully we'll be having a prompt and will be able to follow up on the education aspect, as well.
TW: I'm kind of going off on a tangent, I realize, because this is not what your poster's about, but a lot of the CDC recommendations, as well as [those in] New York state are around this cohort, 1945 to 1965. But since we're at CROI, we're also thinking about HIV, thinking about what folks with HIV are being offered, and what would that look like in an EMR prompt.
JZ: We talk a lot about the birth cohort. But there are a lot of other risk factors for hepatitis C. And there's more and more data, including a Centers for Disease Control and Prevention Morbidity and Mortality Weekly Report from a couple weeks ago showing that there are surprisingly high rates of hepatitis C in people who are younger than the birth cohort -- which was unexpected. So, I think it's important to screen more than just people who are in the birth cohort. You need to look for risk factors. The risk factors for hepatitis C are very similar to that for HIV. So, anybody who's being tested for HIV, you should at least consider hepatitis C testing.
I think we need to be looking at testing a wider swath of people than just the birth cohort. Our hope is that that's something we'll be able to implement, as well.
TW: And women of childbearing age.
JZ: Absolutely. And women of childbearing age.
TW: Right. Can you tell me a little more about what your hopes are in terms of this for practicing medical providers? I know that you said that you're going to hire some patient navigators. But what is the role of the medical provider in having something like an EMR?
JZ: I think the providers play a very important role because we're all part of the health care system. We're the ones who have the first contact with patients -- whether it's patient providers in the emergency department, in urgent care, or primary care clinics; it's all our responsibility to try to identify people who need additional services and get them into the proper places.
So, trying to increase screening campus-wide is a really important part of our plan, including education sessions, EMR-based prompts and other things. Because if we can get the patients identified, then we can actually use our system to take some of the responsibility off the providers by helping them get these people linked to care. But, we need the providers to really step up and help us with the testing part of it.
TW: Do you think having an EMR-based algorithm is something that could be reproduced in other states?
JZ: Yes. It's definitely reproducible. What we learned from doing this is that every institution has their own names for the tests and has their own way that information is pulled from the EMR. However, we actually built this in an external system so that any data could be loaded into it. You'd have to obviously update it and change it for each individual system, but it certainly is replicable.
TW: Right. I'm going to play a little devil's advocate here. There's rumors that there are medical providers who aren't on an EMR; they're still using paper charts. What do we do about those folks?
JZ: I think it's very difficult for people who don't want to use electronic medical records. I think we can do some of the things we did in the old days, because people who aren't using EMRs are probably using paper forms. And including checkboxes on paper forms, reminding providers (the same way the electronic health record reminds them) that they need to be screening is one method. Hopefully they'll join us soon.
TW: Right. All right. Thank you so much.
JZ: No problem. It was very nice meeting you.
This transcript has been lightly edited for clarity.
Terri L. Wilder, M.S.W., is a director of HIV/AIDS education and training in New York City.