Inside theMednet: The Community of Experts Solving Tough Clinical Questions

Paul Sax: [00:00:14] Hello, this is Paul Sax. I'm editor in chief of clinical infectious diseases. And welcome to the Let's Talk ID podcast. In a recent New England Journal of Medicine blog post, I highlighted a few places where clinicians can get input from real human beings, not just AI generated summaries. One, the most old fashioned, was the IDSA listserv. One is a new social media space called Roon, and another was something that just came to my attention recently called MedNet. Today I'm joined by Doctor Patrick Passarelli, the ID editor for MedNet. He's a med-peds ID doc and an assistant professor of medicine and pediatrics at Dartmouth-Hitchcock Medical Center. Patrick, welcome.

Patrick Passarelli: [00:00:55] Thank you so much.

Paul Sax: [00:00:56] So, Patrick, for listeners who may not know you, can you tell us a bit about who you are and what you're doing now?

Patrick Passarelli: [00:01:02] Yes. So I am a med-peds doc and infectious disease trained. So combining those two skills, I was recently hired at Dartmouth-Hitchcock and Geisel School of Medicine to kind of revive their pediatric infectious disease program. So I'm kind of the one-person department Monday through Friday during daytime hours, I do get some coverage overnight and on the weekends, which is really nice. You know, anything you can imagine you do as an infectious disease physician on a daily basis, whether it's seeing inpatient consults or taking calls from community physicians. That's kind of my role. Along with managing the pediatric HIV program and running the pediatric COPAT or OPAT program now, too.

Paul Sax: [00:01:47] So how did you end up in ID? I was interested to know how people ended up, especially med peds, because, you know, as small as the ID community is, the pediatric ID community is even smaller.

Patrick Passarelli: [00:01:58] Well, I'm a little bit unusual in that I kind of trace back my ID roots to when I was probably 11 or 12, and I read The Hot Zone by Richard Preston, which I later realized is somewhat sensationalized, but it's an accounting of Ebola Reston virus, and how it kind of almost escaped into the D.C. community. So I just remember being completely transformed by that book. And then I sought out other books like Level 4: Virus Hunters of the CDC and, you know, the American Plague books like that about infectious disease history. Basically, I kind of always knew I would go into medicine. And I'm also a little bit unusual in that I knew what med-peds was before med school, so I was kind of predetermined med-peds as well. But always as an avenue to do ID. And then when I was finishing my medical training, my wife was starting hers. So while she was in medical school, I worked as a hospitalist for three years. Couldn't wait any longer. And towards the end of her training started my fellowship. And then after fellowship, was trying to be closer to her in New England. So ended up at Dartmouth.

Paul Sax: [00:03:04] So tell us, we're going to talk about MedNet. Now, how did it get started?

Patrick Passarelli: [00:03:07] So it's interesting because I didn't know this until well into me using it. It was actually founded by a brother sister duo, Nadine and Samir Housri, who were looking for answers when their own father was diagnosed with prostate cancer in 2014. And so, Nadine, as a trained radiation oncologist and Samir as a software engineer, kind of teamed up. And knowing that some of the most, you know, valuable opinions and knowledge in medicine come from experts, she sought out experts in the field using her connections. But then based on that experience, they wanted to kind of bring that to a larger audience and allow people to do
that on a more, more accessible basis.

Paul Sax: [00:03:51] It started in oncology, it sounds like.

Patrick Passarelli: [00:03:53] Yeah, it started in oncology. And then as it got more popular over the years, they expanded to include other fields. So rheumatology is actually another huge component of MedNet. I think about 80% of rheumatologists in the US are members, and then the infectious disease section actually started more recently in 2024 and just kind of grew out of popular support and spilled into infectious disease. And our most recent section is general medicine. And the next couple areas to be added are in the surgical subspecialties.

Paul Sax: [00:04:27] So when did it first get started? You said when ID started, but when did MedNet start?

Patrick Passarelli: [00:04:31] 2014.

Paul Sax: [00:04:32] 2014, so it's been around a while? So I guess the reason why I haven't heard of it is because ID is only relatively recently joining the party. And your involvement, how did you get involved?

Patrick Passarelli: [00:04:42] So I got an email when they were building the ID community, and I was a fellow at University of Rochester at the time. And so I just got an email like, would you like to participate in our fellows program, which is kind of catered specifically to fellows and basically just kind of has more active soliciting of questions. You know, not knowing the other fellow's participation, I was always pretty diligent and tried to submit a question a week. And usually it was from, you know, naturally came from discussions on rounds. Or I'd be in clinical conference and it would kind of we'd be discussing a topic where there very clearly wasn't a consensus or people just didn't know. And so that was often an impetus for me to go on MedNet and ask a question. And then I didn't know this because it's all behind the scenes, but as I was graduating, I was approached by now my editor, who manages several sections within the medicine section, and he asked if I wanted to take part more extensively as an editor. So I actually started doing that in August before my job at Dartmouth.

Paul Sax: [00:05:47] You must have submitted some really excellent questions.

Patrick Passarelli: [00:05:49] Yeah, my most popular one was the one on measles that you mentioned in the article. Kind of similar to Nadine and Samir's story, that was for me personally, I was trying to help my dad, who had just gone through chemotherapy for lymphoma, and I just was like, I don't know. He's born before 1957, but would he benefit from another MMR? You know, my parents live in Nebraska, so at the time the Texas outbreak was happening and still is happening, unfortunately. But it was similar to that. I was kind of looking for more personal answers.

Paul Sax: [00:06:22] Just for listeners who don't know the question, I'm going to repeat it. In light of recent measles outbreaks in the United States, would you recommend an MMR booster for immunocompetent patients born before 1957? The answer? Generally not. So I mean, sometimes we check titers, but anyway, really interesting to see the responses. And it's really interesting also to just to see what the good questions are. So what makes a good question?

Patrick Passarelli: [00:06:49] As I alluded to, it's kind of those questions where, you know, you're already among experts maybe in the field and talking to an attending or, you know, you've already gone through the literature and really just can't find any solid evidence to answer a clinical question. Those are the ones that really make for great MedNet questions. Because, you know, if you're wondering what the answer is, a lot of other smart people are to. The questions that I reject are usually ones that are pretty easily addressed by major studies or randomized controlled trials. So it's really that gray zone, which I've been learning more and more over the past couple of years, is really where ID physicians thrive.

Paul Sax: [00:07:26] I highlighted two others, and again, I'm just going to cite them because they struck me as just so spot on for clinicians doing ID practice. One is, do you recommend a prolonged duration of antibiotics and or suppression for patients without preexisting hardware, who have placement of new hardware after decompression washout of a staph aureus epidural abscess? What a great question. And then the other one was an area of particular interest to me. What is your preferred third antimicrobial agent for a patient with treatment-naive Pulmonary Mycobacterium avium Complex (MAC) without cavitary disease, and strict contraindications to utilization of rifampin or rifabutin? Wow, that's good stuff. Well done.

Patrick Passarelli: [00:08:06] Very specific. [laughs]

Paul Sax: [00:08:07] Those are the kind of questions that obviously generate ID consults that other specialists turn to us for. I think it's terrific. All right. So what do you do as editor?
Patrick Passarelli: [00:08:18] So my job is really once a question is submitted, I get notified and then I review the question. If you know there's grammar mistakes or, I was actually an English major, so using those skills, trying to make it more succinct. We don't tend to like parentheses with a lot of additional information. So really just kind of editing the question. My main role is to connect it to experts and then filter those questions and reach out to experts for whom that question is relevant. So for NTM, for example, you know, we have a couple NTM and TB experts that all send questions to frequently. Periodically we'll engage in not just the regular questions, but high yield questions. And then we help make polls out of the questions, too. So if it's something that's amenable to like a multiple choice, we'll release those in like a daily digest and have people participate in those.

Paul Sax: [00:09:10] Do you have experts answer all the questions or?

Patrick Passarelli: [00:09:13] No, and that's where there are still questions that have never been answered. Either the expert just never responded or, you know, they are stumped, I suppose. So there are questions out there that just still don't have an answer. And that's where periodically there are new members that join and will answer questions. And I also get notified about that too.

Paul Sax: [00:09:33] Where do you find the experts? How do you know who they are?

Patrick Passarelli: [00:09:36] So usually it's through kind of their research prowess. We really target people at academic centers. I learned a term that's more of a business term, like a KOL, kind of KOLs of medicine, or key opinion leaders. That's generally how we find people. And part of my job as editor is to kind of recruit new experts as well, although it's not something I have, you know, my primary job is to do the questions, so it's not as often that I get a chance to kind of reach out and cold email people.

Paul Sax: [00:10:05] Yeah. So you brought up a business term. So I'm going to bring up a business issue, which is do you pay your experts?

Patrick Passarelli: [00:10:11] No. And they're so generous. You know, there are various methods of answering. You can tell some people are a little bit more casual, but then some people, you can tell they do spend a lot of time on it and will cite different papers. I think it kind of ranges the gambit. But yeah, we are always very grateful because we know it's people who are already very stretched with their time and are generous enough to participate.

Paul Sax: [00:10:35] This sort of brings us to the key premise of my blog post, which is that this is about human input. But we are, of course, living in a moment where AI can, like, instantly summarize guidelines and look through the literature. So where does MedNet fit into that landscape?

Patrick Passarelli: [00:10:52] Yeah. So I think the biggest difference and it kind of using that framework of like fast thinking versus slow thinking, are AI tools that we're all familiar with now are kind of more the fast thinking. MedNet represents kind of more the slow thinking. I guess another way to put it is that a lot of the AI tools are based on data retrieval, and this is more capturing and trying to scale up the collective wisdom of experienced experts. So to me, it's kind of like the AI tools plus, and vetted by humans and like people that you kind of know as your the attendings that you trained with. And really that sense of community as well. The format is a forum, but then everyone's picture is posted along when they post an answer. So it really just kind of reminds you that when you search an AI tool, you just kind of get that robotic response, whereas it's really nice. I'm a people person, so for me especially, it's like, oh, there's that person and there's their smiling face.

Paul Sax: [00:11:50] So where do you see MedNet going in the next few years?

Patrick Passarelli: [00:11:53] We're kind of slowly expanding into different fields. Speaking of AI, we are moving kind of into that space. I'll just leave the teaser for there, and people can kind of discover in the next couple of weeks what that means. And so kind of integrating that as well into our forum-based site for now. But I'm always hoping for more participation. I kind of take for granted, you know, it's so much a part of my world now that, you know, I forget so many people, friends from other institutions throughout my training. I forget to tell them about it. So every once in a while I'll come up and I'm like, oh, I didn't tell you. This is like a big part of what I do now. And then I'll tell them about it, and then they join and oftentimes find it useful.

Paul Sax: [00:12:33] Yeah, I wrote in the post, I said, you know, AI is trying to make itself better at these common questions that have no answer. You know, these edge cases. And I think that the repository of questions that you have really would be a goldmine for them to train on, frankly. So very, very interesting that you're planning to do that. All right. Anything that has happened on the site that really surprises you?

Patrick Passarelli: [00:13:02] I guess every once in a while there are some vigorous debates. So things that to a layperson would just be like, what are you nerds talking about? People do get really fired up. So that is one thing that surprised me, is that there is some passion that comes with it that everyone's like, oh, wow.

Paul Sax: [00:13:21] One thing that's nice about the site that's different from, say, med Twitter is that everybody's identified. You lack the anonymous people and the bots and the other people who create such an angry and contentious space in social media. So that's a good thing. Can you think of any examples when answers to questions change how you practice?

Patrick Passarelli: [00:13:42] I definitely can. It's more just almost like an approach to a problem or a question that I never would have seen it that way, but definitely in my fellowship. I guess widening the scope of opinions from, like, the attendings I was working with. It's more than just like necessarily a practice changing question. It's more just like, oh wow, I never really thought about it that way, which kind of reinforces that discussion based and community based aspect of the site.

Paul Sax: [00:14:09] Great. Well, listen, I'm going to finish by asking you a few lightning round questions that I sometimes surprise people with. Okay are you ready? What's your favorite antibiotic?

Patrick Passarelli: [00:14:18] As an ex hospitalist, I think I still have to say ceftriaxone. But as an ID person, doxy. I mean, it sounds kind of cliche, but just there's so much that doxy does, so it's a toss up between those two.

Paul Sax: [00:14:34] I think a very appropriate for med-peds doctor to choose both of them. Why not? What do you like to do in your spare time?

Patrick Passarelli: [00:14:41] So I recently learned how to snowboard last year, and so this winter has been a good one to continue practicing. And I just went to Remington Winter course for the first time. And so I got some more, um, blues in. And I'm still progressing. But living in New England is also a great place to be a birder.

Paul Sax: [00:15:02] So we have a lot in common. I'm married to a pediatrician. I was an English major. My wife is also a very avid birder, but I was never able to master snowboarding, and I say that as one of the great disappointments of my athletic career, I found it very, very difficult.

Patrick Passarelli: [00:15:16] Well, it's never too late because I just started literally on the bunny hill and watched like a billion YouTube videos.

Paul Sax: [00:15:24] This has been a real fun discussion. I want to just remind everyone. I've been talking to Doctor Patrick Passarelli, and he is both a med peds ID doctor at Dartmouth, and also the ID editor at MedNet, which I encourage you to check out. Thanks so much, Patrick.

Patrick Passarelli: [00:15:39] Thank you.