 Good afternoon. It's another Hot Topics in Global Health. Today, I have the pleasure of introducing Dr. Steven Louvy, who is new to Stanford. He just arrived here as of September. He's our new director of research at the Center for Innovation and Global Health. And he comes to us from Bangladesh, where he spent the last six years. I always get that wrong, Steve. He spent the last eight years in Bangladesh working at ICDDRB, which is the International Center for Diarrhears and Zenith Research. But he was both at ICDDRB and also at CDC, where he was the country director for Bangladesh. So he had a very unique position in Bangladesh and with the CDC. He also, prior to that, spent six years in Pakistan working at the Aga Khan University, where he helped set up the teaching clinical epidemiology unit. So he's had quite a lot of very interesting overseas experience. And while he was in Bangladesh, he participated in a very unusual outbreak of encephalitis. Has anyone seen the movie Contagion? This is the virus that Contagion is built upon. But Steve was one of the outstanding investigators that figured out the pathway of Nipah virus. He's going to share a little bit about this. But I should let you know that Steve's main area of interest is in water and sanitation. And he will be housed both in the center of innovation and global health and also in the infectious disease section. So welcome, Steve, to Stanford. Thank you very much, Michelle. And welcome, all of you, and good afternoon. I think it's great that, although I've really been out of the medical school structure for a long time, that I still see that there are some universals like the attraction of food to residents. So that's encouraging. And I want to talk a little bit about Nipah and how we approached this in Bangladesh and why this might be relevant to people who are working in wards at Stanford. Nipah virus was first identified in a large outbreak in Malaysia, where the first cases were seen in September of 1998. It began up here in the northern part of the country in Ipoh province. But over the course of months, spread farther down south, ultimately, 283 human cases were recognized and 109 deaths. Paul Schwau, a Malaysian virologist, isolated a novel paramedic virus from a patient's CSF in Sungai Nipah village. Hence, the name of the virus is Nipah virus. Now, the clinical features of the illness in Malaysia were symptoms of encephalitis, moderate disease, high fever headache, Malaysia, which often resolved, but at times went on and produced more severe symptoms with disorientation, hallucination, seizures, and coma. I want to mention that this really was an encephalitic presentation. Only 14% of patients mentioned any respiratory difficulties. And those were typically non-productive cough and chest radiographs overwhelmingly were normal. The pathology, when these people died, this showed a diffuse vasculitis. The brain was the most affected organism, particularly the brainstem. But virus was also identified commonly in kidney and lung. So how did the people in Malaysia get Nipah? This is a typical question that we epidemiologists ask. Well, the outbreak was concentrated among pig farmers. Indeed, 92% of cases reported contact with pigs. And compared to controls, people who didn't have Nipah virus, folks who developed Nipah infection were 5.6 times more likely to have close contact with pigs and to have contact with sick pigs. So then we asked the question, well, where did the pigs get the virus? Well, in 1998, there was a newly recognized syndrome in Malaysian pigs called porcine respiratory and encephalitis syndrome. This is pretty much a mild disease, a little bit of fever, some trembling unusual uncoordinated gait. At times, it was severe. And the pigs developed a barking cough. But mortality was low, 1% to 5%. And if you raise pigs, pigs die. It's just like raising chickens. Not everyone makes it to adulthood. So this was not a particularly outstanding event. That was later appreciated that press was caused by the same virus, by Nipah. So then the question was, how did it transmit between pigs and how did it go all over the country? Well, once pigs were recognized to be sick and to be the carrier of this virus and to be spreading the virus, the price of pork plummeted. And so what did people do? They tried to sell their pigs farther and farther away, which was a very efficient way to spread the virus throughout the country. So the next question we asked is OK, so the pigs spread it to the people. How did the pigs get the virus? Well, there were a lot of wild animal studies done trying to figure out what was the reservoir of this virus. And ultimately, the teropus bats, which are large fruit-eating bats. These are not like the insectivorous bats that live in North America. The virus was isolated from a teropus hypomelanus. And other teropus species have also been found to harbor the virus. So the likely transmission cycle then reconstructing it is that in Malaysia, when they were building the pig pens, they started to put fruit trees next to them so that the fruit could drop down. It would give the pigs some shade. And it would also provide them some food. The mangoes in the trees attracted the bats. The bats shed the virus in saliva. Or they'll eat fruit for a while and then spit it out, spit out what they call the spat, which is the seeds. So that would all be contaminated with saliva. The pigs would get it. And once the pigs got it, they transmitted on to people. The outbreak in Malaysia ended with widespread culling of pigs. Almost a million pigs were killed. And this is a picture where they're bulldozing live pigs into a pit. And this was followed by machine gunning by the Malaysian military to kill all of the pigs. Fruit trees were no longer permitted around pig pens. And the pork industry was really decimated. However, since 1999, there have been no further cases of Nipah virus recognized in Malaysia. So that would have been the end of the story, kind of this weird spillover of this bat virus into pigs. But subsequently, Nipah virus was recognized repeatedly in South Asia, and particularly in Bangladesh. So each of these diamonds are places where an outbreak has been identified. And you can see overall, since 2001, there have been 266 cases and 204 deaths. So actually, although many people still think about Nipah as a Malaysian virus, there are more cases and more deaths subsequently recognized outside of Malaysia. There are teropis bats, a little different species. Teropis gigantius in Bangladesh. And they also are infected commonly with Nipah virus. So the question we epidemiologists ask is, how does a bat virus get the people? So to answer that question, I want to go back to a few months after I arrived in Bangladesh. It was January in 2005. We had just gotten, I was new in the position, and there was a new person in the position as head of the government agency responsible for surveillance and outbreak investigation. And he called and said that government health workers had reported that eight previously healthy people from Bashar al-Upa-Zilla had died within a one week period and asking for some advice and some support in mounting an outbreak investigation. So when we mount an outbreak investigation, we think about person, place, and time. We begin by saying, who are the people that are affected? So we sent a team out to Bashar al-Upa-Zilla with a very general case definition saying we don't know what's going on, we don't know how many people are involved. So let's just find everybody with fever since January 1 and make a list and take a look at it. So we did that and identified 124 people with fever. And as we looked at it, we wanted to refine the case definition because there did seem to be a syndrome of particular concern. And that was fever plus either mental status changes or seizures. So there was a group of these people that looked like they were seriously ill. Among a person still living in that same union, that area, with illness onset in January. When we did that, we got down to a list of 12 people mean age of 16 years, a little over half of them were male. And when you review the symptoms that these folks had, 100% had fever by definition, 75% had unconsciousness, 33% seizures. Note that 92% died and that the median time from the first symptom to death was only four days. So we didn't know what we were dealing with but we knew it was bad. So we've talked about person and the next question we ask is place, where was this happening? Well, there's a map of Bangladesh and you can see that Tongail district there is in the center of the map. And Bashar al-Upazilla is just sort of a small place within Tongail, but Bangladesh is the most densely populated country in the world. There are 1,000 people per square kilometer. So we needed to drill down even more closely in terms of place. And so this is a GPS map of the cases. And what you can see is there's remarkable clustering. You know, there are a bunch of cases all here together, a bunch of cases here, and then another case down the road, maybe one that looks like an outlier, but remarkable clustering in place. Similarly, there was remarkable clustering in time. Most of the cases really occurred over a seven day period. Again, one outlier who incidentally is also the geographical outlier. So we had these cases over time and we were able to collect serum from three of them. We couldn't get serum from the other cases because they were dead and buried. So we collected blood from those three, ultimately sent them to Atlanta for diagnostics. And two of the three came back as positive for Nipah. One was negative, but this person had illness on set on the ninth. The serum was collected on the 11th, so they probably didn't even have a time to mountain antibody response. So at this point, we confirmed that this was an outbreak of Nipah in Bangladesh, but we're still asking that question, how does a bat virus get into people and why do we keep seeing this happen in Bangladesh? So we did a risk factor study. We did a case control study. We looked at the exposures among cases, people who had Nipah virus, and compared them to controls. So the cases we've talked about, all 12 people from our initial investigation, and then we took three controls per case. People who lived nearby was a similar age. Now since most everybody had died, we had to use proxies to find out exposures. And so then we did a typical case control analysis, looking at exposures among cases, comparing that to controls. We're very interested in exposure to domestic animals. There was no pig raising in this community, so there was, and we kind of looked down here and you say, well, none of these p-values look very significant. Maybe we should try to increase our sample size. Oh no, we don't want to do that. We've already had 12 deaths. We look for food exposures, and really there was nothing that looked like it was coming out, except for one exposure. And that one exposure was that cases were more likely to report drinking fresh date palm sap than controls, and much more likely than we would expect by chance. So this got us wondering, what is fresh date palm sap? I think for this audience, the best parallel I can draw is to a maple syrup that you collect sap and similarly in winter in Bangladesh, they shave the top part of the tree, and when you shave the part of the tree in the evening, the sap rises during the night and drips out, and then through a bamboo spout and you can collect it into a clay pot. Most of this is then boiled into molasses, which makes great seasonal sweet cakes, but some of it, real early in the morning, is sold fresh, and it's just considered a delicacy. People just love it. They just, it's something they look forward to and enjoy. So if you look at this, you can actually, so we were trying to figure out, well, could bats have access to the date palm sap? So this is an infrared photograph, so we could look at night, and you can see the whole sap collection system and watching that over the course of the night, you can see a teropus gigantic coming in and that is his tongue. That is his tongue, which we know can shed neepa virus licking the sap stream. And we could watch and see how often these bats come in and so contaminate this food source. And although the tongue aisle outbreak was the first outbreak that we implicated date palm sap, it turns out that this was not unusual. As we got more focused on that and started looking and it's in several other outbreaks, and indeed as we look back over all of the experience, date palm sap is the primary pathway of transmission of this virus from bats to people. If the bat, if the virus stopped when it got into people, then they would never have written the movie Contagion and Michelle probably wouldn't have asked me to talk about this virus. But one of the concerns about neepa is it's a class three zoonotic disease. The zoonotic disease is a disease that moves from animal to people. A class three zoonotic disease is a disease that moves into people and occasionally there is person to person transmission. So this is data from an outbreak in Bangladesh in Faridpur in 2004. The index case was from a village in Malikpur and he had drank some date palm sap. He got sick and when he got sick, his mother, his son and his friend and an aunt came to his house and gave him very close care and really nursed him while he progressed to die. And then over the course of the next two weeks, all four of those people who cared for him also developed a similar illness. And one of them, case B, was his aunt. Now it turns out that his aunt was a member of an unusual religious sect where they practiced unusual practices mixing men and women together, smoking marijuana as part of their religious ritual and they were really seen as somewhat outsiders. But when she became ill, the leader of this sect came to spend close time with her as she was dying. And 10 days later, he also came down with a severe illness. And when he came down with this illness, this was really a crisis in the religious community because he was the spiritual leader of this distinct and closely knit community so that the community really descended on him to be with him in his closing days and hours. And subsequently 23 of his followers developed illness. And then those 23 passed them on to yet another generation of cases. So during this outbreak, we saw five generations of NEPA being transmitted person to person. So once again, the question is all right, so now we know it's going from person to person. How is it moving from person to person? So the team did a cohort study where they looked at which exposures were associated with the people who were caring for these folks who got illness and how did that compare with people caring for the folks who didn't get illness. Touching a patient who later died, touching an unconscious patient, touching a patient with respiratory symptoms. So this doesn't sound airborne. Sounds like saliva, washing hands. After contact with patient F was protective. I'm sure you guys all washed hands before you came in and got lunch. So when we looked at the clinical syndrome associated, yeah, these patients in Bangladesh do have a neurological syndrome. Three quarters of them are unconscious. If you look at the survivors, they're still persistent neurologic dysfunction. But there was also a lot more respiratory disease. Remember we said only 6% had very minor non-focal abnormalities on chest x-ray. Here, chest x-rays were clearly abnormal. 62% presented with a cough, 69% with severe respiratory difficulty. And the case fatality was higher in Bangladesh as well. When we look at the strains of virus in Malaysia versus Bangladesh, there are two things that I wanna point out. Number one is the Malaysian strains look very similar to each other. Not surprising. It was probably one spillover event from bats into people. And then all of the people basically got infected with the same strain. It might've been passed through many pigs. Well, in Bangladesh, we see both that these are different from the Malaysian strains. And even within themselves, there's a high degree of variability. So we're seeing variability in the genome of this pathogen. So the folks at the Australian Animal Health Laboratory inoculated Nipah virus collected from Bangladesh with the strain in Malaysia. And they used ferrets because that's a good model for Nipah. And the mean viral RNA levels were 10 times higher in the saliva of the ferrets that were getting the Nipah strain from Bangladesh. Suggesting that there is some difference in the pathogenicity of these strains. In addition, we had anthropologists go in and talk with the family members where a Nipah case occurred, as well as neighboring families, local health practitioners, and trying to understand what is it within the culture that leads to the increased risk of Nipah. And what we learned was that families provide direct care. Even when you're in the hospital, nurses only spend about six or 7% of their time in direct patient care. They're mostly doing other things. So most of the care is provided by family and friends. And this is really rooted in emotional support. And the cultural expectation is to maintain very close contact and provide hands-on care. And there's a real desire, even at the time of death, for close communication. And family members are the ones that are involved in ritual purification of the body, including cleaning the orifices. And all of this, of course, done in a low-income country where 40% of the country is living on less than a dollar a day. So this is not a place where people have gloves or masks or any of that. So one of the intern physicians who was caring for Nipah patients and fired for in 2010 was renowned for the care he gave his patients and for the time he spent with them and for how closely he worked with them and how seriously he took his responsibilities. He managed 14 patients with meningoencephalitis, including three that were confirmed Nipah, and did the usual procedures that interns do all over the world. Physical exam, IV cannulation and nasogastric innovation, characteristic of the conditions in Bangladesh. There were no gloves, no mask, and no hand-washing facility available to this intern. He developed fever and mental status changes with confirmed Nipah infection and died on the seventh day of his illness. So we got a bad problem. We got an ugly problem in Bangladesh. And it's also a concern that we have with an unstable genome and demonstrated capacity for person-to-person transmission that these are the kinds of organisms, these class three zoonoses that are at highest risk of becoming the next pandemic. So how might we stop this? Well, one thing is, could we prevent the bats from spilling over into people in the first place? So one way to do this, and our anthropologists, and we had a lot of blind alleys, but the bottom line is that modifying fishing nets made of bamboo can be used to block access to the shape part of the tree. And this was indigenous knowledge, though not often used. We've done a, so one of the things then we did is a randomized control trial. So I know here at Stanford, you work a lot on human subjects and get randomized control trials of different medical devices. Here, we do randomized control trials of trees. So we randomized our trees to four different types of interventions. And we had intervention trees, and we had control trees. Because we wanted to know of the available materials, could we, in fact, keep the bats out? And trees have to be matched on their height and shaving pattern, because they have different attractiveness to bats. And so you can see the four different types of materials that were used to try to block the bats. And what you can see is that in the control trees, in the control trees, there were 3,556 bat visits where the bat contacted the sap. So just like our outbreak investigations showed, if it's unprotected, bats are frequently going to visit. By contrast, almost never did we see bats contacting sap, except in one of the jute nets had a little problem and the bat was able to wiggle its head in. But in general, this was physically effective at keeping the bats out. But it's one thing to have something that works well. I mean, if you will, this is sort of like a condom for a tree. If the tree's wearing it, you can prevent infection. But how do we get people to actually use the approach? So we did a bunch of piloting and have had some mixed experience, but some interest in this. And so now, right now as we speak, we're training field workers for a district-level prevention trial. And our objective is twofold. One is to encourage people not to drink fresh date palm sap. We know it tastes great, but it's really best not to drink it because of the risk. And secondly, if you insist on drinking sap, then only drink it from a skirt-protected tree. So the intervention is in 550 villages. And we're using sophisticated behavior change intervention with real behavior change experts. We've got a 45-second television spot, video documentary posters, and a trained local non-government organization group that can actually go in and talk with villagers. Our baseline survey to look at sap collection practices. Last season, lots of process evaluation to see if all the things went out that we are trying to get out, and then an endline survey. And comparing both an intervention area and a control area. So that's sort of the work to try to prevent the spillover. But then we're also concerned that sometimes it's going to spill over, or if it's not NEPA, it's going to be some other saliva-transmitted virus. And there are lots of saliva-transmitted viruses to concern. So there are hospitals, of course, in Bangladesh, but they're a little different than the Stanford hospitals. So Nadia Ali Rimi, an anthropologist, helped us look at three public tertiary hospitals in Bangladesh. So she looked at three pediatric wards and three medicine wards, so one in each hospital. And five anthropologists went and conducted 48 hours of structured observation to just say what's going on in an objective way. And things were crowded. For each one pediatric bed, there was 1.8 patients. So frequently, there were two patients per bed. If you looked and measured the area of the ward during the daytime, there were 4.3 people per 100 square feet in the ward. And if you counted the number of uncovered coughs and sneezes, you know, not the coughing and sneezing, occurred at the rate of almost five times per 100 square foot per hour. In terms of hand washing, there were 16 hand washing stations, but only seven of them had soap. And these were exclusively at the hand washing stations that were reserved for the senior physician. And nobody else was allowed to use them. The patients in the attendants, the ones who were providing all the hands-on care, none of the 12 had soap, and only two of them had running water. So the anthropologists counted 2,709 hand washing opportunities. That is saliva contact, or fecal contact. Now what are you going to do? Only 32 times did they observe hand washing with soap. And they also found that none of the people in the wards could mention was aware of any infection control policy. So this is the environment that we're trying to work to reduce person-to-person transmission within. So we've tried some approaches. One is to say, look, there are a lot of barriers. Can we give people soap and encourage them to wash their hands, telling them before you eat, after you feed the patient, or after you clean the patient, keep the patient's food and your food separate? This is really common, eating the patient's leftovers. This is a poor country. 40% of children are stunted. Male nutrition is widespread. You wouldn't waste food. And so it's a little countercultural saying, no, don't. Don't eat it. Store the patient's food separately. Try to sleep on a separate bed, everything so crowded that people will often sleep with the patient they're with. If you're going to sleep, sleep back to back, rather than face to face. Or we even had them try to sleep head to heel. Again, a little countercultural. And when you comfort and hug the patient, put your head in the chest, try not to go to the face. And trying to keep your face less than one hand's distance. So as we roll this out, what we found is that some of these, there are real barriers to implementing these interventions. One of the big ones is that if there's no hospital infrastructure to wash hands, it's very difficult. You can give behavior change messages, but behavior change messages are not magic. You need to have infrastructure institutional support in order to implement it. Telling people to stay one hand's distance away from the patient's face. People have just told us, we cannot do that. This is, as one woman said so poignantly, even if it causes me to die, I cannot be separated from my son during his illness. So on the other hand, the alternative sleeping arrangements and splitting up the food, that seemed reasonable to folks. So we're doing some piloting. And we've got some funding now also trying to improve hospital infrastructure to improve hand washing. So both low cost mixtures of alcohol and glycerin and actually something that is lower cost still. The picture is not great, but this is a hand washing station that has water. And instead of using soap, you can buy local detergent and mix a sachet, which costs about $0.02 of powdered detergent into a, it's a little bigger than this, but into like a liter or liter and a half mineral water bottle, put a hole in the top. And then that is a poor man's soap. And we've done some work showing that that is as effective as standard bar soap in terms of removing indicators of bacterial contamination on hands. So inexpensive in trying to see if we can work within this constrained environment to improve it. So although Michelle says that I have interests in lots of other areas, we continue to be really interested in trying to reduce transmission of NEPA and other saliva transmitted infections. We think that's really important for the people of Bangladesh. It's also important globally because if and be at NEPA or be at new strains of influenza, strains that are capable of person-to-person transmission is the sustained chain that actually puts the globe at risk and hospitals where you're bringing a bunch of people into this dense environment actually risks being a place of worsening, of increasing the risk of transmission. And I would close by saying that the work in Bangladesh is the work of a very large group of people who have contributed. I would mention particularly the role of the government of Bangladesh and their active interest in collaboration and best practices and support from many individuals who have given much of their time to putting this together. So thank you. Any questions? Have you been messaging this virus now like a household word? No, no. If you go into, I mean, the literacy rate in the villages on the order of 40% to 50%. And if you go and you talk to them about NEPA virus, most of them will not have heard about it. Well, it's been in the newspaper on the radio, but that doesn't necessarily have much penetration in their day-to-day lives. If there has been an outbreak in their village or in the neighboring village, yes, then they know about it. But the attempts by government to talk about this have been, yeah, they'll hold a press conference and they'll say a few things. And the press gets excited when there is an outbreak. But in terms of really trying to push a message out to a group that's not that easy to get to because they're not, they don't have high levels of access to media. No, there's low level of understanding. And then you talk to people and you say, well, why do you think people got NEPA? Most of them do not subscribe to the biomedical paradigm. So you're saying, OK, what are you telling me? There's something that I can't see that's invisible that infects sap and it's going to kill me. You're crazy. I've been drinking sap. My family's been drinking sap for generations. You're nuts. And then when there is an outbreak, the most common response, the most common local explanation is that this was what they call asmanibala, a curse from Allah, a curse from God on high. And particularly, you think about the Faridpur outbreak where it was this religious sect that was so decimated. The local explanations have nothing to do with communicable disease. So we go in and we want to talk about hand washing and viruses and all. We're just going right past each other. So it's actually a lot of work to try to. And recognizing, in this room, I assume that most of you engage in the biomedical paradigm so I could present it from that perspective. But understand that that's not universal. I think this very large public health campaign that may be an opportunity behavior, has there been a cost-effectiveness analysis integrated into that or a consideration of other approaches that might be important? Yeah. About the hand washing. Not so much the hand washing, but with the NACSO group. Yeah. As we were certainly collecting information on cost as part of the district level intervention, because there are a lot of upfront costs in terms of trying to figure out this messaging and trying to figure out how do you connect with people of a radically different world view. And those I sort of view as sunk costs, so you've got to sort. But if we can get the messaging right, then the cost per person of leveraging up is not so much. My own view is, as I look into this, my perspective on this is that it is difficult to do this on NEPA alone, because when you think about all of the kids, there are many more rabies deaths in Bangladesh than there are NEPA. Even in that area in the northwest part of the country where it occurs, it's fewer than 3% of all cases of encephalitis. So on its own, it's pretty expensive. But I think, again, we have to think about, are we only trying to prevent cases in Bangladesh? Or is there some global interest in interrupting transmission of stage three zoonotic viruses, i.e. what HIV was five decades ago? So I think we totally have to think about cost effectiveness in a low income setting. I think we're still really iterating around trying to get the intervention right. With the saliva-transmissible disease, it would also be appropriate for other types of transmission. This particular outbreak was like the saliva-transmissible. Do you think another zoonotic, or stage three level three, where it is communicable to humans would be droplet or airborne transmission and have other interventions that we could start to lay in place now that will prevent droplet transmission from occurring in such high risk settings as those hospitals might change? Yeah, I think it would be great to be able to reduce droplet infection as well. We have one setting, one of the outbreaks, which is suggestive of droplet transmission, but it's not conclusive. As I sort of look at it, and of course we have high amounts of multidrug resistant tuberculosis in Bangladesh, and those guys are in that same crowded ward as well. So is droplet transmission also a concern? Yes. Now, the way funding is so fragmented, we do have a little bit of money working on some TB issues as well. And at least with TB, that folks have been thinking about tuberculosis and low income settings for a long time. And so there actually are some standards you can think about and some issues in terms of ventilation, some things you can do. I think one of the reasons why we're focused on saliva transmission is because we also have a huge problem with influenza, that H5N1 influenza is endemic among poultry in Bangladesh. It is there all the time, has been since 2007. When people get H5N1 influenza, the case fatality rate among reported cases is 50%. The real case fatality might not be that high, but it's still going to be high. So one of the concerns is not only is that both NEPA and avian influenza are class 3 zoonotic diseases that are saliva transmitted. And we feel that there's little attention around saliva. It's not as exciting as airborne, I suppose. So that has been a primary focus of ours. But I totally agree that there is a need to address other issues as well. Any other questions? I will say one last thing is if anybody is interested in engaging around any of these issues, please let me know. We could really, the Bangladesh team could really use some help moving some funded proposal ideas into protocol format. So if there are people who are good writers and who are interested, I'd be happy to hear from you. The preceding program is copyrighted by the Board of Trustees of the Leland Stanford Junior University. Please visit us at med.stanford.edu.