Human Mobility & HIV Workshop - May 25, 2021

Transcript

Proceedings

Welcome - Holly Campbell-Rosen, Ph.D.

Opening Talks – Lessons Learned From Humanitarian Settings
Moderator: Sheri Weiser, M.D., M.P.H.

Mental Health Lessons from Displaced Persons in Humanitarian Settings - Mark van Ommeren, Ph.D.

COVID-19 in humanitarian settings: lessons learned from the AIDS pandemic - Paul Spiegel, M.D., M.P.H.

Q&A  

Session 4: Implications for Interventions
Moderator: Carol Camlin, Ph.D., M.P.H.

Climate-induced migration and food insecurity as pathways towards worse HIV health - Sheri Weiser, M.D., M.P.H.

Harnessing social networks to improve HIV prevention and care cascade outcomes in mobile men - Harsha Thirumurthy, Ph.D.

Optimizing refugee-specific interventions to improve care - Kelli O’Laughlin, M.D., M.P.H.

Interventions to improve HIV care among mobile populations - Kate Clouse, Ph.D., M.P.H.

Human Mobility and HIV Infection – What can we learn from population-based surveys? - Andrea Low, M.D., Ph.D.

Q&A      

Session 5: Panel Discussion – Highlights and Future Directions
Moderator: Holly Campbell-Rosen, Ph.D.
Panelists:
Sally Blower, Ph.D.
Carol Camlin, Ph.D.
Bernal Cruz, M.S.W.
Paul Spiegel, M.D.
Fred Ssewamala, Ph.D., M.S.W.
Frank Tanser, Ph.D., M.Sc., FRGS
Maark van Ommeren, Ph.D.
Sheri Weiser, M.D., M.P.H.

Agenda Item: Welcome

HOLLY CAMPBELL:   Hello. Welcome to today's webinar. I am Holly Campbell, the Program Officer of the HIV Research Program within the Division of Research at the National Institute of Mental Health. We are happy you joined us today. We are very excited about this workshop and its potential to stimulate the field and talk about prevention, treatment, and context of mobility.

(Housekeeping remarks)

Now I have the pleasure of introducing Dr. Sheri Weiser, our moderator for today’s opening. Dr. Weiser is a Professor of Medicine and internist at the Division of HIV infectious Diseases and Global Medicine at the Zuckerberg San Francisco General Hospital at UCSF. Her research focuses on the impact of food insecurity, and other social and structural factors on treatment outcomes for HIV and other chronic diseases. We will have the pleasure to hear about her research in the next session. Please, take it away.

Opening Talks: Lessons Learned from Humanitarian Settings

SHERI WEISER:  Thank you so much, Holly.  Welcome to this session on Lessons Learned from Humanitarian Settings. We have a great program for you this morning. We are going to start from hearing from Mark van Ommeren, with a talk entitled, Mental Health Lesson from Displaced Person in Humanitarian Settings. Dr. van Ommeren is head of the Mental health Unit with the WHO Department of Mental Health and Substance Use. Before his current position, his work focused on developing interagency mental health policy for humanitarian settings with linked implementation tools, and also developing testing and disseminating a range of scalable psychological interventions. And, building back better mental health services across different levels of the health system after major emergencies.

Next, Paul Spiegel will give us a talk entitled 'COVID-19 in Humanitarian Settings: Lessons Learned from the AIDS Pandemic. Dr. Spiegel is the director of the John Hopkins Center for Humanitarian Health. He is a professor in the Department of International Health at the John Hopkins Bloomberg School of Public Health. He is internationally recognized for his research on preventing and responding to humanitarian emergencies. Having published over 100 peer-reviewed articles on these topics. He was formerly deputy director and chief of public health at the United Nations High Commissioner for Refugee and he served as a Commissioner on the Lancet Commission for Migration Health,  the Lancet Commission on Syria, and currently co-chair of the Lancet Migration. Welcome. I will turn it over to you, Dr. van Ommeren.

Agenda Item: Mental Health Lessons from Displaced Persons in Humanitarian Settings

MAARK VAN OMMEREN:  Thank you so much for inviting me Holly, and thank you for introducing me, Sheri. It is a pleasure to speak with Paul, and you will be seeing him later today Thank you so much.

There are different components to my talk. First we will talk about policy, and priority settings and some research samples.

Right now, the highest number of people affected by emergencies, whatever statistics you look at, the highest number of people displaced since World War II. This is a time where the topic of forced migration is relevant. I will be speaking about forced migration, rather than migration, in general. I was of course, talking by humanitarian settings, but I will always refer to migration in all aspects.

In terms of needs, we have to - I think it is very important to mental health very often think of people who have gone through horrible things. They only think of the horrible things, rather than thinking that people usually also already had a mental health good, or not so good, mental health, before these things happened.  The same is true for forced migrants, and for people affected by emergencies. People come with a whole range of mental health and social problems, that societies tend to have, and this adversity tends to worsen many of them. Often, the response can cause some further problems. I think the same is true for migration, as shown on the right of this slide.

One of the most common questions we get is,  how many people are affected? And that is a difficult question. There are lots of studies that have been done. There is an enormous heterogeneity in them, and those of you who do systematic reviews, often an I square of 99 percent different substrata in the systematic review data.

These are the best estimates we could come up with. We estimated together with IHME in Seattle, and these are the current estimates. That one in 20 people affected in the world affected by conflict, whether severe mental disorder, and if you go down to moderate or mild, these numbers grow quickly grow quickly to 22 percent of the population. These are enormous numbers if you are thinking that you are going to be helping all those people.

So indeed, that is the question. The question of public health, the decision-makers gets is,  what do we do for mental health for hundreds of thousands of people? Obviously, some priorities need to be set, but what are the priorities?  People look at those questions in different ways, some people say we should help everyone with severe distress. People at risk of disorders, everybody with disorders, severe mental disorders, everybody with PTSD, everybody who seeks help. There are many different ways of looking at that question.

One thing I can say is that respected colleagues have used, which are among the following, focusing on well-being only, not focusing on mental disorders. Prioritizing PTSD over other mental disorders. Excluding psychosis. Or, spending all resources for maximum clinical affective care for a few people. So known of these are WHO’s views, but they are views that I have people that are highly respect, that do think this way.

There are a bunch of policies out there, and there is one coming out hopefully in the next year, minimum surface package for this area of work. I will quickly summarize one set of guidance, just to set the context for the talk, not to go into detail. This is the 'Sphere Handbook' and Mental health Standard, which has nine actions. One starting with coordination, and without coordination in this area, you simply, will have a lot of lack of equity, because some actions will be wildly available and others none at all.

The second one is situational analysis. The third one is strength in community self-help and social support. This is really consensus-based recommendations. There is not strong data on that, may be a future research agenda for that. There is the psychological first aid, again, there is not so much data for that, but a very strong consensus that should be made available.

Basic clinical care to be available at healthcare facilities. There is good data for that, and we have a manage gap program, as very clear clinical protocols. That is at the clinic, and many people don’t go to the clinic and there are many other places that people can get or should be able to get mental health care. That should be done through making psychological interventions available for people in a prolonged distress, and I will discuss that further in the research part of my talk.

There is protecting the rights of people with severe mental health conditions in the community, hospitals and institutions. From a migration perspective, that is very interesting to see whether what happens to people with severe conditions, whether they migrate, or whether they stay behind. There is also some very interesting data in the migration data that migrants,  when they move, that migrants in new countries have a  substantial higher rate of mental  health problems, particularly in the Caribbean - Caribbean countries, living in the UK having more psychosis. There is a lot of research on that.

And the eighth public health action is to minimize harm related to alcohol and drugs. I will not go into detail on that, but, that is also one of them. That is of course relevant to many people listening to the stock, because when you work in HIV, it goes without saying, it is important. And Paul, who is speaking after me, he will remember we worked together on this, when he was at UNHR.

The next one is 'Building Back Better'. We are quite proud, we have this publication. If you Google Building Back Better term in Geneva, you get this publication. But I know if you do it in the U.S., you get the White House because Building Back Better is a slogan for the current U.S. Administration.

All of this is nice, but we have some problems. The first problem is that there is not enough, well allocated resources to implement what I just said. So, rather than having your beautiful mind things implemented everywhere, you have a scattered plot of various actions that - sometimes when I talk to colleagues in mental health, it is like we are having a vaccination campaign in the north, and primary health care in the south, but not the other way around. Some of these actions I have outlined, and some places in some communities, and not in others.  It is very difficult equally distributed.

Second things is that in all these responses, overall, we pretty much miss most people with emotional disorders. And by that, I mean depression, anxiety, and PTSD. They are not very visible, and many people have them. So, there is a lot to be done.

The way forward would be through of course, priority settings. Giving our resources and doing research. What priority should be there? I propose that the first priority is really human rights and vertical equity. We must be sure that psychosis, severe depression, severe conditions, that there is an answer for these problems within services made available in humanitarian settings of forced migrants and refugees.

Secondly, beyond that, we need to make available care for people with emotional disorders, but not just any care. It needs to be scalable. So it needs to be feasible, affordable and cost-effective. That is where the research challenge is.

Thankfully, there is a growing evidence base that there is a lot of psychological interventions and that these interventions not only work in high income countries but also in low and middle income countries and you do not need specialists to deliver them effectively. And that they also work for severe emotional problems. For example, for severe depression.

In interventions, that we know that work, can be modified through innovative delivery, information technology, through making them shorter, by using interventions in a trans diagnostic way for multiple problems at the same time, and very much focusing skills for self-management. That is how interventions can be scaled if you keep the core focus on a self-management these interventions seem to work.

At WHO, we are not certainly the only ones working on this area. Many others have gone before us, and will go after us, but we very much focus on developing or republishing those who have already developed it, these interventions. Very much focus in making sure that these interventions are open access. There might be quite a bit of work in this area already, but it is not open access, so it is difficult for agencies in the field to use them. There is always the question, are these interventions enough? These simple interventions. The answer is no. How did we do the research around these different interventions? We used the five-step phase protocol. Which, pretty much, follows many different trials, but it usually looks like this.

The first one is the focus phase. And then we do the small feasibility RCT that is not at all looking at whether they work or not, but just testing the RCT. Can we run a clean RCT? Can we get it all done with high quality? And then we do a process of evaluation. Again, it does not focus whether intervention works or not. But very much focused on whether the research procedures, are they feasible and acceptable, and are they working?

Then we do a large, this is maybe an overstatement, but a large state-of-the-art RCP, lots of people. And then we do the process evaluation again, and this time the process evaluation does not focus on the research but  focuses much more on whether what people think of the intervention and how it can be further scaled up.

We have done this, and we started this in 2013. We did not realize it that this was an area that would draw so much interest. When we went to funders, we also managed to get funding quicker than anticipated. Within the last six or seven years, 10 large trials, and some of them will be published nicely, JAMA, Lancet, PloS, and many of them are still under review. They have been done on different interventions, on adults, adolescents, face-to-face, on E, on prevention and treatments. Generally, we have had a positive experience. Most of the research, where it is done, you can see it on the map. Not in Americas. We can discuss why that is, but the reality is the funding, we got was for these places. Of course, you see Europe there, this was one large multisite trial that was done in Europe with refugees.

I am going to give examples of these trials. I have three examples. The first one is done in Lebanon. It is a guided digital health intervention. It is unpublished. Let me share it with you. I've never presented this before. It is done by a team of Lebanese Europeans. It was funded by Elhra in the UK, which is funded by (indiscernible).  And there was also some European Union funding that helped.

So, the research question was whether this step-by-step, this digital intervention, whether it was more effective than enhance usual care in reducing depression and improving impairment.

It was done last year, which was a messy year to do a trial in Lebanon, because the economy was collapsing, there was political turmoil, there was still this ongoing refugee crisis of 1.5 million displaced Syrians in Lebanon. Lebanon has 6 million people, so it is another 1.5 Syrians. It is hard to imagine, it is like having America suddenly having 75 million refugees, if you look at the proportions of refugees there are in Lebanon.

During the same time, during mid-crisis, there was an enormous explosion in Beirut, which destroyed a large part of the city, and the COVID-19 pandemic was quite a challenge. However, for E-Mental Health, COVID-19 pandemic was actually helpful because this was the one-way people can still get help through the internet. The trial was done because the Lebanese authorities required it, both among the Syrians and Lebanese. In fact, we powered it so much that we had a full trial for Syrian displaced people and a full trial of Lebanese people. Later I will show the number of people in the trial, but there are over 1,000 people, 500 Syrians and 500 Lebanese. We analyzed separately and together the data.

It is like an E-learning invention, to learn one psychological technique of behavioral activation, which is known to be in evidence-based technique of reducing depression. It was administered through five sessions of people, e-learning through story. One thing about E-mental health is E-mental health interventions, they work well, but it works better when they are guided. Meaning that when the participant gets some support from a human being, who encourages the person to use the materials. You can imagine yourself,  if you by a self-help book in a store, you might stop reading it after half a chapter. But if somebody out there who you have contact that encourages you to read once a week a chapter, you are much more likely to complete it and do have benefited from it.  In here, part of the trial is we had these non-specialists, laypeople, who were in 15 minutes maximum per week in contact with the participants through text messaging mainly or phone calls, to guide the person to use the program, or get some technical advice or how to use the IT.

This trial, it is actually two trials, one with Syrians and on with Lebanese. It was for adults. They were all suffering from moderate to severe depression. They also had impaired functioning. There was a bunch of secondary outcome measures beyond depression and functioning, which was primary outcome. We had very nice results. This is one of the trials that I feel very confident about the results. I don’t know how many of you are trialist, when you are among trialist, and trialist review other trialist paper, they always start saying, where is the bias? Where could bias have slipped in? Maybe an RCT but it can still be biased. I think it would be very hard to have bias in this trial because we recruited online through Facebook. We do not recruit the people that we thought could be helped. That would be especially likely to respond to the trial. We did particularly exclude people that we felt, that person, lets not refer to the trial because that person may not get there. There was no bias in the recruitment. Through Facebook, they found out about the trial, they joined the trial, and then there was an automatic randomization happening. Again, there is no way for us to have bias in there. And then, the data collection was also automated. Because after different points of time, people were asked to fill out questionnaires. Again, there's no human being who can do hint, hint, and ask people to please, indirectly without saying in so many words, kind of suggest it would be nice if they gave positive answers.

It is single-line, not double-blinds. There will be limitations.  Feel very confident about the results in this trial. It was challenging. One of the big challenges was drop out. A lot of people drop out. Which is very common for E–mental health. It is also very common for research in refugee groups. I would say for migrant, and it must be true for most work, of course, research with migrants, if they are on the move, the likelihood that you can follow them up is an issue.

Another point that I would like to emphasize among challenges is the e- helpers, we like to work with laypeople, it is less expensive than fully licensed clinicians. But, of course, there is an issue about how to get paid, how to get retained, these are issues. And then another big challenge is risk management. Obviously, any trial has to go through ethics and any ethical Review Board would want to know what to do with people who are suicidal. Really, this trial, through the internet, could be anybody, anywhere in the country could sign up for a trial.  If they are suicidal, leave their protocols in place.

We managed this, we worked with international hotlines for suicide risks. We also have separate budgets to deal with crisis management so that people can get follow-up care in case they were suicidal. This did not become an issue, but is important.

This is one of the largest RCTs ever done in mental health in a low income country.  We found that with simple – the beginning of recruitment was slow, but with more advertisement over Facebook, so quickly, we recruited people. That was showing the potential scalability and it can show safety delivered in a public health crisis, and we found improved outcomes.

Now, I will move to another trial. This is a trial in Turkey, also Syrians. All three trials covered the areas with Syrians. This trial is with a group of Turkish and Europeans, a lot of Europeans, because we did two trials. One trials we did in Turkey, one time across Europe. I will focus here on the turkey trial, funded by the EC.  The question is intervention more effective than enhanced usual care in preventing the onset of mental disorders among distressed Syrian refugees in Turkey?

We did earlier is a systematic review, a published review, that is about prevention, interventions that prevent the onset of mental disorders among refugees or people affected by humanitarian crisis. If we had a workshop, I would ask you to guess how many trials we found. Since it is not a workshop, I will tell you, we found zero. Basically, there have been zero trials out there that show in a randomized way, that one can prevent the onset of mental disorders given intervention, among distressed refugees. Even outside, even if one looks in the US or Europe, Australia, there is not that many trials out there. There's about 60 right now. Trials out there that looked at the prevention of mental disorders among people who are already distressed. Of course, there are hundreds of articles written about prevention, but most of the research of prevention doesn’t exclude people – they call it prevention but it is not prevention, if you define prevention as a reduction of incidence. This study was done in Turkey, in Istanbul, and Mardin. We use intervention, which is highly innovative. It is also a cycle educational course, but it is recorded. People go in a room and they listen to the course, but it is not very passive. People are doing exercises while listening to this course. It is accompanied by a self-help, illustrated book, doing what matters in times of stress. You can find that book on our website, in many languages. And this intervention can be delivered by very quickly trained facilitators. It is all automated, pre-recording. So it is not so complicated to deliver it. You can fill up a room with people. It is not like group counselling, where you say 10 people is maybe a lot, but here, we did it with groups up to 30 people. This trial, is for refugees, and they had to have psychological distress, so they had to score above a minimum score on a measure. But, they are also all assessed if they had a mental disorder or not. If they had a mental disorder, they got referred to care right away, and they were outside of the trial. We really wanted to see whether we could actually prevent the onset of mental disorders, to reduce the incidence. This was a good group to test it on because you would expect quite a bit of incidents in this population that is so highly distressed as this population. We were supposed to do t12 months follow-up, which we did not, because of the COVID-19 crisis, but we were able to do a six-month follow-up.

Our primary outcome we did have a positive effect, which is under review, so I cannot tell you the details of that, but it was not all positive. There were some nonsignificant findings. But the fidelities of course were huge, because everything was prerecorded.  

We mentioned some of the challenges related to COVID, but actually one of the biggest challenges, which I should have put in this slide, is the complexity that if you want to do a trial like that, that you are going to have to assess people not only for having the stress, but also for not having a  mental disorder, and then organizing it. That is a challenge for prevention. For pure, intervention trials.

This is the first prevention RCT among refugees experiencing psychological distress without a mental disorder in low to middle income country. We do think it has potential because you can have larger groups, at the same time. A very light intervention.

There is one more trial. This is an Early Adolescent Skills for Emotions. Also done among Syrians, but not in a Syrian country, but in Jordan. This is also under review, this trial. This was also funded by the EU. The research question is again a treatment question, is intervention more effective than usual care in young adolescents.

This is Jordan, which is south Syria. The recruitment was through door-to-door visits, in a month. This is a challenging area because for young adolescents the literature is not so clear yet.  That you can do this modification of interventions, to make it less intense. For adults, we know we can modify it, having laypeople do it, having fewer sessions, making these modifications. For adolescents, we know much less, it is a challenging age groups, 10 to 14.  Anybody who is a parent, knows there are a lot of difficulties in this age group and not so easy to always easy to address.  From this intervention, I was less confident that it would work. The intervention, it is seven group sessions, and three sessions with the carers. The adolescent sessions focused on emotional regulation and problem management, and the caregiver sessions, with coping skills and positive parenting.

This is from Syrian refugees aged 10 to 14, high psychological distress, and living with a caregiver with consent. We were looking for both internalizing and externalizing symptoms as outcomes, which I think is a mistake, because we developed interventions for internalizing, so then to make externalizing a primary outcome,  was a mistake, as we also found  in the findings.  Because in the findings we found it worked for internalizing, it did not work for externalizing, which in hindsight, is no surprise.

So, again, we had a mixed range of findings here. Some very important, positive ones on internalizing, internalizing is a term that mental health professionals use for emotional symptoms like depression and anxiety.

So there were some challenges. The caregivers were almost all female, the fathers did not participate. The scheduling was complex, and the whole intervention – any trial with children get quickly complex. If you have children and if there are carers, and they both get an intervention and it needs to be randomized and stay blinded, the carers also need to give consent for the adolescents. These are logistically very complex trials. But, we found some positive effects. Which is great since there is such a crying out for interventions like this for young adolescents.  So, we think it has the potential for use, and for scale up. If we get a replication, we will publish it. It is actually what we do for all of this work. We always say, one trial is not enough to release it, two trials is enough for WHO to have enough confidence to put it online. We know the FDA also uses a standard of two trials in its considerations for recommending treatments. So it is not a complete arbitrary number and it was delivered by non-specialists, again, not by psychologists and psychiatrists. There are very few of them in their countries.

Overall, the lessons learned, because that is of course, what NIH would find particularly important. I would say the biggest lesson is the – saying this is from a high income country view for a second. Sitting in a high income country wants to do this in a low income country, it is really important to have a very strong local partner, in my case, I was very  lucky because my first trial I did through Pakistan, and the local partner knew more much about trials than I ever did. It taught me a lot. In any case, having a local partner who has experience, or who is really ready to do a quality RCT is really important. One cannot do that unless one spends a lot of  time on the ground, legal challenges can be big  obstacles. In Lebanon, simply, also need to study the local population and also recruiting. Of course, inherent in places where there is a lot of adversity, that there further unexpected adversities.  There is an expectation in all this work, that it cannot spend months of training people to do the intervention, because it will never be scalable. It needs to be a light intervention.

Cultural adaptation is an obvious requirement. And I would it is very important in the trials, to set up the care for people whom the trial was not enough, or the intervention was not enough, or people with severe adverse events, that is not always easy to make sure that these services are available. A little is easier in countries in the three examples I gave, because Lebanon, Jordan, and Turkey have mental health systems. But there are other countries where it is more complex.

Where to go from here? I would say that the issue is still that we have a sense of what works. We have interventions that work. But, we are not getting them at scale. We need to find ways to do that. We need to have our arguments. We need to have cost-effectiveness data, and we have for some of this work we have cost-effectiveness, but certainly not for all. We need to cost them and we need to all the time, work on sustainability. None of these interventions work in a vacuum, in systems. Of course, with migrants, if migrants are on the move, I have no doubt you are discussing this in these two days in this webinar, research gets complicated because you may lose your subjects and how do you follow up in case there is a need to follow up. Overall, there is a huge call, for any intervention science, to make sure that what we know works and that we also know how to implement it so it is more likely to be implemented. And with that, I thank you.

Agenda Item: COVID-19 in humanitarian settings: lessons learned from the AIDS pandemic

PAUL SPIEGEL:  So, I will have start right in, if I may. Let me share my screen.  Hello everyone and thank you very much for the opportunity to present. It is great to hear Mark, and to see how much the field has progressed, often, thanks to him and his colleagues. We are always trying to have more evidence-based interventions in humanitarian field. It is often overlooked. As Mark clearly mentioned, it is extremely difficult to undertake research in this field.

I will talk about COVID-19 and some of the lessons learned from HIV/AIDS during that time. I'm actually excited to present this for the first time. It is the first time that I have tried to connect the two. So I am looking for to any comments that you may have.

As a little bit of background, what I am going to do is compare some of the work that my colleagues and I did around 2004, 2005 and 2006, and then throughout the 2000's, looking at HIV, particularly on refugees, because I was working with the UNHCR at that time. At that time, like currently with COVID, there was very little information about how HIV and now COVID affects migrants, what makes, if anything, what makes them different? Yet, never a lack of suppositions and in trying to figure out what may occur.

I am going to talk very briefly about the magnitude of the displacement.  Mark has one similar slide, I am going to make the comparison. And then I will talk about the transmission of HIV and COVID, what we know, and some of the suppositions early on that we made for both. And then I would like to move into human rights and equity components, looking at those three areas.

Just to go back, this was something that came out in September 22 of last year in John Hopkins Magazine, some of the work that we have been doing. I'm going to highlight this in the next slide because I think it is prescient. John Hopkins scientists mobilized early during the pandemic to understand the threat COVID-19 posed to refugees and help inform Bangladeshis response. Their models have led to more questions than answers. That was September 22, I would say that is still the case, it is incredibly changing situation of which I hope to discuss with you.

Making comparisons between 2006 and now 2021,I went to 2006 to look at the global trends, and there were 32.9 million forcibly displaced persons then according to UNHCR. That means refugees, asylum-seekers, internally displaced persons., Stateless.

Mark showed this one, 79.5 were waiting any day now. It usually comes out in June, for the next year, in the end of 2020, but we already know it will over 82 million likely, with what has been happening, unfortunately.

Just as a comparison, if you look at the end of 2006, we had 32.9 million forcibly displaced persons. At the end of 2019, 79.5. It increased 46.6, 3.6 million persons per year being displaced. And if you take that group of 79.5 million or 80 million, it is the 20th largest country in the world, larger than the UK and Italy, not combined, but separately. We are talking a massive amount of people. That is just displaced, there are many people that are affected by conflict, persecutions, who are not displaced. Unfortunately, we do not have a very good number. Were talking about a lot of people.

Now, let us talk about transmission of COVID-19. This was just this past Thursday, looking at the airstrikes in Gaza. Again, talks about maybe they will turn into a super spreader centers. We have been hypothesizing a lot of different aspects in terms of COVID and what may happen. But we are not necessarily been seeing a lot of what we have been hypothesizing about.

One of the issues that we have looked into a long time ago was that there was a "myth" of wars will spread HIV, and therefore people affected by conflict will have significantly higher HIV prevalence, and with movement, they will spread HIV.

Looking back, there was this article in 1991 by Smallman-Raynor, who talked about Civil War and the spreads of AIDS in central Africa, hypothesizing some aspects of what could happen due to war and HIV. Much later, now, there was Allen, who looked at the displacement in IDPs northern Uganda, and they said that the 18 year war between Uganda government and the LRA has increased HIV rates in parts of the north, to 11.9 percent, which is really double the rest of the country. That is true, but I will show you the next slide the problem with that.

Then that became world vision talks about ponds of politics in northern Uganda. So the hypothesis, or in this case, just stated, that HIV in conflict are going to spread. But we have looked, and this is just one of many examples of data being not well considered. This is a paper that was published in Lancet, a long time ago, in 2007, but we compared here Gulu in northern Uganda, and we compared this to the national, which is here. National HIV. This is Gulu.

In theory, correct, if you look at 2002, Gulu is double what is happening in the national. Gulu, which was in conflict and displacement the entire time, went down, considerably from 1993 and it followed, not dissimilarly, what was happening nationally. One example if you are just looking at a point prevalence, you may make some mistakes.

Then there was Nancy Mock, from Tulane, she did a simple overlay of looking at how to measure conflict and looking at HIV prevalence. You can clearly see here – and it is not causative, clearly, but where we have a lot of war and where we had at that relatively lower HIV prevalence.

And then we had Strand. This I also found quite interesting – a little bit more sophisticated, where he  looked at HIV prevalence on the y-axis, and on the X axis he looked at the level of conflict, using the (indiscernible) database. And it again, you see a line clearly that once you have the higher the level of armed conflict, the lower level of HIV prevalence. Again, not causative, by any means, but insinuate more ecological.

So, what we did is we started in the early 2000's, we started to measure HIV prevalence amongst refugees. It is an ugly slide, I apologize. Here are refugees in various countries, mostly in camps, from Burundian refugees in Tanzania. And then what we did was we looked at the host populations and we looked at the comparing primarily antenatal center surveillance. There were some, but I have not marked them here. Some of them were actually population-based surveys. Most was antenatal surveillance.

What we found consistently was refugees had a similar or lower prevalence than the surrounding host communities. This was extremely helpful for a stigma and discrimination point of view within UNHCR. It allowed us, amongst other reasons, to eventually we became the 10th cosponsor of UNHCR in 2004.

What are some of these risk factors and what are some of the reasons behind this? I first wanted to say clearly, that this does not mean that conflict does not have an effect on HIV, nor does migration have an effect on HIV, of course it does. To think about this to say that conflict will increase HIV and migration will always increase HIV, is incorrect. One needs to look at the HIV at origin. The HIV prevalence at host area, and now we are talking about where people are moving. And then the length of time of the conflict itself, and of existence of camps.

Looking at the increased risk, there may be behavioral changes, there certainly is an increase in gender-based violence during this time, possibly transactional sex, due to the differential power dynamics. And there may be an increase in resources and services depending on where you are if it is the beginning of an emergency. Perhaps, if you are mixed in an urban population where you do not have access to medical care.

Conversely, in conflict settings, you often have reduced mobility because logistics transport is not functioning. Reduced accessibility to clinics and HIV prevention services. Yet, in some cases, you may have increased resources and access to services in the host area, so if people move into an area where you have nothing, particularly in refugee camps, which are often sourced and funded by international organizations, you may have access to, actually, to better HIV services than the host communities. I am looking at the time, I will move a bit.

That is an example of our thinking, in terms of HIV and transmission, how it may change, it is certainly very context specific. Now comes COVID, and early on, very early on, we work with modelers at Hopkins, Shawn Truelove led this modeling. We want to look at refugee camps and worked closely with UNHCR, and we worked in the largest refugee camp or set of camps in Bangladesh, and then we expanded that to other camps.

Perhaps not surprisingly, what we found, first let me go into the characteristics. We felt that there was a high potential for transmission. You have a massive population in a very small area, very high density. They had relatively low capacity to treat, in terms of numbers of hospital beds. And limited possibility for physical or social distancing.

What we decided to do was to look at low transmission, which would have been more on influenza like  moderate transmission, which is what we were seeing. This was so early on that we were actually using data mostly from China and Europe, and a little bit from the US. This was in March and April of last year, which has some biases, of which we all know.

We predicted, depending on the transmission, out of the 600,000, these are days here, that regardless of which transmission scenario one has used, within 100, 200 days, or even 300 days, there will be tremendous amounts of infections. Infectivity would be very high. The difference in terms of the reproductive rate would purely be the time, which is quite important, because it did allow for improvement and treatment in other areas. And we looked at deaths. The deaths were high, but what was interesting, and we even saw it there, was that the deaths proportional to China, to Europe, and the US, would be significantly lower. And that was primarily using the demographic characteristics of what was happening in those countries. And here, you can see the population age distribution, the camp, versus Bangladesh itself, versus China.

While we knew that the proportionality of the severe cases would be lower in these camps compared to many of the other countries, and this is the same for many camps, we also knew that there was going to be still quite high and overwhelming existing services.

The paper goes on, we think that it was helpful in terms of the government UNHCR, WHO, and NGOs in the field, because they increase, if you look at the right, it increased treatment centers, but we painted quite a severe picture of what was going to happen early on.

Now, what did happen? This is a data hub, all of this is down here, you can access this. I accessed this on the weekend. Here if you look at testing and positive tests, for example, and even tests per million, this is Bangladesh itself. This is the host community in Cox’s Bazar, and this is the Rohingya. Currently, there is testing per million almost a Rohingya, then there are for the others. And you see low positivity rate. So that is a good sign in terms of that there is a fair bit of testing and the positivity rate is low.

Per million, in terms of cases, again, you see Bangladesh highest,, lower and then even lower amongst the Rohingya refugees. And then the deaths per million, again, you see that there for whatever reason, I do not think we have very good reasons, we have many hypotheses. What we were predicting in the Rohingya camps and in most of the camps around the world, has not yet come to fruition. I wanted to say, yet, and I will explain in a second.

What are we seeing? There are a lot of different issues. IOM, I think, did a really interesting report. It made it very clear that there is a lot of distrust amongst the Rohingya towards authorities for understandable reasons. So, there were a lot of – this was more of a qualitative study, although they tried to go broad in terms of sampling. It found that there was a lot of distrust so therefore people were not going to clinics, people were burying their dead elsewhere. There were a lot of rumors about if you were found to be positive, you would be taken to an island that the Bangladesh government wants to relocate Rohingya. So for a lot of reasons, I am sure in Mark's mental health studies, a lot of this may come up as distrust of authorities, so we may not be seeing the whole picture.

Some of the questions thus far, no major outbreaks? Is transmission and effects amongst these populations different or is it similar to national local communities? Is it not being reported? Are we missing it? Has it happened already? There has been a tremendous amount of transmission amongst these groups, and yet, because of the age structure, we just did not see the hospitalization and deaths? And a whole other area that we are studying, that I won’t get into now, but of course relates to HIV and other aspects, is that because we were focusing on COVID-19, how has this affected other diseases, like HIV, malaria, and Ebola in West Africa? Malaria, there increased deaths, excess deaths, due to malaria, more than there were to Ebola because of the concentration on Ebola at that time.

How can we know what is happening? I want to mention this. This was on the 21st. Today is the 25th, just four days ago. The Indian variant is now in the camps. There is a significant increase in numbers right now. Bangladesh puts the camps under lockdown. There is a lot of concern that, maybe we did not see, either we missed it or we did not see what did not happen as much as we expected this first round, unlike what is happening in India. Are we going to see a second wave and see the amount of death, sickness and death, that we expected and hypothesized, but will it be in the second wave?

There is a tremendous amount that we still don’t know yet and we need to look at the surveillance data, mortality data. We have looked at a lot of these. Our hope is that serological studies, they were done in the Rohingya camps, but unfortunately, they are not being reported at this point, because it is very political. And this is before this current wave.

That is looking at transmission. I think for me at least, it is extremely difficult to hypothesize what, in terms of transmission and ultimately morbidity, mortality, what is similar to a refugee, or to a internally displaced person to the host population, and from their country of origin? And what is different. Over time, we are slowly learning but it appears to be very context specific. I don’t think it makes sense to be able to say it is one way or the another.

When I was in Liberia, I took this picture in 2011, this was in the Liberian paper. “I have heard that the Ivoirian refugees are given enough attention and aid”, and he says yes, “I have also heard that a good number of them have AIDS to spread”.

The amount of stigma and termination against refugees and migrants for being a refugee and migrant, and for spreading disease, is rampant. I am sure you know where I am going to go looking at HIV, and looking at COVID, and what has happened.

Early on, when I was at HCR, we worked with colleagues, and we put together a HIV and AIDS for human rights everyone. We developed cartoons that we were able to, and we met with local populations and the refugees to be able to try to dispel some of these myths and perceptions. This is one way we did it in multiple languages, to again, tried to address, as opposed to just putting up signs saying certain things that we wanted to try to work in comic books, particularly for children.

What we did find, and this is some of our work with UNAIDS, is that refugees do not increase HIV risk. Rather, it is potentially their context and their setting and their environment where it may put them at increased vulnerability. This has been – and I will go through some examples now, but we had the opportunity, at least when UNHCR joined you in AIDS, to really band together to try to reduce stigma and discrimination. So we are seeing the same thing, not surprisingly with COVID-19, this was just one example, I can’t remember, of new humanitarian looking at the Rohingya refugees that Bangladesh hosts. Huge among of tension in Lebanon right now with Syrian refugees. And I like the quote from the Secretary-General Antonio Guterres, "We must act now to strengthen the immunity of our societies against the virus of hate. " We are seeing so much populism, and so much anti-migrant and refugee, even in pre-COVID, we were seeing this, and COVID has exacerbated it.

I think I am getting into this a little bit later, but I thought this picture was powerful. March 18th, just a while back, looking at migrants who were caught trying to cross into the US border. I will talk briefly about title 42 in the US that is being, in my view, abused as a public health tool to stop asylum from occurring in the United States.

Inclusion and access. There have been some positive changes, I must say, compared to HIV and COVID. This is perhaps a complicated slide, but this was something we published a long time ago, because what we found consistently, over time, was that refugees and internally displaced persons, were not included in countries national AIDS control plans, and were not included in, certainly at the beginning, into global fund proposals. This was always very frustrating, global funds based in bGeneva, we would go there often to give talks to try to show them that in fact, refugees were actually being excluded and that the global fund, could just add money to the national policies, a lot of the time the governments were saying, we are not include refugees who are not our citizens into this. But when we looked at the amount of money relative for a national plan, compared to refugees, it was very little. If it was additional, most governments do not have a problem including them.

To walk you through this, there were 60 countries over a period, that at least had refugees in their plans. This was the number of those countries in round one that had refugees and submitted to the global fund in round one. This was no mention of refugees, so the majority did not mention refuse. A majority did not mention IDPs. The yellow means that they just did not submit it that time. The green means they referenced refugees or they referenced IDPs, but it was just in the introduction, there was no activities or interventions and only this, the light blue is probably the most important when you walk through this, to say, well, what happened and did they actually have a plan? Sorry, I said the national plan, this is in the global fund proposals, did they have some, in the proposals, some interventions specifically to address either refugees or IDPs? The bottom line is they often did not.

Over time, we looked at access to anti-retroviral therapy. So early on there was not a lack of access to HIV treatment, yet, we did further studies, afterwards, looking at there were concerns early on that refugees should not receive ART because adherence that they are constantly mobile, which is not really the case because often you stay a refugee in that country for decades. But we did a bunch of studies looking at anti-retroviral adherence among refugees in Kenya and Malaysia. To no ones surprise, we found no difference in their adherence to hosts among refugees. Again, using data to try to dispel myths.

Over time, we did see the inclusion of refugees in host countries national strategic plans. We did see an increase in access at similar levels of nationals to anti-retroviral therapy, but the problem here was, this is what was access in terms of on the books. Regulations, that they were allowed to access, but in practice, this was often not the case.

So, I know it is very different, between something like HIV and SARS-CoV-2. So we know they are varying differences in the diseases themselves, but I would like to equate them a little bit with the clear differences between accessing anti-retroviral therapy, and accessing COVID-19 vaccine. We are very clear what is happening now, and I am sure this has been discussed in terms of the vaccine nationalism.

UNHCR, early on, but this was just in April, they called for equitable access to COVID-19 vaccine for refugees, as many of us have been doing for a while.

There is some good news relative to what was happening with the provision of anti-retroviral therapies early on in similar circumstances.

This is some new data from UNHCR recently, that 157 countries with refugees or other persons of concern to UNHCR, and 80 percent of them, 124, committed to include them or will do so when their plans are finalized. When you compare that to early on with the national AIDS control plan, there is nothing like that.

Another 32 countries are in the process of doing so. However, this is understandable, out of the 124 countries committed to include them in vaccine rollout, only 26 percent, or 32 started receiving vaccines. I think the difference here, currently at least, is that most of those countries, are hardly getting vaccines for the reasons that we know. Similarly, when you compared to the global fund, and you looked at Gavi and COVAX, early on they created a humanitarian buffer. It has not been funded sufficiently to have much of an effect, but at least they were considering how to provide vaccines in these humanitarian settings. That was not happening early on with HIV.

Looking at the time, moving through restrictive laws, policies and practices. I am sure many of you know, this was in 2015, it has changed a lot, and I won’t go over everything, but for the longest time, including if you remember, the United States, did not allow people who were coming over and were going to be resettled, they did not allow people with HIV to be resettled, to actually come here, and that was a big change, when eventually the US changed their policies, and then the AIDS conference was allowed to come back to the US after many years.

There are still areas where countries that are deporting people with HIV. If you are refugee, you will be deported if you are found to be HIV-positive. This was  occurring when I was still at HCR, and sadly, it is still occurring. In those situations what happens, is usually the government gives UNHCR enough time to be able to resettle that person to a third country, as opposed to pushing them back across the border. But they did do that recently in Syria.

This is really in words, what's that other slide was showing. When you refuse to grant asylum on HIV, or someone is HIV-positive, or if you, in some countries, where it was required to have an HIV test before receiving antenatal care, and if you were positive, you would be sent back or imprisoned. One can imagine the implications of these people not wanting to get sufficient access to services, and pushing everything underground.

There was a lack of privacy for refugees who underwent HIV tests. There were a lot of different data confidentiality and protection issues that came up with HIV. I think that set the stage for some improvements over time at least. There was a big push by UNAIDS to reduce these practices.

Now, at is happening in the United States with COVID, I am sure most of you know, Title 42 allows the government, and particularly CDC, to have an order to say that they used Title 42. In March 20th of last year, they said on both borders, northern and southern borders, that we would stop people from coming over because of COVID – in order to stop the spread of COVID. They were also allowed to expel migrants who were waiting for the court date, to be able to know whether they would receive asylum or not.

What has happened, is they implemented this. Unfortunately, though, it is very, very clear, this was the previous administration, however the current administration has still not rescinded this. What is very clear, is this was done not to reduce COVID spread, because Americans, people holding green cards, students, commercial traffic, the exceptions were all over the place. This was used against both US law, and against the international refugee law, to  stop people from seeking asylum.

We have been pushing, we came out early on to our left here, together with Columbia University, and a bunch of others, signed letters to say that this should not be used. It is an abuse of public health. There are many other countries are able to allow people to come over, yet address some of the concerns about COVID. At this time, COVID was so much higher, the transmission within the United States, then where these people were coming, we were more concerned that we wanted people to come over, but we wanted to make sure there were enough prevention measures so they would not catch COVID, actually was the concern then at that point.

We did a global call for action, I am part of a group called Lancet Migration, which part of the Commission Lancet, who were looking at various issues in terms of migrant and health. And I know, I remember seeing the lecture yesterday, forgetting the definitional issues of economics migrants versus forcibly displaced, the key was to include refugees and migrants in all areas of response. This is still not happening in terms of COVID vaccine, across the board, in the US, particularly it is state-by-state, but particularly, what the state requires in terms of identification.

Most recently, just May 20th, five days ago. The UN Refugee Agency, which rarely ever criticizes the US, urged Biden to rescind the border order, meaning Title 42, to not allow people to not seek asylum because of COVID. And then a colleague and myself, we wrote an Op Ed on the hill, earlier on, to try to explain how these public health orders were being abused within the US.

So, I am finishing slightly earlier. A couple of key points: Number one is we do not know as much about SARS-CoV-2 and COVID-19 in general, I mean now, globally, but particularly in humanitarian settings. As with HIV, in such settings, it is very, very likely context specific. A lot of us are extremely concerned that what we are seeing in India, will move into – already started in Nepal and Pakistan, into the massive refugee, Iranian refugees in Pakistan, as well as the Rohingya refugee. As we said, the numbers of cases are increasing significantly amongst the Rohingya, as we speak.

We need to look at improved inclusion of refugees in these national donor plans for COVID-19. We did see there had been a significant improvement compared to the past with HIV, but I think the worry is that it is easy to include them right now, when there are insufficient vaccines for the whole country, but we need to be able to look very, very closely. I know I put in refugees there, but honestly, undocumented migrants in countries is the same. We need to make sure that we include them both from a public health inequity point of view, but also from a human rights point of view.

Stigma and discrimination remains. I think it is false to say definitively that refugees, migrants, do not transmit disease or refugee and migrants do transmit disease. They are just humans and it depends on the context. It depends on their access to prevention and care, it depends of the economy, it depends on so many different factors. Making an assumption in all cases is not the way to go. Analyzing all the context specific areas and then trying to make a decision about how best to support those people who are on the move, as well as where they are moving to and where their transiting through, makes the most sense from a local public health point of view, as well as an equity point of view.

With that, I am going to end with various acknowledgements, I am sure I am missing some people but this is a group effort of thinking over a long period of time. Thank you very much, I will stop there, and I will open, I assume with Mark, for various questions.

Agenda Item: Q&A

SHERI WEISER: Thank you so much, Paul and Mark, for those outstanding presentations. They were very informative. I will start here with a question for Mark from Ricky. What challenges do you anticipate transforming research into front line services, particularly with task shifting in professional groups?

MAARK VAN OMMEREN: I think I got the question, but I would love to hear it once more.

SHERI WEISER: Of course. What challenges do you anticipate transforming research into front-line services, particularly with task shifting and professional groups?

MAARK VAN OMMEREN: I think everything is about vested interests. So professional groups tend to often have vested interests that they think that their profession should do it all. We have a history for mental health, international mental health, for the last 30 years, where traditional psychiatrists were not so happy about integrating mental health into primary health care. They have come aboard on that now, but it has taken a long time. But now we are moving in psychology, where clinical psychology interventions are being task shifted to lay people. We are seeing the resistance.

One of the things I try to say to my colleagues who were in the field facing this resistance, I say, well, if you were going to teach laypeople in the community, community workers, psychological interventions, but the clinicians – but the psychologists in the capital haven't learned these interventions, it is not going to work. It is like someone else told me, it is like organizing - if you introduce an intervention in the lower level of the system, and it is not available at the higher level, it is not going to work, it will not sustain, it needs support throughout the system.

The solution to the challenge is making sure that whatever intervention is introduced at the lower level, that you make sure it is also available at the higher level.

SHERI WEISER:  Thank you. I have a question for Paul, starting with a question from Carol. Given that displaced populations eventually leave humanitarian settings and integrate into destination communities, what policies, services, or interventions are needed to reduce  HIV risk during that migration and resettlement process? In other words, there is a temporality to risk, are there certain moments these processes where migrants are particularly vulnerable?

PAUL SPIEGEL: Thank you. Perhaps first, a clarification. Is that in many situations, so people move whether they be a refugee or IDP, often, the most risky time is actually during movement for a variety of reasons. People are moving, often women and children, sexual violence. That is usually the riskiest time during that time and that is when it is necessary to try to be able to support in many different ways. One of them is to try add support the movement if they can, at least when they are in a safe area outside of the conflict.

There was a supposition there that they integrate, and I think it depends on the setting. Perhaps an economic migrant, at one point, as we are seeing here with the difficulties of their movement, but eventually they may integrate into the US, for example. That question of integration is do they have access? Do they have a car? Do they have access to medical care? Do they really integrate or not? In a refugee setting, they often do not integrate. They are either in camps, although the majorities of refugees and IDPs are actually outside camps. You can think of them, in a way, as an undocumented migrant. Sometimes, they do have access to services but often they don't, and that is one of the biggest problems that we find,  whether you're an undocumented migrant or an IDP or a refugee, if you are not in a refugee camp where services are generally free and of decent quality, generally, most are not in those situations. So how to best provide prevention and care to them can be very complicated when they are not part of the social welfare system of the country of origin or where they are located.

It is a big challenge, often but we are trying to do more and more is we are trying not to provide independent parallel services to refugees and IDP's, sometimes you do not have a choice because the government either pushes them into areas or camp like situations in very remote areas. We are trying to do is, for the most part, for those refugees and IDPs that are outside of camps, in cities or villages, to try negotiate with the governments and the money that would go to parallel services, we try to improve the local services for those communities and those that are displaced.

SHERI WEISER: Thank you. For both Mark and Paul, what advice do you have for trainees interested in conducting research in humanitarian crises? What skills are most important, beyond the skills needed to work in low resource settings?

MAARK VAN OMMEREN: I think it's probably good to think of teamwork. If you are a trainee, to hook up with the team that is doing a larger effort. That would be a very good place to start. There are so many logistical challenges when you are working in the context you do not know very well. It is probably good to be part of something that helps, that is dealing with the ethical issues and logistical issues. Security issues, sometimes. I think that is really a key part.

I think from a training perspective, I think it is very useful to have both qualitative and quantitative skills. I think that is just the purest one of the two, I think is very useful if you can combine them.

PAUL SPIEGEL: Perhaps I will add, agree 100 percent. I think it is very important to be strong both in qualitative and quantitative, and I know now that I've moved from the UN to Hopkins, I think sometimes some people shy away from a quantitative because it can be overwhelming. I would recommend doing both quantitative and qualitative and push yourself for quantitative, if you're still concerned about that.

There something larger I would say we need to think about, there has been a push for a long time - two big areas of change. One is what is called localization, which primarily means more and more allowing money and research to be done at the national level by countries where the refugees or IDP's are.

On top of that, there have been more discussions in the last while about the decolonialization of aid. It is really important. Now it even links with Black Lives Matter. But I think it is taken very seriously by a lot of us, because when you look at how aid and assistance is developed, it was primarily western human rights. Which is not the human rights component, I think is still quite important, and that is changing as China and others are becoming much more involved.

When you also look at where the big schools are that are teaching these sorts of skills, they are primarily in high income countries. When you look at how donors, particularly let's say NIH, CDC and others, it is very complicated to be able to maneuver and try to get these grants. I think it is extremely difficult for those schools, let's say professors in low income and some middle countries, that don't have the help from the administrative and others, to actually just fill out and apply to a NIH. We are perpetuating a lot of issues and making it easier for people in Europe and America, to be able to do research, and it is a bit more difficult to access these funds.

Finally, I will say in terms of donor funds in terms of projects, most of the donor funding from the US and Europe, goes to UN and international NGOs, and then it trickles down to the national NGOs.

Most of these national NGOs will never be able to have both the administrative capacity to financially respond and to report back that is required, compared to the massive international NGOs. It is something I think we need to move and we need to think about who we are training and how we can support national researchers and national NGOs moving forward.

SHERI WEISER: That's a great point, Paul. I am wondering to that effect, speaking of deep colonialization, if there has been a change in recent years in the way that your forming partnerships to better prioritize community-based organizations, community experience, and devising the programs and solutions.

MAARK VAN OMMEREN: I would not say there is a linear line there. Things are changing all the time, how we work and learn. Certainly for a resource priority setting, we have learned we should not just do this with so-called, experts. We should talk to people who are implementing, people with lived experience.

I think there is a shift overall, certainly for mental health, and I think the shift is involving people with the difficulties. And, I would say in mental health, we are better at it in certain settings than other settings. I do not think we are doing such a great job at in humanitarian, in refugees, I do not think we do that so well, but mental health is an area with much more lived experience, which are first people who are affected.

Working with local partners, I think it is varies and it depends on who you are. At WHO, why do I like the trials that I was describing? Why am I doing them? It should the universities be doing trials, not the WHO. But I am doing them because I would like to publish WHO open access materials and I need to know whether it works, and universities, so often have the same vested interest in making it open access.

So we do it. But then, I do not think I am required to build local research capacity. So I go with existing research capacity. If I worked in another job, I would be very much vested in building local research capacity. Everybody should, it is a good value and it is important. But I do not think that is my job, so I'm going a little bit off with you now, we work with people based in low and middle income countries with their own organizations there, but they are often the stars of the country.

PAUL SPIEGEL: Maybe just to add, there needs to be a shift, I think it needs to be and there will be a shift. Everything from universities to NGOs. We have to look at the business models. At Hopkins and many other universities, the faculty bring in their salaries through grants. So that is not necessarily an incentive then, to be working with others and saying well, we will take a smaller amount and work with others. Because that is not the incentive the way it is now.

I think certain business models have to change and similarly with international NGOs and the UN, to keep their lights on, lots of other things, they need to take indirect costs, it would mean a shift of academic universities, we are talking about research, but also international NGOs, to be smaller, perhaps. To not have the same amount of influence and power.

So it is probably not in these group's interests and so, I think it will have to come from donors, certainly PEPFAR has made some deliberate changes now in terms of where that money is going, and it is going to more local and governments. I think that is a good way to be able to move forward.

But I think for us to see major shifts, it will have to come from the donors. Certain donors require us to, and I think that is the way to go, to be open source, everything we do is open source and share our data. That is also a trend that is , and I think eventually will be the norm, which is very positive. But I think we have a long way to go.

There was a humanitarian summit in Istanbul, a while back, and the idea was to have I think it was 25 percent by 2020, but I could be mistaken on that, of the funding going to governments and national NGOs. It is not even close, it is about 2 to 3 percent – most of it still going to international agencies and the UN.

SHERI WEISER: Absolutely. So there is a question for Dr. Spiegel, what can US citizens do to encourage government action to adopt more humane migration policies?

PAUL SPIEGEL: You may have caught my accent, that I am not a US citizen, I am a Canadian citizen, but I can tell you what I would do if I was a US citizen. I think as always, it is contacting your representative. I am getting contacted by a bunch of media organizations. Just in the New York Times today actually, there was an article pushing back against this Title 42 and why, what is happening. Why is this being used or misused?

So I think certainly one way would be speaking out. Another way would be working with some of the key local organizations. So one would be advocacy, but particularly evidenced-based advocacy. At the Center for Humanitarian Health, we really move towards that, but we try not to be partisan at all. We just use the evidence. Try to say this is what makes sense for policies. So I think doing research, and one thing that I noticed, is that again, the incentives in academia may not be the incentives from moving beyond publishing your papers. For promotion and other things, you’ve got to get your papers.

I came from a very different background, I’m a  Professor of Practice, so it is nice to publish papers, but I want to do that next step and actually advocate. I think that researchers in academics can do much more and do not stop just when you publish your paper, but start working. I will just give one quick example.

We have analyzed all of the migration detention standards in the US with CBP, customs border patrol, ICE and ORR, and there are a lot of disparities going on amongst that. So even though we do not want detention to occur, we want to make sure that there is dignity and access to proper healthcare.

So we are now working with members of Congress to try to at least explain to them what we have seen and maybe they can use that to try to make changes.

SHERI WEISER: Great. Thank you. We just have a few more minutes left, Fred asked, just to the point that we are discussing before, of decolonizing this kind of work. He asked, how about including those who we work with in these kinds of presentations as a start and how can we achieve that and engage them in these discussions?

MAARK VAN OMMEREN: I think it is a very good question. The Turkey trial for example, that I presented, the first author is C. Acarturk. He will do a wonderful job presenting this and some of the trials that we have already published in Pakistan, they get first author, they deserve first author, because they lead the work. So they are the first authors.  And they should be presenting too, or co-presenting and be there. Again, Acarturk is a very senior professor, so I do not think he needs our help. But there are others who definitely – I think it is good for NIH, it is good for international work to have international folks speak all over the world presenting saying what they are doing. Their perspective. It would be interesting to see how they see the discussion of colonizing and all the research grants. How difficult it is to write a research grant when university structure is a little bit different than the US structure, where it is normal that you spend a good amount of your time writing these grants.

SHERI WEISER: That is great. I'm going to ask the last question because we only have two minutes left. One from Carol, I'm going to ask you each to say something for less than a minute. What policy changes is needed globally to prepare for the increased magnitude of forced migration?

I gave you a challenging one with one minute to answer.

PAUL SPIEGEL: One thing that comes to my mind is perhaps not an answer per se, is it is going to get much, much worse with climate change, clearly. So I think we need to be able to already consider and think ahead and put both resources. Rather prevent or rather mitigate what we know is going to be happening. It is just like the idea for future epidemics or pandemics, were never good at, we are much better at responding because we do not want to put time and effort and money that may not be used in case something does not happen. Therefore, we are way too reactive.

I think that is one aspect, is we are going to need to recognize the people who are moving due to migration and that is going to be causing more forced. And to try to think ahead of time. I would say, just as I am thinking through, one would be to be able to consider some sort of – which is a  pipedream, but minimum amount of healthcare for when people are crossing borders at least until they can pay for something for themselves. But the problem is that would always go onto the country – the cost from the country.

The other aspect I would like to see and we have been stopped at every way because of data protection and confidentiality issues, but better understanding in terms of let’s say, health passports, something that we can enable in terms of continuity of care to know what is happening. We have tried that, but it is extremely difficult as you can imagine from a legal point of view.

SHERI WEISER: Thank you so much, and Mark, I am wondering if you don't mind, if you have anything to add to type it in because we are at time. I think we are about to have a five minute break. So thank you for outstanding presentations and excellent discussion. And now we will have a break.

HOLLY CAMPBELL:  Thank you, this is Holly. I would like to also announce that we will have two quick polls, just some questions that come up, if you feel like answering them, that would be great. Just a little but a fun. Thank you.

(Break)

Session 4: Implications for Interventions

CAROL CAMLIN: I am just going to give us one more moment to re-gather. I believe we should start now. It is my pleasure to serve as the moderator for session 4, Implications for Interventions. Welcome back, everyone.

I am very excited about our panel of speakers. We have five panelists speaking for the session. Including Sheri Weiser, who is a professor of medicine and internist the Division of HIV Infectious Diseases and Global Medicine here at the University of California, San Francisco. Her research is focused on the impacts of food insecurity and treatment outcomes for HIV and other chronic diseases, both domestically and internationally. She does a lot of work on sustainable food insecurity and likelihood interventions.

We have Harsha Thirumurthy, Associate Professor in the Department of Medical Ethics and Health Policy at the University of Pennsylvania. Harsha is Associate Director for the Center for Health Incentives and Behavioral Economics and also affiliated with Penn’s Population Studies Center. His work, he uses insights from economics and psychology to design and evaluate interventions that can improve health outcomes in low-income settings.

We have Kelli O'Laughlin, an Assistant Professor in Departments of Emergency Medicine and Global Health global health at the University of Washington. She is a global health research scientist focused on assessing the health needs of refugee populations and on designing and evaluating refugee specific interventions to improve care.

Kate Clouse is an infectious disease epidemiologist and Assistant Professor at Vanderbilt University's School of Nursing. Also affiliated with the Vanderbilt Institute of Global Health, Division of Infectious Diseases. Her research specialty is in patient care engagement, with a focus in South Africa.

Andrea Low is an infectious diseases physician, and clinical epidemiologist Assistant Professor of Epidemiology at the Mailman School of Public Health at Columbia University, also the Clinical and Scientific Director on the multi-country PHIA Project, which conducts population surveys in 14 countries in sub-Saharan Africa.

Without further ado, I am going to pass things over to Sheri to give us her talk.

Agenda Item: Climate induced migration and food insecurity as pathways towards worse HIV health  

SHERI WEISER:  Wonderful. We know that climate change is the biggest public health threat of 21st century, and affects every stage of life from birth through older age. Today, I will be focusing on two of the most immediate and harmful consequences of the climate crisis, namely food insecurity and forced migration.

Starting with migration, we expect that there will be at least 200 million climate migrants by the year 2050. What this means is that 1 out of 45 people in the world will have been displaced by climate change. At the catastrophic warming levels predicted of 2 to 4°C, migration is going to be simply necessary for the survival of many.

Most climate-related migration is expected to occur within resource poor countries, which may lack the resources and adaptive capacities to absorb these massive population shifts. This figure shows projections of internal climate migrants for the three most effected regions, sub-Saharan Africa, Southeast Asia, and Latin America. In all scenarios of the missions, show no change in our admissions to aggressive climate action, show climate change will be a driver of internal migration, so even in the most optimistic scenario, we are talking about 35 to 72 more migrants by mid-century.

Climate change can either influence migration directly, such as through forced displacements, from sea level rise and hurricanes, and also indirectly by amplifying socioeconomic and political drivers of migration.

For examples, massive crop failures and livestock death from extreme heat or drought can disrupt livelihoods and cause food insecurity, which then increases pressures on affected populations to migrate. As you can see in the bottom figure on the right, mobility patterns can include migrating away from climate affected areas, such as coastal areas inland, and from rural to urban areas. People also migrate into areas with climate risk, particularly if they are living in a region that is highly susceptible to climate change. Finally, mobile populations may actually fare the worse in terms of their climate health risk because of their inability to leave.

We've been hearing about many of the negative health consequences of forced migration, I will go through this very quickly. Infectious disease outbreaks are very common within refugee camps. We have been hearing about the fact of disruptions in livelihoods and social networks can put people at risk for malnutrition. STIs, of course included HIV and poor mental health. From migrants who move from rural to urban areas, there is also risk of chronic disease due to lifestyle and diet change, and also disruptions in healthcare access. Of course we have been hearing about the flipside, since for some mobility can  provide a pathway to health improvement by offering safety from violence, reduction in food and water insecurity, and access to better quality healthcare.

In terms of food insecurity, climate-related disasters have also begun to reverse the decline in global hunger we've seen in recent decades, and it is now estimated that one of nine people in the world are hungry. Climate change will increase this number by 20 to 50 percent by mid-century. We know that climate related food insecurity disproportionately impacts people of color and of vulnerable populations like children, pregnant women, the elderly, and refugees.

The regions of the world that have the highest prevalence of hunger like sub-Saharan Africa, Southeast Asia, and Latin America, are also those most vulnerable to climate threats. As you can see that the red and orange colors in these maps that represent both high hunger and climate variability.

Climate change negatively effects all dimensions of food insecurity. So it decreases food availability by causing crop loss, animal death, and migration of fish. We see reduced food access through its economic impacts like income loss and unemployment. It contributes to reduced food utilization through worse in diet quality, and food and water contamination. Finally, the stability of the entire food system can be compromised through climate induced market volatility and political instability.

Food insecurity similarly negatively effects health in many ways. For example, we and others have shown that food insecurities associated with STI risk, with increased mortality from Ebola virus disease. There is a huge body of literature linking food insecurity to chronic disease risk such as hypertension, diabetes, coronary artery disease, chronic kidney disease. And we and others, have also found that food insecurity can contribute to neurologic and mental health problems, like cognitive decline, migraines, depression, anxiety, and substance use disorders.

How is climate change linked with HIV? Some of the same effects of climate change that I've just gone over, like food insecurity and migration, also contribute to the spread of HIV. This is supported by recent data. So for example, the recent study that used demographic and health surveys or DHS data, from 19 countries in sub-Saharan Africa, found that each additional year of experiencing drought led to an 11 percent increase in HIV prevalence.

Looking at the converse, we recently analyzed  DHS data from 23 countries in sub-Saharan Africa, and found that each additional year experiencing excess or extreme precipitation led to 25 percent higher odds of HIV among women. A final study, also with DHS data from 25 countries, found that a one-degree Celsius change in average monthly temperature was associated with an 8.5 percent relative increase in the proportion of the population with HIV.

It is not surprising that these massive public health threats are linked since they both preferentially target the most vulnerable among us, and not to be a broken record, but many of the same regions with the greatest burden of HIV, shown in the top map in the pink and purple colors, strongly overlap with those with the highest climate vulnerability. In the bottom map, in the orange and red colors, I already talked about that these are the same regions with the highest prevalence of hunger.

Here is a conceptual framework at we developed to understand all the intersections between climate change, food insecurity, migration and HIV. Starting at the left, you see global climate change, which includes rising temperatures and sea levels, more extreme weather events, biodiversity loss, among other things.

These factors drive migration and food insecurity, which are also bi-directionally linked. We know that climate change is climate related food insecurity, as a huge contributor to migration, as I discussed.  Migration, in turn, can also increase the risk of food insecurity. For example, if people are forcibly displaced to crowded settlements. Food insecurity in migration in turn lead to poor HIV outcomes through several pathways, including increases of infectious diseases, gender-based violence and behavioral risk factors.

As an example, we know that both food insecurity and migration can increase gender-based violence which is in turn an important factor for HIV transmission and worse HIV outcomes.

HIV morbidity and mortality could then contribute to decreased to household resources and labor availability, which then could lead to further depletion of natural resources to deforestation, overfishing and land degradation that perpetuates a vicious cycle between climate change and HIV. I will use this framework to present data on the pathways from climate change to poor health, starting with climate driven migration.

In a qualitative study among HIV infected individuals in the regions of Kenya, led by Tammy Nicastro in our group, floods were actually viewed as his huge contributor to migration because of destroyed crops, homes, and infrastructure. Here's a quote: Floods have increased because you find that when it rained last, places like Modi, were so affected to an extent that those who list there were forced to relocate to other places. There were some homes that were destroyed completely.

Also in our work, we found that there may be gender differences in climate-related migration. In a recent analysis, using DHS data from 23 sub-Saharan African countries, in a just did analysis, we found that women have higher odds of migration in both conditions of drought and extreme rain, but there is no statistically significant association for men.

Turning to food and secured in malnutrition, as I mentioned, these are both important. Impacts of climate change and key drivers of HIV acquisition risk and worse outcomes. There's actually been a lot of exciting research from the Planetary Health Alliance looking at climate change’s impacts on micronutrient deficiencies and also increased chronic disease burden. Elevated CO2 levels have actually been found to contribute to 3 to 17 percent decline in the amount of protein, iron, and zinc in important crops like wheat and rice. And the figure at the bottom, show the risk of deficient intake of these three nutrients under the atmospheric CO2 levels predicted in 2050, with the red and orange colors indicating the highest risk areas.

Climate change is also driving a rapid decline in the pollinator population of the world, and a large proportion of dietary nutrients globally come from pollinated crops such as fruits, vegetables, and nuts. An estimate half a billion deaths per year already are due to inadequate pollination. Finally, over one billion people in the world rely on fish in their diet. However, since 1996, fish catch has been falling and this is part related to climate change, and this puts many people at risk of zinc, iron, and vitamin A deficiencies.

As an example of how climate change affects nutrition in a meta-analysis led by Mark Lieber on our team, we found that drought conditions are associated with 46 percent of higher odds of both wasting and underweight prevalence. In a prediction model, we estimated that climate change is actually going to increase this prevalence of malnutrition by greater than 50 percent by the year 2050.

In Tammy's qualitative study, participants talked about how loss of crops, animals and income due to extreme weather, worsened food insecurity, diet quality, and also contributed to weight loss and stenting in children. “Weather changes affects yield on my farm. Too much rain or drought interfere with the growth of plants and lowers the quality of yields. This interferes with our children’s growth since they are forced to eat food that are difficult to chew.”

Now look at how food insecurity effects HIV health. There is a large body of literature showing that food insecurity worsens health along the entire HIV cascade of care.  So we and others, have shown that food insecure individuals are nearly three times as likely to become infected with HIV, have three times the odds of antiretroviral non-adherence, and are even two times as likely to die compared to individuals who are food secure.

I would like to quickly highlight another pathways through which climate change can drive poor outcomes, but also may have these synergistic effects with food insecurity and migration. These include higher prevalence of other infections, behavioral risk factors, and increased violence.

Many studies actually show that climate change can increase or alter at least, the distribution of infectious diseases globally. This is particularly true for vector and waterborne diseases. In turn, diseases such as malaria and diarrhea can negatively effect the health of people living with HIV.

As an example of this, using Uganda National Panel Survey Data, Epstein, in our group, found that as precipitation increased, self-reported diarrhea, cough and fever decreased. But you can see that this levelled off and started increasing again at the highest levels of rainfall.

Going back to Tammy’s qualitative data, participants described increased opportunistic infections which they contributed to cold and wet living conditions, more diarrhea outbreaks which they linked to contaminated water from floods, and increased malaria incidents that they also perceived to be related to standing water from floods. And all of these factors contributed to worse HIV health.

Another important pathway towards worse HIV health, in Tammy’s study, participate talked about how infrastructure erosion from flooding and extreme weather events undermined people's healthcare engagement and adherence to ART. I will read the quote here. “When it rains, reaching such places is a challenge since the roads become muddy and impassable. Reaching Minyenya clinic is a hustle because the roads are in deplorable condition. The fare is hiked by motorcycle operators. It is very hard to find any means of transport.”

As another example of how climate change can impact health behavior, we use DHS data to examine the association between drought in the previous year, and self-reported HIV testing in 11 Eastern and South African countries. We found that drought was associated with lower odds of testing among both rural and adolescent populations. We hypothesize that these populations are more vulnerable to income shocks and food shortages from drought, which then could impede their access to medical care.

Looking at the final pathway, Adrian Epstein and I, tried to understand  whether drought was associated with intimate partner violence using DHS data in 19 countries in sub-Saharan Africa, and among over 3000 women in 19 countries, we found that women living in severe drought had higher odds of reporting having a controlling partner and experiencing physical or sexual violence.

Women living in mild or moderate drought also had similar outcomes. There you can see results were slightly attenuated consistent with a dose response effect.

I'm timing this and I have five more minutes, just in the last five minutes I want to turn to a more hopeful part and talk about solutions because there is in fact a lot that we can do to tackle how climate change affecting HIV and other health problems.

According to the Lancet Commission on Health and Climate Change, acting on climate change can actually be the greatest global health opportunity of the 21st century. This is because we know that climate change solutions have many healthcare benefits.

Just as an example, if we decrease fossil fuel emissions by driving less, we will decrease the 7 million deaths a year attributable to air pollution and the 5 million deaths linked to physical inactivity.

For approximately 10 years, I have been working alongside several colleagues, with my co-PIs listed there, on an intervention study in Kenya. We call Shamba Maisha, which means farming for life.

This climate adaptive intervention aimed to address the broader drivers of food insecurity and poverty, which in this case was frequent drought and unpredictable rainfall patterns occurring in the setting of climate change. So the intervention included a loan to a human power water pump for irrigation and training and sustainable farming practices, and financial management.

For example, farmers are taught to practice regenerative agriculture. So helping to replete the soil of essentially depleted nutrients. In our pilot work, we found improvements along many of the proposed pathways, including improvements in food security, diet quality, mental health, empowerment and even reductions in violence, as shown on the quote on the right.

In terms of HIV clinical indicators, people in the intervention group had better immune function with their increase in their CD4 cell counts. About 165 cells compared to control participants. They also had 7.6 times the odds of being virally suppressed.

We also learned and are learning, that solutions need to be flexible due to unanticipated climate challenges. For example, this water irrigation intervention worked really great for drought, but when the weather shifted over time to were excessive rains, crops were flooded and eaten by hippos, so we had to devise different solutions such as potential crop loss insurance for flood damage, and digging trenches to keep hippos away, as shown in the picture.

Ocean farming I think is another great example of regenerative intervention, that can improve both health and livelihoods. This is a form of aquaculture that uses a vertical approach to diversify what can be harvested from the ocean. You can see in the picture on the right, this include extracting sea salt from the surfaces, cultivating kelp, seaweed and scallops through suspended ropes and nets, and catching fish and growing oysters via cages on the seabed. These ocean farms can reduce food insecurity and overfishing. They are also low cost, zero input, and easily scalable solution, which provides a climate adaptive livelihood intervention, and particularly for women. They are also sustainable, since the components feed upon and support each other.

Seaweed also has other climate benefits since it can be added to livestock feed to reduce methane emissions and it sequesters carbon, helping to mitigate ocean acidification. A final environmentally sustainable model, that has been engaged as a strategy to address food insecurity in the US, is urban gardening.

Kartica Palar and I, conducted a small qualitative study among 45 LatinX, largely immigrant individuals living with HIV or diabetes in San Jose, to look at the health impacts of an urban gardening intervention. Participants noted improvements along nutritional, mental health, and behavioral pa, as you can see.

Just before I conclude, I want to mention that  there are still critical gaps in knowledge. There is virtually no data on how metrics of climate change such as severe weather events affect HIV incidents and treatment outcomes. We need to better understand those pathways through which climate change may affect HIV prevention and treatment outcomes using longitudinal studies. And there is a need for intervention research to improve HIV health in the context of climate change.

In summary, we have seen how climate change drives food insecurity and migration, which are important pathways towards worse HIV health. I hope I have also demonstrated the opportunities to design interventions that benefit both the environment and human health. Going forward we know that health considerations should be at the forefront of climate solutions, and that the healthcare sector has a crucial role to play here.

There really is no greater threat to humanity as health professionals trained to care for patients and protect public health, we should be compelled to act in this space. I will end there. Thank you very much.

Agenda Item: Harnessing social networks to improve HIV prevention and care cascade outcomes in mobile men

HARSHA THIRUMURTHY: Thanks everyone. It is always nice to follow Sherry, big shoes to fill here. Thanks to Carol and others for organizing this workshop, and I am very pleased to have this chance to tell you about some work we have been doing on social networks.

So, my name is Harsha Thirumurthy and I will be talking about harnessing social networks to improve HIV prevention and care cascade outcomes in mobile men.

To motivate this work, it is worth recognizing a few points. One is that we all know that there are sizable gender gaps in utilization of HIV services. You see this in the figure that is shown here for UNAIDS Global Report. It is worth noting before we dive into the rest of this work and continue in the talk, that closing these gender gaps is important first and foremost, for improving men’s own health outcomes.

When we focus on HIV and having high access to HIV services and high coverage of testing and treatment, can be really vital for reducing HIV mortality and more generally, utilization of health services that are available. Can be vital for improving men's health outcomes.

The second, this is a point that often gets made as well, increasing use of HIV services among men is also vital for reducing HIV incidence among women. Especially young women in sub-Saharan Africa. So that is the overarching motivation for the work I will be talking about today.

Over the last decade or so, we have seen of course a rapid expansion and access to various types of HIV services, and HIV testing services are no exception. In this review that my colleagues and I published earlier this year, we note that there has actually been remarkable scale up of both community and facility-based HIV testing services in sub-Saharan Africa. Each of these testing modalities have a lot of advantages that we note in this article, but they also have disadvantages in terms of what those modalities cannot accomplish. It is worth noting that these gender gaps and HIV testing based on a previous figure, have persisted in many communities despite the scale up of various different types of testing services.

That only underscores the need for alternative approaches that can reach the men who have not yet learned their status as well as the last 20 percent and others who are high risk and have not been tested frequently.

So it is with that rationale and the need for increased coverage of HIV testing, that we have been looking at network-based approaches and potential that those may have for increasing coverage for HIV testing. We think in this work that social and sexual networks have a lot of potential to increase testing coverage among hard-to-reach individuals. For the purpose of this talk, I will focus primarily on hard to reach individuals, such as highly mobile men who work in fishing communities in the Lake Victoria region. As well as men who purchase sex, that is another population that is hard-to-reach. It is communities where there is a high prevalence of transactional sex, this is important population in which increasing coverage of services is important.

It is also worth noting that network-based approaches may have potential in other populations, other settings as well, it is not what I will focus on in today's topics, but contemplating the potential that that has elsewhere is addition.

So the reason why we think network-based approaches are useful is that we think it is possible to really leverage the power of social and sexual networks by combining it with technologies such as HIV self-testing which are portable and easier to utilize in communities outside of facility and community-based testing services.

So the work we are doing, is combining HIV self testing which has become part of the standard of care in many countries following the WHO guidelines. We want to combine the self-testing with social networks and explore whether that can be useful for closing gender gaps.

So to outline the remainder of this talk, what I would like to tell you about his a few recent studies we have done of network-based approaches that utilize self- testing. We call this approach secondary distribution, and you will see why we use that term, as I tell you but the studies that have been done, in the first set of studies I will focus on women's sexual networks as being the primary feature that we seek to leverage for increasing testing coverage. And then, I will move on to tell you about an ongoing study that Carol Camlin and myself and others, are currently leading, which is focusing on social networks of men in fishing communities in Kenya.

The recent studies that I mentioned are focused on an approach that we call the secondary distribution of self-tests within sexual networks. There are few studies that we published to both first pilot the approach and assess its feasibility, its safety, and acceptability. Then, a randomized trial that has been published as well as one that has recently been completed.

In these studies, what we have done is recognized that women use health services at higher rates than men do. Particularly for women whose partners are hard to reach, even with existing facility and community-based testing services, that self-test may offer a potential avenue through which it is possible not only for the men to learn their status, but also for women to have greater awareness of their partner's status.

In the first study we did, we recruited about 300 women, both from antenatal and postpartum clinics, as well as female sex workers, and provided them with multiple self-tests, along with education on how to use the self-test. We also provided guidance for strategies to use to introduce self-testing to their partners. The importance of using discretion when raising the topic of testing. Then we sought to assess what outcomes were achieved when it came to HIV testing among the partners of those women.

What we found in this study is that, in all of the populations including female sex workers, there were very high rates apartment testing that took place. In the left panel here, you see the outcomes that are achieved for women in antenatal clinic studies and on the right is for female sex workers. In general, there was primary partner testing rates that were in excess of 90 percent. In dark blue, you can see the couples testing rates that were achieved as a result of this self-testing intervention.

In addition to this descriptive study, we followed this with a randomized control trial that was done among pregnant and post-partum women, and it confirmed the very promising pilot study results that we had obtained, and showed that this was a much more effective strategy for increasing partner testing than the current standard of care in many clinics, which is to invite male partners for clinic-based testing.

This study is one illustration of how sexual networks can be used to promote testing. There is also another feature of this particular intervention that I think is compelling, and one that we explore further in a subsequent study, which is suggestive evidence that this availability of self-tests coupled with the use of sexual networks facilitated safer sexual decision-making by women. So there was a lower likelihood of both sexual intercourse and higher use of condom use when women were able to learn their partners status and those partners have tested HIV positive. This is one implication of the intervention that we explore further in this study.

Following these two studies in Kenya, we received support from the National Institute of Health to carry out a much larger randomized control trial, among women who are at higher risk of HIV exposure. In this cluster, randomized controlled trial, completed last year, we were working with women who have higher risk on the basis of HIV negative women who had higher risk on having self-reported two or more partners in the past four weeks. In these intervention clusters, women had sustained access to HIV self-tests, unlike the one-time provision, there is a sustained provision of free HIV self-tests. Whereas, in the control clusters, we had something resembling the standard of care.

You can see in this study, we recruited participants among whom there is a very high prevalence of transactional sex. So 81 percent regularly reported exchanging sex for money or material goods. About 66 percent of participants reported sex work as a primary or secondary income source. Many women reported that they did not know their partner's recent HIV status, 42 percent saying their partner had not tested in the past 12 months. From the standpoint of HIV risk, it is also notable that condom use was not universal despite the lack of knowledge of partner status.

One of the encouraging findings from this trial is that, consistent with the previous studies we had done, there were significantly higher primary partner and couples testing among the women who benefited from sustained access to HIV self-test. We actually see higher partner testing and couples testing, not just initially, but also over time so women were able to learn their partner status on a frequent basis and they also reported high acceptability of self-testing by their partners.

In addition to the primary partners, given that many women reported two or more partners, and reported in engaging in transactional sex, we also see that in data we have obtained in over 11,000 transactional sex encounters, that there was a significantly higher partner in couples testing with transactional sex partners.

We also noticed higher condom use following learning that a partner tested HIV-positive, and lower condom use by learning that the partner was HIV-negative. So both findings are interesting and raise important questions that we can perhaps explore in the discussion.

One notable result is that, despite these effects on partner and couples testing, there was not any significant difference in HIV incidence between the two groups.

Finally, before I wrap up, I would like to shift from the reliance on sexual networks and approach to both increased testing coverage and to increased awareness of partner status in reducing HIV risk, to a reliance on social networks of men in the very same communities where these other studies have taken place.

In this study, which is getting underway at the moment, funded by NIH and co-led by Carol Camlin and myself along with (indiscernible Name) and Zachary Corina in Kenya, we are looking at the potential for social networks of men in fishing communities to not only promote HIV testing but also to facilitate linkage to care and linkage to prevention, along with retention and care prevention among men.

Very briefly, I would like to tell you of the study. The overall study design and rationale. In the study, it is a cluster randomized trial with the clusters not being geographic clusters, but rather we are defining as close social networks of men in fishing communities. Following extensive social network mapping exercise in the communities, we are identifying close social networks or smaller groups of individuals who were mostly connected to each other. Viewing those as networks within which we may be able to promote public health objectives.

We want to leverage, specifically, the role of th network central individuals in those close social networks to promote testing and ongoing engagement with HIV services. The intervention here is building upon the prior work, but also adding on some new features.

Number one, it is relying on social networks rather than sexual networks. It is using self-tests as before, and it is also introducing financial incentives to offset many of the costs associated with seeking HIV services.

The outcomes will focus on the study begin with HIV testing uptick among men, and continue on to linkage to ART PrEP after the testing takes place, as well as viral suppression and PrEP adherence. The goal here is really to view the network central individuals as being primary agents who can promote health outcomes within their networks.

I have told you briefly about the rationale, but I want to mention once again, that we are relying on self- testing because in these communities, accessing clinic based services can be challenging and self-testing has proven to be highly acceptable to men in these communities. There is an ongoing scale up of self-testing that is taking place in Kenya, and we know that there is a need for linkage interventions following self-testing. That is one reason why we introduced these other features.

In the social networks, we are selecting highly connected men to act as the promoters who may be able to overcome barriers to seeking HIV services among the men they know. The financial incentives, we hope will offset the transportation costs, which are essential to incur if someone wants to go to the clinic and initiate ART or continue to go for their monthly or tri-monthly appointments.

The current status of the study is shown here, it also illustrates the sequence of events that we think will be necessary for the study to take place. Number one, following a pilot study that we did last year, we have conducted a social network survey in the beach communities to first figure out whom men know and who they interact with. And then following this network mapping, we are going to be identifying the close social networks and randomizing them to be intervention or control clusters. The promoters who we identify each of these social networks will undergo training on not only how to use self-test, but also how to promote self-testing among the men and their networks. How to facilitate linkage to ART and PrEP and continued engagement in ART and PrEP over time.

We view the promoters as being both central to the success of the intervention, but also as being the individuals that it is crucial to train properly on how to use self-tests and how to promote engagement in HIV services. Following the randomization and training of promoters, we will be then observing outcomes three months later to see whether testing has taken place among the men in the close social networks, and then we will also be looking at six and 12 months into initiation of ART and PrEP, as well as many of the biological outcomes that matter when it comes to ART retention, as well as continued use of PrEP.

This is one example of how we are using close social networks. I realize that the studies I have presented, both the ongoing and previous ones, may raise a lot of questions that people have about how networks can be used to promote HIV prevention and treatment. I welcome the discussion and I want to acknowledge the many individuals who have been part of this work. Thanks.

Agenda Item:  Optimizing refugee-specific interventions to improve care

KELLI O'LAUGHLIN: Thanks, Harsha. Hello all, my name is Kelli O'Laughlin, and I'm a faculty at the University of Washington. I am very pleased to present today, and I want to acknowledge funding support for the National Institute of Mental Health, as well as from the University of Washington Center for AIDS Research.

First, I am going to share with the background on HIV and refugees. There are 40 million people estimated to be living with HIV globally, with over 28 million of those people living in sub-Saharan Africa. 6.6 million of refugees live in sub-Saharan Africa, however, the number of refugees in this region living with HIV/AIDS is not known. Refugees face hardships which may increase their vulnerability to HIV, compared to non-displaced populations, as Dr. Spiegel nicely introduced for us this morning.

Specifically, hardships faced by refugees include difficulty meeting basic needs. Disrupted social networks, limited livelihood opportunities. Threats to security. Prolonged displacement. Potential exposure to sexual violence and/or transactional sex. And at times, difficulty access condoms or treatment for sexually transmitted infections.

It's important to note, especially in the context of this workshop, that refugees are an inherently mobile population. By definition, a refugee is a person who has fled war, violence, conflict, or persecution, and has crossed an international border to find safety in another country. Many refugees then continue to move even after that initial period of displacement, reasons for this vary, but they include the need to access basic needs. Pursuing educational opportunities, seeking livelihood opportunities, visiting family and friends, and attending events such as weddings and funerals. And then, generally working to integrate with the local community.

There is a growing recognition that refugees mobility is a positive asset that can contribute to their lasting protection. With that understanding, the medical, policy and research community must collaborate to understand the implications of mobility on the movement in the context of HIV to enable refugees to engage in a care in a manner that is acceptable to them and that optimizes their health.

With that, I want to tell you but the study site where I am conducting research in Uganda for the last 10 years, in collaboration with local partners, to study the HIV cascade of care. This is Nakivale refugee settlement in Uganda.

The country of Uganda in Eastern Africa, hosts approximately 1.4 million refugees, and the Nakivale refugee settlement is located in southwestern Uganda. It was established in 1960 and is approximately 71 mi.² with the current population of about 130,000 people living in the settlement.

The majority of refugees in Nakivale are from the Democratic Republic of Congo, Burundi, Somalia and Rwanda, as well as a minority from other neighboring countries. Within Nakivale, there are four health clinics which serve refugees and Uganda nationals and provide free HIV testing in antiretroviral therapy.

HIV prevalence from the refugees countries of origin, ranges greatly from 0.5 to 6.1 percent. Routine HIV testing studies that our team conducted in Nakivale,  demonstrated that prevalence of HIV of 4.5 percent. The HIV prevalence in the surrounding region of Uganda, outside of the settlement, is 7.3 percent.

I am not going to share with you about research that our refugee focus team has conducted in Nakivale related to understanding and optimizing HIV care engagement in this setting.

Our refugee focused research team consists of a number of collaborators, presently we are working in partnership with the United States High Commission for Refugee’s, Medical Teams International, the Infectious Disease Institute at Makerere University, and we benefit from the contributions of the community advisory board, that consists of refugees from numerous countries of origin, that vary in age, gender, as well as their position in the community. It does include refugees that are living with HIV.

The first study I want to tell you but today was one that we conducted on the HIV care cascade in Nakivale

refugee settlement. Our team conducted the study in 2013, to access the cascade of care among refugees and Ugandan nationals. In this study is 6850 individuals that participated in clinic based routine HIV testing, 276 individuals were found to be living with HIV. Of those newly diagnosed with HIV, only 54 percent linked to HIV clinical care within 90 days of their diagnosis.

This highlighted the need to investigate linkage to HIV care in this setting and opportunities to improve engagement in HIV care. Our team conducted a follow-up qualitative study in Nakivale, to understand barriers to HIV care and potential approaches to overcome them. We enrolled participants after HIV testing, and invited them to interview with us 90 days after their new diagnosis. We conducted semi-structured interviews with 54 clients and 8 health worker staff to assess participants experiences accessing HIV clinical care.

From these interviews, we learned that refugees living with HIV faced daily challenges that are barriers to engagement in HIV care. Including food shortages, problems maintaining shelter, health needs, difficulty accessing water, limited childcare, difficulty with transportation and limited emotional support.

For some individuals, especially those able and willing to disclose their HIV status to others, they were able to benefit from social support, allowing them to overcome these daily challenges and successfully engage in HIV care. For others, particularly those with anticipated stigma, posttraumatic stress disorder, depression or anxiety, they often defaulted to non-disclosing their HIV status, and by not sharing the status, they remained isolated and unable to benefit from the social support of others and faced more difficulties, successfully engaging in HIV care.

Equipped with this information, we began a study to enroll a cohort of people newly diagnosed with HIV in Nakivale refugee settlement to assess which correlate quantitatively with failure to link to HIV clinical care. Enrollment site in Nakivale were three, one at Nakivale  Health Center, Juru Health Center and Kibengo Health Center.

Adults accessing HIV testing were invited to participate, and again, this included refugees and Uganda nationals.  To conduct this evaluation we surveyed participants prior to HIV testing to evaluate their demographic information, perceived health, HIV testing history, distance to clinic, time living in Nakivale, and travel history.

We also used validated surveys to screen for depression, generalized anxiety disorder, PTSD, social support and stigma. We enrolled 5568 individuals in this study, 65 percent of whom were refugees or asylum-seekers, and 34 percent of whom were Uganda nationals.

Comparing HIV-positive and HIV-negative participants, among the HIV-positive participants, they were more likely to be female. Uganda National or screen positive for PTSD, and they were less likely to be living with a partner or to report being in good health.

We asked people the question how long they had been living in Nakivale refugee settlement. This graph shows a comparison of participants testing negative and positive for HIV and the length of time they reported living in Nakivale. Nearly than half of those newly diagnosed with HIV, 48 percent, and well over half of those without HIV, 58 percent, reported living in Nakivale for over five years.

We then asked how long they planned to remain living in Nakivale. In this graph shows a comparison of participants testing negative and positive for HIV and the length of time they planned to remain in Nakivale. Twenty-six percent of those testing HIV-negative thought that they would remain for more than five years. Compared to 34 percent of those who had tested positive. A large majority of both HIV-positive and HIV negative participants reported they did not know how long they would remain in Nakivale or they did not answer the question.

This time to remain being higher among HIV-positive matches findings from our qualitative study on adherence to HIV medications published in Conflict of Health in 2018. In that study, refugees spoke of missing ART doses when travelling away from the settlement. They also voiced concerns that leaving the settlement could disrupt their access to ART and prevented some refugees from leaving the settlement.

We asked participants if they had been away from Nakivale for more than or equal to one continuous month in last year, and 33 percent of HIV positive participants reported that they had. This compared to 25 percent HIV negative. Among the HIV positive participants, the reasons the ones that were away, the reasons that they were away, were mostly for employment and trade, but also family related and some school related. Among HIV-negative participants, the reasons were similar, and the distribution was similar but also included some other minor reasons including jail and health-related.

I want to point out that for this study we did ask about one continuous month or more away from the settlement in the last year. We did not ask about shorter periods of time away, movement within the settlement, which is quite large geographically, or the impact of movement on HIV care.

This graph demonstrates the proportion of participants linked to care by years Nakivale among those who had been diagnosed with HIV. Importantly, we found no statistically significant difference among these categories of time and linkage to care. However, this may have been related to the small sample size or missing data from participants who did not answer this question.

Health status, health limitations, prior HIV testing and distance to clinic were not associated with linkage to care. In evaluating the cohort of individuals newly diagnosed with HIV and the association with each of the following variables, screening positive for PTSD, depression, anxiety, presence of support and stigma with linking to HIV clinical care, the only significant finding was the presence of social support was associated with linking to HIV care.

Take away from this research, is that social support is associated with linkage to care among those testing for HIV in a refugee settlement. More information is needed on characteristics of mobility and the impact of mobility on HIV and care engagement for refugees.

Now I will share with you our current and future research relating to enhancing HIV care engagement for refugees in Nakivale refugee settlement and elsewhere in Uganda.

First, I want to tell you about an intervention to optimize linkage and retention to HIV care in Nakivale, which is currently underway.  It is called, Early Client Led ART Delivery, or Early CCLAD. This study is funded by the National Institute of Mental Health through K-23.

The background for this intervention is that we wanted to enhance access to HIV services for people living with HIV in the refugee settlement. We wanted to include task shifting, when possible, given the overburden to clinics. And we wanted to increase participant agency and seize the opportunity to foster social support, given our prior research demonstrated that social support is associated with linkage to care.

Using an intervention mapping approach in Nakivale, we designed Early CCLAD. The description of this intervention is that newly diagnosed clients placed into groups of 3 to 12 people based on geography. They had their first meeting at the HIV clinic with the clinic counsellor, and subsequent meetings were in the community without any staff. One group member, varying each month, would attend HIV clinic prior to group meeting, and pick up each group member’s ART, and then go back into the community to meet with the group and distribute the ART and have a meeting. Also we made a plan that clients were to attend HIV clinic twice yearly, and that the outcomes would include meet again, clinical attendance, medication pick-up, viral load, mental health and social support outcomes. Just to make this very clear, in many countries in sub-Saharan Africa,  differentiated HIV care includes client-led ART delivery, but this is for people stable in care who have demonstrated that they can attend clinic and that their viral load is adequate and then they are allowed to enter. But this is at the time of diagnosis.

To date, we have enrolled 26 participants in each distinct groups. We are collecting monthly data on team meetings, we are assessing their HIV clinic data by pulling records from information from HIV clinic records. We are also conducting intake and six-month survey that looks at PTSD, depression and anxiety and stigma and social support. And we are carefully conducting exit interviews for anyone stopping early, and we plan to conduct these for everybody at the completion of the study.

Briefly, I'm going to tell you about one more study that we are just starting called ART On The Go. This is a study funded by the University of Washington Center for AIDS Research, with support from the National Institute of Health.

The objective of this study is to understand what aspects of refugee movement are associated with HIV care engagement to find this linkage to care, retention to care, ART adherence, and viral load suppression, and to inform future interventions to optimize care engagement for refugee populations during times of movement.

The design of this study is to prospectively enroll a longitudinal cohort of two different groups of people. One, newly diagnosed with HIV and two, those already in care, to assess mobility and evaluate linkage of retention in care.

We're going to do this in two parts, two areas of the country. The Southwest, as well as in the North, in various refugee settlements, with adults. This will be offered in nine different study linkages.

We will be conducting monthly surveys using interactive voice response technology, over a six month period, and our aim will be to understand reasons for and characteristics of travel in the preceding month.

Just to tell you briefly, interactive voice response is a technology that allows a computer to interact with humans through the use of a voice in a keypad. So participants can use any phone, and it does not need to be a smart phone, it can be a dumb phone, and call into a IVR number. They audibly hear questions and they select their response by pushing the number on the keypad, and then they can hear the next question. Presently, we are in the final stages of study implementation and we are about to start enrolment.

Conclusions, is that understanding of mobility and the impact on HIV care engagement among refugee populations are needed to design HIV care and engagement preventions.

As social support is associated with linkage to care for refugee populations, investigators should also assess the impact of movement on social support.

Opportunities to foster social support for mobile population such as refugees, especially during periods of movement, may improve HIV care engagement.

I have a number of people I want to acknowledge, particularly mentors, Ingrid Bassett, Rochelle Orlisky and Connie Kellum, and Andrew Kambubu, as well as collaborators at Medical Teams International, UNHCR, The Uganda Ministry of Health, incredible team of research assistants in Uganda, and study participants and funders.

Agenda Item: Interventions to improve HIV care among mobile populations

KATE CLOUSE: Thank you, Kelly, for that great presentation. Great to be back again today, I am Kate Clouse from Vanderbilt University. I’m going to be talking today about interventions to improve HIV care among mobile populations. I will also be talking about an ongoing study that we have called, CareConekta, on behalf of my study team who are watching tonight in Cape Town.

Just a quick outline, I will be giving some background information and then talking broadly about interventions for mobile populations. I will give that case study of CareConekta,and then move into future directions. Let's start with some background.

Possibly, like many of you, my interest in looking at mobility really started with understanding poor engagement in HIV care. Patient attrition from HIV care is an ongoing challenge, particularly in sub-Saharan Africa. Dropping out of care, as you probably know, may lead to disease progression, transmission of virus, and drug resistance.

There are many factors that can contribute to poor attrition, and these are really well-known in the literature. They can be structural, social and personal barriers. It is also expected that mobility may impact engagement in HIV care, and the literature on that has been growing in recent years.

In sub-Saharan Africa, attrition among peripartum women is higher than the general adult population, which indicates the need for targeted interventions. This really is the area where I've been focusing my research over the past few years. In South Africa, where I mostly work, the postpartum period is the time of highest drop-off from HIV care among peripartum women.

Looking a little bit further into that, let's talk about mobility after childbirth in South Africa. There is a cultural tradition of the mother returning to her family in a rural area for help with the baby after giving birth. It is also common to leave the infant with the family in the rural area while the mother returns to the urban area to continue work.

A few years ago, in work that was supported by a K01 award supported by NIMH, we interviewed 150 peripartum women in Johannesburg to ask them about their actual travel or intended travel during this pretty brief window of being pregnant and post-partum. We found that 44 percent reported that they intended to travel, and nearly all of this travel would be after delivery. The map here shows the distribution of destinations among these 44 percent of participants who reported travel either before or after delivery. The darker colors indicate a greater proportion of respondents reporting that area. Participants reported intended travel to eight of South Africa's nine provinces and four other countries, just within a short window of time from being pregnant to the postpartum period.

I'm sure a lot of you are familiar with a lot of data challenges that we experience when working with mobile populations. Particularly, it can be difficult to assess mobility prospectively, we typically have to rely on retrospective or cross-sectional data, and this can be self reported mobility, I could be household surveys, it could be medical records, and even if you do have electronic medical records, this still may require additional manual matching.

Also, movement between facilities, which we have become known as silent transfers, and complicate our understanding of attrition and actual engagement in care. These maps, which are also from like K01 work, show movement between health facilities among those who were considered lost to follow-up from HIV care during pregnancy. Through a fairly laborious linking of lab data manually, we were able to find evidence of continued care both locally within the province, which is shown on the map on the left, and nationwide, as shown on the map on the right. This demonstrates the mobility and movement between facilities is happening, but the healthcare data system that we have are not designed to capture this movement.

I wanted to take just a minute to talk about different types of mobility. We think that there may be different modes. The first is a predictable type of mobility, which can be a regular pattern between urban and rural areas that is motivated by employment, which we know is circular movement, and something we can come to expect and also the individual can expect it.

But there is also unpredictable mobility that happens frequently, this can be a retreat when meeting social support. This often is due to TB, or now COVID, and the individual may be unable to work, eat or pay rent. Pregnancy often falls under this category, even though we might have anticipated that it would be a predictable type of mobility, often the implications of it ends up being sort of an unpredictable type of mobility.

We're thinking about intervention development or how providers approach mobility and mobility-related policies, you really need to consider both of these types of mobility.

With that introduction, let me move into interventions. I am really going to focus on interventions that have been out for a while and are available to read in the literature, because I knew that today and yesterday in this workshop, a lot of my colleagues would be talking about ongoing, current studies. That is been really exciting. I just want to give an overview of what's out there.

Cell phone interventions, obviously we have known have been around for quite a while now. These capitalized on the widespread availability of basic mobile phones worldwide. The great thing about cell phone interventions is that phones travel with the individual, which makes them ideal for a mobile population. The very earliest intervention studies showed efficacy of one-way messaging for improving retention in HIV care. Cell phone interventions are also useful as a survey tool, there is evidence of widespread implementation. As one example, mom connect in South Africa, which is a national registry for pregnant women. They can also allow for peer-to-peer interaction like group chat, and gamification, in more sophisticated methods.

Some considerations with cell phone interventions are automated messages versus human power interaction. There has been a real progression of intervention design from one-way to two-way interaction and then a case management approach. All of this facilitates a deeper level of engagement but requires increased resources and time. Questions remain about the appropriate amount of interaction by the patient, whether it should be passive or active interaction and a balance of engagement and burnout.

In terms of physical tracing, that is another type of intervention that has been explored by several different researchers. When individuals are physically searched in their homes, what this can do is allow for improved estimates of engagement in care. This usually is reserved for those who are defaulting treatment or lost to follow-up, or it is conducted on a sample of a larger population, in which to extrapolate results.

Studies have also shown that same day tracing for missed visits are shown to be effective. Considerations are that this can be very resource intensive in terms of staff time or travel. It can be a substantial effort for each individual traced. There will always be a proportion of individuals unable to be traced, and the individual may not be home due to travel or mobility.

Looking at biometric devices, we have seen several studies that have demonstrated feasibility and acceptability in multiple settings in sub-Saharan Africa for fingerprint or iris scanning devices. These can allow for ease of movement between facilities without compromising data quality. Studies have found improved data linkage through these methods.

Some considerations that the authors note is that this may require substantial capital investments, and the potential for implementation challenges related to the hardware and software failure were noted, for some inconsistencies there. Fingerprint available may vary by occupation and by age.

Multi-month dispensing is a fairly new type of intervention that may be appropriate for local populations. Some large RCTs in sub-Saharan Africa have recently demonstrated non-inferiority of six month ART dispensing versus the three month standard. Obviously, this provides a greater supply of drugs for mobile individuals. It may also result in cost savings to the healthcare facility and the individual. Some considerations are that requesting a longer supply requires planning for travel and anticipation of travel. It may only be appropriate for clinically stable individuals.

Now I'd like to move on to talking about our study that we have ongoing, it is in our R34 study funded by NIMH, and it is called CareConekta. As part of the background of how we got to develop smart phone intervention to study mobility, as I mentioned earlier, cell phones are ubiquitous and that is true in South Africa as well. Smart phone ownership has rapidly increased, as well.

Mobile health interventions have been shown to be acceptable in local populations in South Africa. The frequency of use and possession of cell phones may allow for ecological momentary assessment, EMA, which is real-time data collection of participants in their natural environment. You can use this information to develop interventions.

Earlier studies, as we have seen during his workshop yesterday and today, had demonstrated the potential for tracking mobility in the aggregate using cell phone provider data. And then some recent New York City-based studies have explored wearable GPS devices for tracking daily mobility among populations living with HIV. With that background, we developed CareConekta, which is a smart phone app for android phones. It prospectively collects individual GPS location, but in order to ensure privacy, it does not give us the exact location. It gives us a 1 kilometer fuzzy radius that is a random location within a one kilometer circle. This enables us as researchers to better understand mobility. It allows for real-time intervention to improve engagement in HIV care.

We have designed it so that it acts as a clinic Finder tool, you can see the screenshot on the right shows the location of clinics around the Cape Town area. We are first implementing this in pregnant women living with HIV in Cape Town.

I will briefly walk you through our study arms and aims. It is a RCT and we have to study arms. The control arm is the CareConekta app with the clinic finder and no additional staff contact, but our intervention arm adds contact from the study team when an individual is travelling. Travelling, in our study, means they have gone more than 50 kilometers for more than seven days.

We have two aims, the first is to characterize mobility among South African women during the peripartum period and its impact on engagement in HIV care. There are two sub-aims for this. First, to use the GPS location data from CareConekta and spatial analysis to characterize peripartum mobility within the complete observational cohort including the frequency, distance and timing The second is to assess the association between mobility and engagements in HIV care for the mother and the infant.       

Our second aim is to evaluate enhanced CareConekta as an  intervention to improve engagement in HIV care. And there are two sub-aims. The first is to assess the acceptability and feasibility of the standard and enhanced app with a focus on identifying user preference and implementation outcomes. And to evaluate the initial efficacy of the enhanced intervention, using notifications and staff contact in improving engagement in HIV care, by assessing the association between study arms, engagement, and HIV care at six month postpartum.

Where we are in the study, we started enrollment in December of 2019, and then we paused enrollment due to the national COVID response in South Africa in March 2020, after we had enrolled 68 participants. Our enrolment was paused for six months and we were able to resume enrollment in September of last year, and we completed our enrollment goal in February of this year. Follow-up is ongoing and we anticipate completing by the end of this year.

Since follow-up is continuing, I cannot share results right now. I tell you a little bit about our implementation experience so far. Certainly, running a mobility study in the middle of a global pandemic and a national lockdown is going to affect mobility. So, we are finding that.

We are also identifying some phone and GPS challenges, that I have listed here; a lack of data. Also, just a changing in health landscape, where products come into the market pretty quickly. One that was not really available, or certainly not widely available when we started this study and when we were planning for it, it is WhatsApp data, which is cell phone data that you purchase that only operates for WhatsApp.

And limited phone storage may lead to the participant to uninstall the app. Participants may change their phone settings themselves to turn off the data or the GPS. Obviously they may lose their phone. Any app that you have is going to have minimum technical requirements, whether that is an operating system, version of the phone and operating system, cell phone provider, battery life, storage space. And then, if you need to update the app, that often requires recontacting the participants.

Were finding that we have more intense staff needs then we have anticipated, particularly with following up with participants.

Moving into future directions. Implementation considerations for mobile populations. Mobility is going to happen. For lifelong care, such as ART, it is unreasonable to expect an individual is going to stay at the same facility or even a single area. So, we need to be able to support and encourage individuals to access care where it is easiest for them at that particular moment. There is no signal source of mobility data that is going to be complete. We are pretty much always going to have to triangulate between sources. Whether that is an app, phone calls, interviews, and medical records. It is pretty great human resources need required for managing mobile individuals. There are questions remaining of is it possible to automate interventions.

Of the interventions that I mentioned earlier, there is a limited evidence of these being brought to scale. And questions of how to scale these interventions to a full population.

In terms of new areas for research, there is movement away from novel apps and more to existing mainstream apps and social networks, particularly WhatsApp. Greater potential for big data to identify movement and merge patient records. There is increased capability of video sharing and cheap data to allow for this. Dynamic choice interventions allow a tailored approach to individual needs. New therapeutics have been talked about in this workshop so far, including long lasting ART and PrEP, that will be highly appropriate and important for mobile populations. And then certainly, to better understand the COVID-19 lockdown impact on HIV care and mobility.

In closing, mobile populations need differentiated patient center care that considers their specific needs at that specific time point, with the understanding that their needs likely will change again over the near term and the long term.

With that, I would like to acknowledge our study participants, our funding from NIMH, certainly, my CareConekta team at the University of Cape Town. Thank you.

Agenda Item: Human Mobility and HIV Infection – What can we learn from population-based surveys?

ANDREA LOW: Thank you Kate, for that excellent presentation. I am going to be delving a little bit into what we can learn from population-based surveys, in terms of the question of mobility and HIV infection. I know that we have talked in quite a few interesting presentations about the use of DHS data, so I am going to be talking about the PHIA project, in particular. I did want to thank Holly and Carol and the NIH for hosting this workshop.

First off, I just want to touch on a lot of what we have been discussing so far. One, about the changing patterns of migration, the climate related impact that we can anticipate. As well as some of the factors like, travel conditions, the epidemiology in both the source and the sink, as Sally was talking about yesterday. Some of the context of migration and how that impacts both the effect on treatment as well as HIV risks.

Now, I think, as Kate was discussing, some of the different ways to measure mobility and HIV, there is a real spectrum of how we measures things like this. A lot of this data can be from health facilities, especially HIV surveillance data come from antenatal clinics, but a lot of other data on the status of the epidemic, does come from a wide variety of sources, including health centers and pharmacies.

Those people who are not health seeking, who do not have the where with all to actually present for care for any of these settings. The issue is that when you're trying to make estimates of who is on treatment, who is missing out from various prevention efforts and different programs, you are really just capturing data on these different types of populations. This group really remains unknown.

In response to that, PEPFAR in the past decade, really wanted to expand beyond some of the work of the DHS and look at population HIV impact assessments, or the PHIA project. These, similar for the DHS, are household based surveys, where we go into homes and we test people for HIV after administrating a questionnaire. Originally thought to be done in about 13 sub-Saharan African countries and Haiti, we are now at the second round of implementing these surveys.

As I mentioned, it is really about how to accurately quantify the impact of all of these different resources, prevention, treatment resources, that we have been using to great extent. Particularly, in sub-Saharan Africa.

The primary outcome that we wanted to look at was HIV incidence. Now we were able to look at that because of this new assay, which is a way to see if someone was HIV infected, was infected in about the past six months. We also wanted to look at HIV prevalence at a more precise level, so that we do that at the subnational or the regional level, as well as viral load suppression, which is a viral load less than 1,000 copies per mL.  We also did a lot of other types of tests, depending on the country.

These surveys are huge, they require a lot of resources. We have to do a lot of community mobilization, to make sure that people want to participate because they were new, whereas, the DHS are usually quite old and trusted. People go into people’s homes, explain the survey, get consent, get individual interviews, and then, unlike the DHS, we actually do venous blood draw in the homes, where we then do testing, with the return of results. The rural people linkage to care and then our samples to both the central and satellite labs for verification of results as well as a number of different tests.

This data has been disseminated in a number of different settings. You can go onto our website to find out more.

The current status is that since 2015, we now completed 16 surveys in 15 countries. We are in active data collection in Mozambique and we are doing the third round in Eswatini, and we have done second-round surveys in Malawi, Uganda, Lesotho, and Zimbabwe. And as part of this, were not tested over 400,000 adults and children. This extraordinary wealth of data is part of these surveys.

Some of our primary results, if you look at some national levels, I know in several of our talks, we have really highlighted the heterogeneity of the epidemic, both at the national and sub-national levels. You could say that here where we can see the prevalence of HIV are really high in these two countries in Southern Africa. And then it actually seems to almost follow a downward trend as we moves northward, with some exceptions particularly when we think of looking at Lake Victoria, as Kate discussed yesterday.

We can also see, in terms of viral load suppression, there is a really big variability within countries. Tanzania has over 73 percent suppression in some of its regions, versus less than 50 percent in others.

When we look at national levels of incidents, as expected, it was highest in east Eswatini, it was also higher in women in every country. And it was surprisingly high in Zambia and Namibia.

In Namibia, this is one of the first countries in the world to achieve the 90/90/90 results. Part of the reason we were surprised by that finding. But overall, when you look at these, I think it reaffirms the fact that most countries are doing well in their second and third 90 results, but the real gap is in the first 90. I think it is also important to highlight that most countries are getting quite close to the first 90, but you'd still see in Tanzania, Cameroon and Cote d’lvoire, that they are falling quite far behind, particularly Code d’lvoire. West and central Africa are showing real gaps here or challenges in their epidemic response.

Here is more recent data from PEPFAR in 2020, I will note that this is before the COVID pandemic. It is based on program data. So the 73 percent viral load threshold set for epidemic control of the product of the three 90’s. You can see that it was conceded by Namibia, they are now close to 80 percent population viral load suppression. You can see that most other countries in this region have either achieved that or are close to achieving it. But Tanzania still lags and western and central Africa, including Nigeria at this point, are quite far behind.

Thinking also of heterogeneity, if you look at infectiousness, I think we talked viremia yesterday, just that measure of how many – the proportion of people in your total population who are not virally suppressed. You see these little pockets, these hotspots, across each of these countries. These are hotspots for more than 10 percent of your total population are infectious with HIV. You can see the heterogeneity across every country, quite a few around Lake Victoria, in our high burden countries, quite a few other hotspots.

And thinking more about some of the gaps. Men,  Harsha mentioned the different ways that we are trying to figure out how to reach men. If we are already falling behind in the first 90, men, is even worse in men. You can see a lot of cases, more than 10 percent difference in terms of awareness. There are number reasons for this. We've gone into them, aside from mobility, there is the stigma, and in many cases, HIV being seen as a woman's disease.

One of the things that was heartening to read was that we found that men, once they were aware, for the most part, but not universally, in most countries, there were almost likely to be on antiretroviral as women. A lot of people still accessing care once they are aware.

With a focus on mobility, the mobility questionnaire or the module, was not core in PHIA one. Only some countries included it. Here most of the countries asked have you been away from home for more than a month in the past 12 months, this exclude Namibia, who asked over the past three years. Kenya asked three nights in the past six months, and Lesotho, focused only on external mobility, with the bulk of it in South Africa.

It can make these cross-country comparisons a little bit tricky. If you look within countries, you can see that men tend to still be more mobile than women. And it tends to be concentrated amongst young people. Although, again, that varies pretty considerably, as well.

Now, keeping in mind the difference and definitions of mobility, you can see if you look broadly at mobility in HIV prevalence, you can see that there is a trend for higher prevalence in these countries. Not so much in Zambia woman and Zimbabwe, but Lesotho, in woman, that was the sort of the starkest impact in terms of higher HIV prevalence, and Tanzania, both men and woman, and Ugandan men, you see higher prevalence was stopped

Again, this might speak to the difference in definition of mobility, as well as the heterogeneity of mobile populations and the risk within the countries.

We also looked at treatment outcomes. So amongst people living with HIV, how many migrants were actually virally suppressed. You can see that this varies as well, although everyone is on the negative side of odds of viral suppression. The pooled data, about a 25 percent reduction in the odds of viral load suppression.

Just to highlight here, Lesotho, our external migrant population, over 50 percent reduction in odds of viral suppression.

In Namibia, we also included a more detailed questionnaire for migration. We were able to generate a compensate variable that incorporated not just the away from home variable, but also whether they had ever lived  outside the country, or outside of that region. We drive to this variable we called a life, migrant, or significant migrant. They did have lower viral load suppression than non-migrants, but this was all being driven by not being aware of the status. So, if you are non-migrant, they had achieved their 90-90-90, this is regardless of sex. A migrant had infected 83 percent proportion who are aware. You can see looking at a different measure of migration, where we are looking at their arrival within the community, if they arrived less than two years prior, they had only 71 percent who were aware, versus those who had arrived there longer. This really had a lot of downstream effects on viral suppression. So their viral suppression overall, was quite low. Particularly when you consider the achievement overall in Namibia.

We look at that again, across our countries. You can see here that in many countries, the migrant men are falling behind. The already lower levels that we see in men, but in some countries you are also seeing the impact on women. It is just never quite as profound.

I will note that these descriptive statistics were not incorporating age, which is a huge confounder of both migration and HIV infection and the 90-90-90. But when we do run these analyses, incorporating and adjusting for age, you tend to see these similar patterns.

We also wanted to look at risk. Here we used our recent data to look at the associations between whether someone was recently infected and the recent behaviors. You can see that someone who is also recently arrived in that community, is almost 4 times more likely to have a recent infection. That is only when you incorporate community levels of viremia into the model.

We have a viremia map here. We have it more on a focused or lower resolution area here from Namibia, and you can see that Viremia, which I found fascinating when we spoke to the Ministry colleagues, a lot of this seem to be to be tracking around the B1 highway, which goes from  South Africa to Angola. You also see this world pop data on migratory flows and how that intersects with a lot of these little pockets of infectiousness.

I will also note that thinking about some fascinating lectures we saw yesterday, around fishing and risk in these communities, and although we did not have the numbers to do extensive analyses in these populations, men who reported fishing as their occupation, were much more likely to be HIV-infected, and much less likely to be virally suppressed. This was similar in truckers. These were very high-risk populations within the migrant population. For women, these high-risk populations were the domestic workers and manufacturing workers.

We also looked at mobility and other factors in adolescent girls and young women in Lesotho. Overall, their prevalence is one of the highest in the world of 11.1 percent. This is girls age 15 to 24. You can see again, these hotspots, this is just of HIV infection, not viremia, but a lot of it is either urban or clustering along these border areas. Here what we found fascinating, was along this little hotspot, which seem to be linked to the semi-pass border, which the Ministry of health colleagues felt was significant in a potential driver, or as Frank Tanser mentioned, these corridors of transmission which might be at this point, crossing borders.

Girls who had been mobile themselves were at an almost twofold increase risk of HIV. One of the benefits of these surveys, is you can also incorporate data from other household members. We looked at partner factors, and there having worked in the past year, was significantly associated with higher HIV in female spouses, the young female spouses, and of course, migration plays a big role in employment in Lesotho. And we also found a maternal education was quite strongly protective against HIV infection in their daughters. We are currently exploring in this country, and across the fields, how mobility, parental factors, and partner factors, all intersect in the risk for HIV in young women and in young men. We hope to be able to present that data soon.

Now, Alaya Khalifa, who is a doctoral student at Columbia, has been doing a larger pooled analysis looking at transactional sex as an outcome. Building on some of the data we see here, and elsewhere, about the risk of transactional sex and HIV acquisition. You can see here that by pooling, she was able to have a sample now of over 27,000 women, and their male partners, living in the same household, where we have data on both the woman and the men. Of this 27,000, 23 percent of women were in a couple where at least one person was mobile. Two percent were both. Eleven percent where the partners mobile, and nine percent where the woman was mobile.

If we look at our outcome transactional sex, we can see here that both mobile had the highest risk of engaging in transactional sex. Women only, were a close second. When she ran the model, adjusting for confounders, you can really see it is the woman's mobility that seems to be driving this association. The partner association will adjust, is no longer really significant, and just as an aside, we are looking at food insecurity, and the role it might play in these associations, and others.

You can see here, similar to data I know that Sheri has been exploring for a long time, we also found that food insecurity was associated with higher onset of transactional sex.

Of course, when you look across countries, the patterns are pretty similar. We should also note that in Lesotho, where the migration is external, this is the only country where the partners mobility is associated with the woman's odds of transactional sex in the past year.

We also, as I mentioned, looking at food insecurity overall. How that intersects with mobility, and with different members within a household. What we found thus far is that severe food insecurity is associated with two-fold risk of HIV acquisition in women. And yet, if we are thinking about ways of medicating, climate change effects, the receipt of food support for the past three months was strongly protective against that risk. These mediators again, in support of Sheri’s work and others, a lot of these behavioral changes were associated with severe food insecurity. Including sexual violence, and high risk sexual behaviors. Having sex people without a condom, people whose status was unknown.

Finally, just thinking about some of the issues with using data from population based surveys, of course, you have to consider missing respondents. There is always the risk that the data isn’t missing at random. We can see here, our response are actually quite good, especially for something where you are testing people for HIV and asking them about very sensitive behaviors in their homes.

In some countries, especially places with a lot of mobility, like Namibia, you will see significantly big drop-offs in the household level. We see drop-offs in the male level. There really are the unknowns of some characteristics in those people who were not included. Although we do include you are eligible if you – sorry, if you slept the night before, even if you're not a household member, we aren’t capturing people living in hotels, or in barracks, or in guest houses, or that sort of thing. That is something we often think about when we are trying to interpret our data.

In conclusion, these enormous surveys, they have provided a lot of high quality actual data. A lot of this has been incorporated into governments policies, trying to target the 95 targets now. There are a number of different analyses that you can do, including mobility analyses and including looking at temporal, geospatial data. We have, especially people in this group, will be heartened to hear that the PHIA 2 surveys, Is now a standard question module. We are asking a lot on questions on internal and external migration and timing and distance. We are also looking at things that should capture multi-month dispensing. PrEP use and HIV self-testing. And I do note just for Kelli and others, that we are just launching our PHIA survey of refugee populations in Uganda. I am always very eager to talk to others to compare data.

Finally, just to acknowledge our partners, these are enormous projects, cast of thousands working on these. We have to highlight the work for the United States Centers for Disease Control and Prevention, our funder. The Ministries of Health and other government partners. And really our local staff, who go through tremendous challenges, including being chased by alligators, and canoes. Having to scale cliffs to get to the enumeration. Satanism, all sorts of strange accusations. And now, they are being accused of spreading COVID. There are just all sorts of challenges that we never predicted. I think the quality of data in our risk response rates is really reflecting the excellence of our field team. So a big thank you to them. Thank you for joining here today. Thank you.

Agenda Item: Q&A

CAROL CAMLIN:  Thank you. Andrea, and thank you to all of the panelists for this really excellent and interesting session on interventions and implications of mobility for interventions, to improve HIV care and prevention. We have until 40 past the hour. Whatever hour that may be, to all of the participants, to have a discussion, which is terrific. I so excited about that. There are some questions and answers that have been posed to panel members and some of them have been answered in the Q&A box, but I will read a couple of them, and I have a few additional questions myself, to different members of the panel.

Let me start with a question for Harsha Thirumurthy. A great presentation, thank you. Posted home self-help testing, how common was it for men who test positive, to actually go to clinics to initiate ART? That is the first question. The second part of the questions is, what were strategies in place to encourage men who test positive, to seek further HIV care, given the current difficulties with getting men involved in HIV care?

HARSHA THIRUMURTHY: Thank you for the question. It is a good question. I should note that in the two studies that we have done, where self-tests were provided to women, the focus there was primarily on facilitating partner testing and raising women's awareness in their partner testing and raising women’s awareness of their partners status. The linkage intervention was not one that we developed as part of the larger intervention itself. It is one that we have identified as an important priority area, and hence, you can see how it is something that is emphasized in the (indiscernible) study instead.

Just to say what was done, the self-test did provide information on where people could seek confirmatory testing, as well as information about the availability of ART and PrEP. That was written information that was provided along with the test kits. In both of those studies, we did not see sufficiently large numbers of men who tested positive to actually be able to say definitively, what the linkage statistics were. There was a lot of wide variation in the estimates in linkage there. Nonetheless, it is something we have identified as a high priority for self-testing interventions.

CAROL CAMLIN: Thank you. Let me move to a question to Sheri Weiser. What sorts of interdisciplinary collaborations are needed for health research at the nexus of climate change, forced displacement, and food security? And secondly, how have you developed these cross-disciplinary partnerships?

SHERI WEISER: Thank you for that excellent question. Certainly, this work does require transdisciplinary partnerships. So, moving beyond just health and social science, and also bringing in partners from earth and climate sciences, where relevant law, agriculture, and even potentially things like urban planning, and certainly, health policy and many others. We have developed some of these partnerships. For instance, in our livelihood intervention that I described in Kenya, it was a partnership between Kickstart, an agricultural organization, the Kenyan Ministry of Agriculture, and we had transdisciplinary partners, including Parsha(?), economics, and many others. We even involved the Kenyan Wildlife Society, and the hippos were eating the farmers crops. They trained our farmers, which they just do for free, on how to dig the trenches to prevent the hippos from eating their crops.

We are also trying to do this at UCSF and we  actually were just funded, pilot funding to start to launch across UC Center on Climate Health and Equity. And we are trying to develop partnerships in many places, from the earth sciences, veterinary science, all different schools from across the UC system. Stay tuned. We are working on that.

The other big piece, particularly where we are dealing with something that has such inequitable impacts, is partnerships with environmental justice organizations and affective communities to help really devise solutions that are community led and really putting equity in health at the forefront.

CAROL CAMLIN: Thank you. Let me pose a question to Kate Clouse. You mentioned a lot of the challenges related to mobile phone based interventions that you have been facing with the CareConekta intervention. I'm curious about whether you can share nuggets of wisdom for those who were interested in developing in health interventions for mobile populations.

What would be the main recommendation you would make if you could start over and redesign CareConekta,  given what you've learned through your implementation research conducted as you have been conducting the study. What would you say?

KATE CLOUSE: Thanks, Carol. The study is still ongoing and so we are constantly learning every day. But I would say the landscape is shifting quickly. The study has been going for two years. Obviously, there are about two or so years planning before that. In the four to five-year period, the in-health landscape has changed quickly, and it is constantly changing. One is just to expect that and try to be thinking as forward and as future planning as you can be.  We are still finding things that are surprising as all of the time. I brought up that example about the type of data, the WhatsApp data, that we did not know about when we first started this.

I think my own personal and probably my team as well, our thoughts are shifting away from developing novel apps, unique apps, itself, to really going to the widely available apps that everybody is already using. Meeting people where they are. So if that is WhatsApp, if it Facebook, it is usually WhatsApp in South Africa, but for whatever setting you are in, using what is available that everybody is already using, starting with that, as opposed to trying to introduce something totally new.

CAROL CAMLIN: Kelli O'Laughlin, it was striking to me the finding around really protective effects of social support for refugees. I am just curious about what your thoughts are around the implications of social support among refugee networks as a potential unit of intervention, if we think about social networks among refugee populations? What would you envision being a potential way to kind of harness that positive finding?

KELLI O'LAUGHLIN: Thank you, Carol. That is a really great question and that is pretty much where our team is right now, trying to figure out next steps. I think that there early CCLAD, is one opportunity for that, I am hoping that I will have the opportunity to do that in a multi-site trial setting, because currently it is a pilot.

But to basically use groups of newly diagnosed refugees and put them together based on where they live, the language that they are speak, their cultural background, if that is their preference, has really been exciting, in terms of fostering relationships.

I will even give you an example. I was really disappointed that one person left the group, they were not showing up.  I was like, why is that happening? In the exit interview, they had moved away, but they were still WhatsApping with their group all of the time. There were still getting that social support, they were being connected, they were not actually attending the meetings.

I think it is exciting and I'm hoping to basically build on the early community-based groups for newly diagnosed people with HIV.

CAROL CAMLIN: Thank you, very much. I had a question for Sheri Weiser, but it could be extended to anyone in the panel as well. Maybe I will ask it of Sheri, and then ask if anyone else has anything to add, because it is a hard question, but given the conversations that we had yesterday and today about the dynamic nature of mobility and its temporal dimension, given that there are different needs of mobile populations at different stages in the migration process, what role do structural interventions play, including agricultural interventions, play in addressing the needs of mobile populations?

You know, we think of structural interventions as being longer-term interventions. And sometimes as being upstream interventions to address social determinants. Sometimes migration of mobility is not predictable and sometimes people lack land or legal status and the access to services that legal status can provide.

Are there ways for us to be thinking about temporality of interventions that would allow us to more flexibly address the temporal nature of mobility and the needs of people at different stages in the process, both during migration and perhaps after, if not before?

SHERI WEISER: That is a great question, Carol. In the short-term, when you're dealing with people who just recently migrated and are not near any form of permanent settlement, that would not be a time where we were thinking about livelihood interventions, but you certainly would be thinking about things like addressing their basic needs and other types of interventions like food support, making sure that they have a safe place to stay.

Whenever you are dealing with people who had recently migrated, in addition to structural interventions, we've heard so much about the need to have some sort of mental health and social support integrated with that. I would view as a multimodal intervention to address the sort of multifaceted experiences that they are having. And then I think once people do settle, there is a lot of innovated interventions I have seen around recent migrant communities, that are doing really interesting livelihood programs or food and nutrition programs, incorporating culturally tailored cooking classes to bring in healthful foods. For instance, like the urban gardening intervention that I talked about earlier. The partner that we worked with, had this sort of tailored cooking class to the immigrants from a certain region in Latin America, to ensure that they would utilize the vegetables that they grew.  Because before, when they first started it, they were not utilizing the vegetables until they got culturally appropriate recipes. So I think cultural tailoring and then livelihood interventions can be very helpful in a sort of longer transition base.

CAROL CAMLIN: Thank you, anybody else have anything to add to that tough question that I just posed?

LOW: Specifically, about southern Africa, one thing that I have been exploring with some of our partners, is why people don’t use cross-border migrants, don’t access healthcare in South Africa. I think one of the structural interventions really needs to be a focus on almost an international universal healthcare program.

I know that there have been laws in South Africa that do support that. But, migrants are either not aware that they are legally allowed to access that care or they are illegally, so that they are not certain about their rights there. I think in many cases, they are still going to be stigmatized if they do try to access care in many places in South Africa. I think there has to be a broader idea both of access to treatment and prevention services, as well as ideally, if it can be done with confidentiality in mind, the transfer of electronic health records. Ways to maintain some kind of idea of how many people are falling off of treatment across all of these different borders.

HARSHA THIRUMURTHY: I will add just one very brief point. I think, when we are talking about climate change, there is also I think, a greater need to look at the potential for crop and weather insurance, and the extent to which that might actually prevent some migration from happening, that is really being driven by an economic need that people have. I think if you go very far upstream, I would put that into the category of some interventions as well.

CAROL CAMLIN: Thank you all so much. We also have more time after our break to discuss these issues in more depth. We will be welcoming back many of the panelists from yesterday. I think that will lead to some very fruitful conversation to integrate hopefully, some of the things we talked about yesterday.

I am going to now call us to a five minute break, and have us resume in five minutes to then begin session five, the panel discussion on highlights and future directions that will begin in five minutes.

(Break)

Session 5: Panel Discussion – Highlights and Future Directions

HOLLY CAMPBELL: Good afternoon and thank you to the panelists and attendees. My name is Holly, I will be the moderator. Today, we have a great lineup of speakers. We had a slide there to show you who they are. Could I get the slide up again? Thank you.

So, you have been introduced to all of these speakers. I will just briefly tell you their names and encourage you to look at their bios in the bio document. At the top left of the slide is Sally Blower, and below her is Dr. Carol Camlin, and next to her is Mr. Bernal Cruz, and Dr. Fred Ssewamala, Dr. Frank Tanser, Dr. Mark Van Ommeren, and Dr. Sheri Weiser.

So, we have a great panel lined up. The title is Highlights and Future Directions. Our goal is that whatever your role in the field of mobility, that you will come away with something, a new idea, a fresh perspective, a new strategy that you can apply to be more successful with your work.

I have a few starter questions to get to the panel going. Then I have some specific ones. I would like to engender an atmosphere of collaboration, and discussion among the panelists. I think that is the most interesting type of panel. When you ask each other questions, and maybe even have controversial things to say. I hope that will happen.

My first question is very similar to one that was just posed to Sheri Weiser in the last session, with regards to collaborations. I think I will ask this of Mark van Ommeren because the nature of his work being very complex and requiring lots of collaboration.

The question is, the nature of collaborating is essential. I wanted to get at how do you form the right collaborations? If you go back to Sheri’s answer, it was quite extensive, the breadth of the type of collaborators she is involving in her research. Just getting at, I think my question was, after having been at a workshop for the last two days, what collaborator do you now think you need on your team? And you have not been here the whole conference, so we may ask someone else as well. Is there a field that is not represented on your team, that would enhance it?

MAARK VAN OMMEREN: Well, yes. I am quite clear as to what collaborator are missing from my team, the team I work with, it is a health economist. It is quite clear that we have many questions around cost-effectiveness, cost-benefit, et cetera. To do this well, we need - I know something about health economics, I have been called for many health economic publications, but still I think we need someone who is really an expert on that. I think because we need the answers to these questions, and what we find is that very often, we publish, and we promise at the beginning as well, that we will do an efficacy, and cost-effectiveness, and a process of evaluation. And we definitely do the other too quickly, and then cost-effectiveness comes out two years later as we struggle with what was good. We do not get the same value. That is absolutely what is missing in terms of skill.

But in terms of type of collaborator, that is a broader question, and I think one likes to collaborate where there is easy trust. That is for sure. There is a lot to say but a lot of interpersonal skills that are beyond the tick marks of qualifications.

HOLLY CAMPBELL: Thank you. Would anyone else like – Dr. Fred. Thank you.

FRED SSEWMAMALA: I think when you talk about collaboration, then we are thinking about collaborators that we are working with as researchers. We should also think about collaborators that are giving us the data or are connecting us in the field.

For example, people with lived experiences. I work in Uganda, I have done this work for the last 20 years. I think that more than my economic friends, public health friends, social friends, from the global North, the people that are going to be extremely helpful and we invite them to go with me, to conferences, to go with me to presentations, and people with lived experiences. Because they help you ask the right questions. Even before you go into pieces, I think that is a group, we tend to kind of, ignore in a way, and we think it would be better collaborators, that they will give us levels of support, and that kind of stuff. I think if we are going to create meaningful change in the work that we do, we need to make sure that our collaborators are empowered and they do not think that they are simply collaborators. They have to feel that they are part of the implementation. That they are part of the people really posing the right questions for us.

I know that the training that we have, we tend to value our land friends, if I may use that, but I think the communities that we work with, I think are the best collaborative, because even when we are going to be able to disseminate the work. Frankly, I was surprised that many of us did not bring our collaborators on board to be co-presenters. Because this is Zoom. If it Zoom, if we can’t invite them over Zoom, then for sure we can’t invite them next year in Canada for the International AIDS Conference, because that is more money.

I think those are the things that we need to think about when we talk about collaboration. Is it meaningful collaboration or are we simply paying service I terms of collaboration?

HOLLY CAMPBELL: Thank you. Would anyone else like to speak to this?

SHERI WEISER: I want to build upon what Fred is saying, and just an acknowledgement that we are fortunately, at a point in time where there is a lot of movement in research and community-based partnerships to break down those typical power structures that are in play. These are cross-sectorial partnerships. Not just bringing community members and community based groups and tell them what intervention we are recommending, and have them help us. But have them really at every stage of the research, informing our thinking, involved in every stage of the process. Involved in designing the right intervention for a group in implementing the research and disseminating the results.

HOLLY CAMPBELL: Great, thank you. Sally.

SALLY BLOWER: I had one thing to say, Holly, what I think we need is people from big data science involved and people in bioinformatics, and people from physics. Because People are thinking about networks now, and people have been thinking about networks for a long time, but they are all social networks. People have come in from social sciences, and that sort of thinking about networks has come in.

People in physics, think about networks in a totally different way, and they have different methodologies and different techniques. Those are the people I think to bring in.

HOLLY CAMPBELL: Great. I have another question that kind of overlaps with what we just discussed. I don’t know how much value it will add. It is a bit similar.

I recently heard Dr. Anthony Fauci of the National Institute of Allergy and Infectious Disease, suggest that high income countries should invest in the resources necessary for low to middle income countries to begin production of COVID-19 vaccines.

So, rather than producing it in high income countries, give the resources and let folks in low income countries make their own vaccines.

I was thinking in terms of refugee camps, but this might apply to – I don’t know how it would apply to other populations who are mobile. But I was thinking, have other attempts been made to empower long-term residents of refugee camps in a type of capacity building? To shift ownership of something to do with healthcare? I am looking at Mark.

MAARK VAN OMMEREN: Well, we see at times, refugees being given the capacity to provide mental health interventions.  Earlier we talked about this task shifting in psychological interventions. That can go also to refugees themselves to provide care.

So, of course, that is a very different nature than Dr. Fauci’s example. But I think certainly we want to move to a model where support is not so externally driven, right? I think that is the higher point. I think that there have been many examples.

One of the reasons why I always do so much community participation, and that is a big theme in this conference, is it is not just about participating, it is about empowerment, and having people being able to do things for themselves and having responsibility.

HOLLY CAMPBELL: Does anyone else want to contribute to that question?

BERNAL CRUZ: I can contribute. I missed out on a large part of today’s conference, but I am glad to hear that community participation and involvement has been a big topic today. It sounds like it was yesterday, as well. I had the pleasure of working with fieldworkers, they were offering mental health and psychosocial support to arriving migrants from Venezuela. Really, the task at hand is to empower and educate those arriving to basically be agents of resources and be able to promote wellness, in general. Emotional wellness being a big factor of that. Implementing  very specific adaptable interventions that can be implemented in very remote areas. Including, other things, mental health, psychological first aid, that sort of thing.

I think the big impact that I walk away with understanding, is not only to the communities begin to feel more self-sufficient, but the participants who were involved, actually have a great more buy into what is being implemented, because those in the position of providing the service can certainly identify with them. Right?

This goes back to the idea of collaboration and intervention, by the people that you know, and you trust, you build relationships. Which in turn, actually can be conducive to forming even greater relationships, and going back to what Dr. Frank was saying. Figuring out some ways to maybe incorporate those who have been affected involved, and pursuing the results in subsequent research, for instance.

HOLLY CAMPBELL: Thank you. I have another general question. If you could immediately, and without the constraints of having funding, implement one game changing innovation in the context of your work, what would it be, and why? Frank, are you up to answering that? Should I repeat it?

FRANK TANSER: I will give it a go. I think the game changing intervention is going to be around digital connectivity. Not one thing in particular, but the digital connectivity M-health, having one key database with a unique identifier, where people can link, where migrants can link from one health center to another.

I think once that is in place, things will change really quickly for migrants. Where is now, it is so difficult to change locations, start all over again, in many cases. I think, that for me, is the one thing that will make a difference.

HOLLY CAMPBELL: Thank you. Anyone else?

SHERI WEISER: I will answer that from a different perspective, which is, changing the incentives and, I think, Paul and Mark had mentioned earlier in their talks, the incentive structures of funding bodies to enable more transdisciplinary research to really happen because those innovative solutions really do require a lot more cross talk than is happening.

As an example, in climate and health, you have your climate funders. The climate solutions are not being informed by health. And then you have health solutions, may be related to migration, that may not take into account all the climactic factors and thinking about environmental sustainability when you promote an intervention.

Even related to NIH, I recently learned that NIH only funded around $9 million last year in direct climate and health funding, even though there was more related funding, that may go up to $100 million this year. But that contrast will say, one to three billion in major health issues like HIV and diabetes. We are not even incentivized to bring health researchers to the table to address some of the biggest health crises of our time.

I think, if we could change funding structures and incentives for researchers, that would help with innovative solutions.

HOLLY CAMPBELL: Great suggestion, I will bring that back to my boss.

SSEWAMALA: Holly, I think, probably it will be hard for us to get a (indiscernible) just one. If we look at the cut through work we saw here, is that poverty still plays a very big role on economic stability. The issue is that what is included in our recommendations as a control, and that normally, what we are trying to say is that we know that that is the big elephant in the room and it has one. We go around everything else and then we do not really want to tackle that.

When you tackle issues of poverty, the question is whether it is at a 1 out of 34, the question is is that sustainable. Even they don’t give you more than two years to establish it, it is inadequate for two years, if you come back and say, that it is not sustainable, but you did not give it a chance.

If money was there, I think some of the structural barriers, which really, poverty being the key, will definitely need to be addressed. If we think about women in sex work, if you think about some of the mobile populations from the Latin America coming up into the US, you think about some of the refugee situations going on, not all of them are in poverty. But you think about issues of war and some of that.

There's something about poverty that is really making it hard for some interventions to take effect. Plus, our need for sustainable interventions within a very short period of time.

I am always reminded that I come from Uganda. Uganda has been independent for the last 50 years, we are not sustainable. I'm not saying that interventions should take 50 years to get sustainable, but at least we should give them a chance to grow, and then start asking about the sustainability question. We ask it very prematurely. That is my view.

HOLLY CAMPBELL: Thank you. I was going to talk about the heterogeneity of the populations you all study. I was thinking about Carol's talk and the genderization. I was thinking to what degree do your data lend to the assumption that interventions should be tailored specifically or can they be in the sense, trance diagnostic – that is terminology for mental health. I think you know what I mean.

CAROL CAMLIN: Thank you for that question. I have been thinking about a lot of the questions related to, not only the talks today, but yesterday, in thinking in my own mind about how I would draw it altogether.

I remember that phrase from PEPFAR, know your epidemic. I am thinking we should be applying that same thoughts to the study of mobility. I think there is a lot to be said for knowing the mobility in the context of mobility in particular populations and in particular settings.

Knowing it through a lot of means, knowing it at a high level in terms of drawing upon the birds eye view that mathematical modelers can provide us, given access to good data. As well as, the kinds of knowledge that we gain from working, as Dr. Fred said, in partnership with people who are experiencing these phenomena in the communities in which they are working through participatory research methods, like the one that Mark Padilla was talking about yesterday, through other means of working in a participatory way.

I think, in that process, we are likely to target. I think, we will need to be flexible and responsive to the dynamic nature of mobility in particular places in time.

There are aspects of women's mobility that lead us to think about particular interventions that would be especially important for women in certain settings in sub-Saharan Africa. There would be other kinds of interventions needed for specifically refugee populations.

So, it is the forces that are driving mobility, it is a context in which mobility is taking place, and it is also where people are in the temporal dimension of mobility in the migration process, that implies a certain set of solutions.

I actually feel like we need to look more closely and be smarter about planning interventions. Which, actually, would lead me to another question about, that I hope we will talk about it at some point, but not right now. I will put a placeholder on it, which is thinking about the whole notion of hotspot targeting, and how does that need to be redefined and rethought of in light of presentations that were offered yesterday. And in thinking about the ways that people are thinking about interventions today.

HOLLY CAMPBELL: I had a question for Sally along those lines. I wanted to ask her what theoretical concept will be the biggest barrier to changing the paradigm of how HIV transmission happens among mobile populations?

SALLY BLOWER: One thing, I said very briefly, is that the HIV transmission model structure 99 percent of people are using now in modeling transmission models is very old. They were developed 30 years ago, for MSN communities in resource rich countries.

So, they were based on looking at behavioral, sexual heterogeneity, and not mobility. Those were taken and used for sub-Saharan Africa.

The epidemic is generalized, so it is not in a high-risk community, it is spread throughout the countries, as everybody knows, and desecrated mobility. All of the models just were plunked down and the results are obvious.

I mean, you treat the highest prevalence places, you will get the greatest reduction. And that though, is a function that the models do not have mobility in. So, I think to move beyond it, the people need to look, throw away transmission models. I spent my career working on it for decades. So I don’t say that lightly. But start fresh, start with new thinking.

HOLLY CAMPBELL: Thank you. I would like to open it up and see if there are the questions for panelists. No pressure, if you do, that would be fantastic.

SSEWAMALA: I have a question for the panelists. By the way, this was an excellent workshop. I am coming to the full notion of mobility, given the work that I do with women engaged in sex work. But over the two years, I saw that the definition was very fluid. I'm trying to figure out where to place school going students who transition from high school and then to universities in different places, where would we place this rural/urban migration young girls, who are moving out of rural areas, and ending up on the street. Where do we place them because they were not represented here?  Do they fit in our definition of mobility or would they fit in something different?

CAROL CAMLIN: I think, Dr. Fred, that you have helpfully pointed out to all of us, that there are so many migrant populations that are easy to recognize, then there are other migrant populations that are not easy to recognize. And I think, often about, not only the kids who end up in the streets when they were migrating to urban areas, but also, girls who were said to be house helps, and was faced vulnerabilities in that context. They really lack mobility, once they are in that destination. And they really lack access to services.

Then we have adolescents living with HIV/AIDS who are in boarding schools, and they are not able to access that phone-based intervention, because no phones are allowed in schools.

There are all kinds of particularities to the needs of different mobile populations. Again, for me, it circles back to the question of knowing your epidemic, knowing her mobility, which also means knowing the mobile populations, the typologies and mobility in place and space and the time, in order to think about solutions.

HOLLY CAMPBELL: I was thinking about the IOM’s definition of migrants because I think there total, as of a couple of years ago, it was 272 million people, and that includes refugees, IDP's, and so on. It is a much larger number than the number that Mark presented today, the 79.6 million, I think.

It includes international students, but I do not know if it includes - I do not think it includes what you described, which is a much more subtle form of migrant movement.

CAROL CAMLIN: Technically, they are internal migrants if they are making moves. In Kenya, it would be defined by the basis of county or sub-county. Uganda would be a district or subdistrict. Even down to the parish level, where we could define internal migration in different ways.

But the question is, yes, do we - you challenge us to think about being inclusive of mobile populations in the ways that we think about mobility.

BERNAL CRUZ: We go back also to the issue of context, right? Internal migration within one country for instance, my own country of Guatemala, happens mostly because of job opportunity or people trying to explore different industry. But you look at other countries where internal migration is happening due to environmental issues and, or, conflict, internal conflict, so on.

They might still be under the same definitions of internal migrant, but, the drive, the causation, the root cause of the actual migration itself, lends itself to a number of different kinds of questions and vulnerabilities subsequently.

So we can continue to explore topologies and the definitions in which we want to put these different people who are on the spectrum of mobilization. But the interventions and the way that we conceptualize it, the questions that are posed, given the condition are entirely different, regardless of how we want to categorize them.

SHERI WEISER: I will add that the rural to urban migration has a whole host of other things that we have to take into account, in terms of chronic disease risk, and lifestyle changes, that compound the other usual issues that would be in other forms of migration.

SALLY BLOWER: One thing that we have not talked about in this workshop that I find really interesting, is about mode of transport. Because talking, as presumably everybody knows, distance away from healthcare facility affects accessing treatment and deterrence to treatment.

One of the studies that we did recently was modelling, if you had bicycles, versus if you had to walk. And showed how the coverage of treatment of the population, HIV infected population, would dramatically increase.

I was wondering if anyone in the real world was doing those kinds of studies. Looking at transportation and how it could affect treatment access and adherence. Maybe not.

HOLLY CAMPBELL: That would require some research on my part.

CAROL CAMLIN: Say that from qualitative work among female market traders and other women who are migrating and highly mobile, this doesn't precisely answer your question, Sally, I don’t think, but, there are issues around using public transportation and disclosure of status.

Specifically, women will talk about how the pills shake when they are travelling. Inadvertently, it discloses they are HIV positive and the other people that they are traveling with. They have different methods of - or if someone does not necessarily want to disclose their status with the people they are traveling with, so they disguise their drugs. They take drugs out of certain packages and put them into other packages. There are many ways that people are managing the processes of disclosure that are aligned to their mode of transportation. Which potentially could potentially affect their adherence. There are important particularities, and granularity to be attentive to when we are thinking about interventions.

SHERI WEISER: I have a question for some of the panelists. I think Mark, you raised this today, and a few people raised it yesterday. This notion around intersectional stigma, which there is a huge field in HIV research, the intersectional stigma related to HIV status and sex work status, and race, et cetera.

But what I felt was very interesting and unique that came up here, was the stigma around refugee status and migrant status. I was wondering if any of you have been doing work on how migrant or refugee related stigma sort of intersects with these other forms of stigma to influence care and treatment and HIV prevention?

MAARK VAN OMMEREN: I have not done that research, but I was yesterday in a conversation with colleagues on TB, and we talked about when someone has TB and has a mental health issue, how that is an absolute double hit in terms of their capacity to function in the community, and so, it is a social protection issue and a double stigma issue. Double discrimination issue.

I think in general, it is the same thing most likely applies. If you are not well and have mental health and TB, being a refugee and having HIV. They are two huge hits. I think it justifies prioritizing, and it is difficult. It is difficult to prioritize, but if you want to say, okay, I want to prioritize all people with mental health problems all people with HIV, all of those vulnerable groups, you have prioritized, in some places, a majority of people. But once you realize that people have a double whammy or a triple whammy, those people are so vulnerable. They must somehow get priority attention without hopefully creating some kind of stigmatized problem by singling them out. It is clear that when people have so many things going on in their lives, at the same time, it is a very difficult to survive and also very difficult to access services.

CAROL CAMLIN: I have also been thinking a lot about this, Sheri, this question, if we are drawing attention to mobile populations as populations that have a particular high need for HIV prevention and care, how do we go about doing that without inadvertently facilitating stigma against migrant and mobile populations?

I feel like there are some opportunities to address this through very community engaged research. So, I mean, one thing we did in the mobility study, is have annual community meetings involving different stakeholder groups where we had community discussions, engaging the community in the research process, and also feeding back findings and testing our interpretation, and inviting community members to participate in the interpretation of the findings and co-create the interpretation of the scientific knowledge from the study. Which also led to discussions about potential interventions for mobile populations.

One thing we were careful to do, from the very beginning, was to engage the community in talking about what the benefits are of mobility to their community. What are the things that migrants bring to the community? How have households in his community benefited from being able to have a migrant going to and from, Kampala, wherever it is, so that people are reminded of all the positive economic and social benefits, you know? Gets migrants school fees paid. Kids health is being sustained. A lot of great things are happening because of mobility in these communities.

I felt like it was always important to keep that value and dignity of mobile people centered. Not only thinking about, we have to talk about human rights of migrants, not only that, but also talking about the value of migrants. The great strength that they bring to communities.

HOLLY CAMPBELL: We were technically supposed to have a Q &A session already. However, the audience was just reminded to give us questions. There are none at this time. So if you want, for another five minutes or so, we can continue with this type of discussion. Dr. Fred, did you have something you wanted to raise? No?

SHERI WEISER: I wanted to mention a point that Mark Padilla typed in. He mentions that he has published ethnographic research on the intersection of deportee stigmatization and addiction stigma in the Dominican Republic, where we argue that it contributes to HIV addiction Syndemic.

HOLLY CAMPBELL: I was wondering, after hearing Mark Padilla's talk yesterday, whether or not you were motivated to change how you engage with the population you serve going forward? We talked about this in general, already. But his approach is very different. I was wondering how - to what extent you might take on his methodology or approach in general?

BERNAL CRUZ: I was not actually at Mark's presentation, but I had the pleasure of talking to him at length in preparation for the panel discussion. My new future in research has got me nervous, because frankly, I have to formulate some questions. It is that exciting part where you have questions that you really don't know how to formulate them. I have had to submit some papers for revision, already. I am a big fan of course, of participatory based research and community based initiatives, as you heard me say yesterday. In fact, I just sent a document arguing for why I should be allowed to engage now in participatory based research involving visual methods, literally, like going to some villages in Guatemala, and doing some film. That would be co-created and really the basis of collaborative knowledge, creation, to promote knowledge and to promote what I am preoccupied with, which is retention of youth in places like Guatemala. Ultimately, in the well-being of the community at large.

HOLLY CAMPBELL: Thank you. And Sally, do you have your hand raised?

SALLY BLOWER: I have a controversial question for maybe everyone on the panel and you a. We talked a lot about hot spots, and a lot of people, Kate, and Frank, and I, and other people, have pointed out problems with using hot spot targeting. That UNAIDS strategy at the moment. Does anyone have any ideas how we can change UNAIDS strategy? That is my question and nobody wants to answer, I bet.

SSEWMAMALA: When you said change a strategy, which means you don’t look at the previous areas which have high prevalence and incidence rates, but you just go nationwide. I am just trying to understand the question.

SALLY BLOWER: Let me be clear about it. Sorry.

We have been talking about mobility, and Frank talked a lot about mobility, and Kate, and the genetics that they have. It shows all kinds of non-intuitive things.

So, I'm saying, if we bring in mobility data, and look at that with the prevalence data, we would be able to develop different targeting strategies. Like what is the source? Which is the sync? Which ones are linked? With our analyses, we found that there are risk hubs, and there is inflow risk hubs, some of them are outflow. So they are putting out risk. Some of them are in flow.

Those are low prevalence areas, that are actually very susceptible because people are either coming in who are likely to be HIV infected, or people who live there are going out and getting infected.

So, I think there will be a more sophisticated way of looking at targeting strategies if people consider mobility, as we all agree in this workshop, as well as HIV prevalence. Putting it together. Was that clear?

FRED SSEWMAMALA: I think that is a great approach and that was very clear. I think that you have to think about this in terms of prevention, I think there is care and support. So if you are going to think about prevention, then frankly, we shouldn’t just be focusing on – that really addresses your question of combining the two together. There is outflow and then there is inflow.

If you are talking about prevention, then even those known hot spots, we have to think about the outflows. If you think about inflows and its care and support, then you need to handle that.

I think your question is the right question. (Laughs) Seriously, how do we think about both prevention and then care and support for those who are already living with the disease. That is how I would look at it. It is the entire continuum. We cannot just say let’s focus on this and this alone, we have to think about the entire continuum.

HOLLY CAMPBELL: Well, thank you. I am sorry we have to cut it short. It is time to end the workshop. Thank you all so much, this was a great panel, and it was wonderful to hear from all of you. Thank you to all of the other participants who gave talks. I hope you enjoyed the workshop. We will be in touch about your slides. And, the video of the workshop will be up at some point.

Thank you, and thanks to the audience as well, have a great day.

(Whereupon, the workshop adjourned.)