 Good afternoon everybody, I think we are on time. You're very welcome to this session on country integration stories, we've got four quite fascinatingly different cases, there isn't really a much of a common thread through them. We're all in some way relating some kind of lesson around integration. Thanks, particularly to all the four speakers who all uploaded their presentations on time. We've got four presentations today, Henry is going to talk to us about their experience in Tanzania with integration with gen expert machines, which I know is a huge huge interest for lots and lots of people, particularly TV program around the place and the like We've got a very interesting story to tell about integration of DHS to with, with WhatsApp, the rapid probe. Sessions which, which touch on an area which we probably, it's a little bit new for us in DHS to DHS to is always been historically about routine data and routine systems. The whole area of emergency response to better. We've got a presentation I know niluka is being joined by a colleague remotely as well for part of it. And then we have, it says Adrian here, but Adrian is not here. We have a whole team with me will, I think, lead us off and introduce us to her co presenters as she goes along. If you want to stand here. If you have the microphone and you can just stand if you like to move up and down that's also good, but we have a portable mic you can just pick up and go. All right, other than that. I'm keeping we're going to try to be a little bit strict on we try to keep it to 20 minutes. If we do that we know we're going to have a good bit of space at the end, where we can have a little bit more discussion and interrogation of the speakers. All right, so everybody clear. Henry, are you happy to kick us off. Cool. Okay, we need to find your presentation. Which is your. Are you going to start. It's here. Okay, so everyone. Right. Do I never get through this. It's on. I can just. So, hello everyone. My name is Henry Calisti, I will be presenting one of our case on integration between the HS2 TB and lipos tracker with the gene expert machine. Physically, of course, this, you might be aware of this that T tuberculosis is one of the highly infectious disease. We're about coat of the word population is actually infected, though, out of those who are infected, it's five to 10% of the people who actually developed to the TB disease. This is as per the WHO findings. TB is preventable and uncurable. Usually it is treated using antibiotic. However, it can be fatal in case the treatment are not done on time. So the testing of the TB is usually done by a various approach. But one of the crucial approach that is being used is the gene expert machine, which is the rapid molecular test. And the good thing about this test is its capability of also detecting resistance to antibiotic in particular if I'm seeing, and just within two hours. The issue that we were actually working on is the fact that despite the ability of the expert to be able to test as fast as possible. But still an issue has been on having accurate timely and complete diagnostic and logistic data from this laboratory machines. And the reason behind this is of course gene experts machine mostly are installed in a particular designated facilities. And usually they are less than the actual number of the facilities that are in place. So you may find one gene expert saving multiple facilities. That is, of course, an ideal situation, meaning that cases of TB or samples of TB are collected from one facility probably transported to another. Facility for testing and then the results being recorded in. I mean, the result being also transferred or transported back to the particular facility which was requesting. So here that is where the main problem starts whereby health workers are then forced to manually transfer this data from the gene expert machines. To the particular electronic system. In our case it is DHS to it's here. So this lead to the issues with quality. Such as receiving incomplete in accurate and usually late data for the patients. While definitely once someone is diagnosed with TB must start treatment immediately because any delay can of course result into more transmission of the particular disease. So, our main response beat was to try to explore an alternative approach. And this is because of course it doesn't mean that there is no approach to linkage between gene expert machine and other system. But probably the one that were in place they were somehow facing some other challenges like economic of scale issues like private local context, etc. So the need was to come up with an alternative approach that could of course help to address all of this. And what we did is to engage by using the participatory design approach where we were lucky enough to encounter. This name re or Tifa project who among their objective we are also to establish the linkage between. The gene expert machine and the and the the case based system so that they can. First track the issue of getting results so while we were trying to find out a way we can do such a proof of concept and then the opportunity to come so with the little resources that we had we collaborated so closely with the laboratory people and the national TB program from Tanzania. And our case was to deal with to start with the 10 color models gene expert machine, so four of them. What we wanted to achieve with of course to have one a situation where as per the diagram, you can see, usually the usual setting is you have the gene expert machine, which normally comes with the gx software installed in a particular machine. And then you have the electronic system in our case which is case based system that is used to register the the tb cases from the facility. So the concern is how then you will be able to obtain this data from from these machines which are in the facilities. The situation is of course they will have to be connected someone to the router, where they can be accessed online, but the issue is how then you will be able to read this result from them, and being able to push the result and link them to the community case in the in the DHS to case based the system. So what we managed to do is of course to come up with an interfacing adapter or may call it a system that you can of course either install it direct within the same machine that is having the gene expert machine installed or you may install it in the in the DHS to instance, the one that is having the case based system, or it can also be an independent system that is aiming to pull this result from those multiple gene expert machine and send them to the central DHS to instance. So after managing to do this, the next thing was of course to make sure they are linked with the particular case within the DHS to system. So, simply the features is the communication is we relied on the standard the architectures and protocols to make sure that it is generic enough to be able to be adapted in any situation. So we are actually using the no communication protocol to SPIP. And as I said, you can install in the same or separate machines. It has a very simple UI that will allow the user to be able to do this initial configuration for the linkage without a need for technical expertise. And then once configured, the adapter will be able to start transferring those approved results. Of course, it supports both automatic and on demand use for the transfer of those results. The technology used, of course, make it all supported to be purely open source. And the good thing is also it is multi platform based on server client texture. The pushing of the data is definitely independent of the external I mean of the adapted self. And the user will be able to either log in straight to the particular interface, but they, they are also able to log in directly by using the, the, the external application which is linked to for instance, in this case, the same access that you would use with the DHS to instance, you can use it to do to access the adapter and then be able to push the data accordingly. So, this is just a snapshot of the simple UI that is used. Once logging, you can connect and then after connecting the thinking can start. Of course, the appropriate status of whether it is not safety or it is already saying that there is any failure, which is a very simple interface for any user who is even non technical to be able to maintain or manage. So, they synchronize the results. And this is one of the very crucial parts apart from just the linkage itself is the linkage of the results itself again to the appropriate case within the DHS to in our case DHS to it here. Basically, once the results are moved, they are linked to the appropriate case, and then the user will be able to find the results in the appropriate tracker stage. For example, this is the tracker which we are using to keep the case for TB, and that is the stage for results in particular the DST results. So the results that you are seeing there are actually automatically pushed from the gene expert machine. So this is one of the approach that can help to solve those issues that I mentioned about the accuracy, but also to have the results right in time. So, among the achievements that I've already highlighted, technically, but the most important thing is that ability to come up with that interface for transferring the data and the good thing it doesn't only support the diagnostic data. It also support the logistic data for maintenance of the gene expert machine. And what we are much more happy about is after this proof of concept is the opportunity we have ahead to engage the most stakeholders so that we can scale up and connect all gene expert machine. So as to strengthen the TB program, MND through accurate timely and complete diagnostic and logistic data, as well as maintenance of the gene expert machine through the timely intervention. So our very future thing that we wanted to accomplish so as to have a full scale solution is to have a standard centralized dashboard within the HS2 that can be used as a standard method as a package, let's say, to support the linkage between gene expert and the HS2 so that the testing data as well as the logistic data from the machines to be able to be managed within same dashboard. Of course, as well as supporting other technology, the serial and FTP as you may encounter some other gene expert machine using other old technology, and this can be an alternative. Thank you for this. There's no link in the slide. Yes, of course. This thing definitely as I mean the, the one that I've showed you here is the actual things by well of course if I mean I would appreciate it. If you can accept me introducing my colleague who is who can go more technical in this. Yeah, so please. Yeah, yes. Yeah, thank you guys. Yeah, we understand that the different gene expert machines and different modules that can take a multiple samples to being tested at once. So what we really like done is each particular sample is has its own ID. And that's supposed to be punching in the machine so during the synchronization process that is a unique idea that we use to interact with the DHS to to know what to track the entity instance is and to go into a specific stages that taking is a result. So you are controlling the DHS tracker with the for the testing as well because the sample collection and send for testing now. Yeah, we, we were limited to to just take the data from the gene expert to DHS to as per the project so requirements that we had, but now we are experimenting on how we can get the samples being registered from the ETL. But the other advantage of our modules that we came into advantage of is, we have integrated it with smart machines that don't need to register the sample into the machine you just put to the sample and scan the barcode or the lab number so after the results are released, the electricity module can link, but for the gene expert machines that we have in Tanzania that are not smart enough to scan on their own, then lab scientists have to enter the sample and then after the results are released we can definitely link to the DHS to. That's another question. Yeah, that's why I mentioned in my presentation, it's not like there is no solution in place, but the challenge that we had was in the in terms of economic of scale as well as issues related to privacy and local context. Yeah. Thank you. I think we're going to have to move on to the next presentation. Hi, what you guys have done is, is really in demand a lot of good people and the final connect a bit talks HL seven to the machine to the Gen DX. That's the bit that you always find is provided. Yeah, one last question. Thank you. I think we need to have a lot of machines. We'll do it at six o'clock. But thanks very much for showing us what he's like. That's the best one. Good afternoon, everyone. I am to hang and I'll be talking to you about integrating details to your therapy pro for. Adverse events following immunization or if a monitor in drink of it 19 vaccination all out in in the search. So, after a couple of authors, so I'm here to talk to you about what we, what we actually worked on. So, my outline vanished. But maybe I should give you like a background and context of off list to briefly. So listen to is a small mountainous country of about 30,000 square kilometers. So it's like a tiny, a tiny dot within South Africa. And we have according to our population projections 2022 we estimated to have a population of about 2 million. So you can see things are manageable that side. Yeah, so we have the the world's second highest HIV adult prevalence and TV prevalence. The prevalence of our 23.5% in the 15 to 19 population according to the fear 2020 is the fear is like the population level surveillance for the suit to sorry. The, and the TV incident so far on 650 cases spend 100,000 population with a coin infection rating TV TV heavy coin infection of around 59% according to WTO 2021 report. So, if you are 2020 indicated that the suit is amongst the countries that have met that I've made good progress towards UN AIDS 9595 targets with the project projected numbers that 991 90.1 991.5, and the 6.5. And when it comes to the COVID-19 pandemic, we had our first case reported in May of 2020 of 2020 largely due to lack of testing facilities in the suit to so we export testing. So we we delayed to detect any new cases of COVID-19 in the suit to and total cases that we ended up having is 34,000 and deaths that were reported were around 706 as of May 2023. And we've, we launched our vaccination program in March 15, 2021. And so far, individuals who received recommended primary dose. So the 4th of June, where 945,000 and coverage among the top plus population who are the people that we are currently vaccinating is around 60%. So you can see the map that shows you how the suit is how the suit is and So maybe I should also talk briefly to you about our experience I kept experiencing supporting and management of information systems in the suit. DHS2 based integrated HIS was developed and the HIS covers 12 health programs like the 12 health health programs that our HIS health information systems, it covers 12 health programs and DHS2 is actually operational now in 199 facilities. When I say 199, I'm talking about the public facilities in the country and some private And we have an EMR at land level electronic medical system we are using BAMNI for that, which we also supported implementation because the suit is very small, we are in about 93% of the health facilities in the country. So we are implementing one EMR solution in all the facilities in the country. So we're actually pushing that to cover the 199 facilities. But we can comfortably say that around over 90% of our clients who are currently on HIV program in the country, they are covered in the EMR solution. So COVID-19 system was developed for it's a tracker based system that was developed and rolled out when the when the vaccine program started in 2021. And we call that system within that system, it's a tracker based system and within it we also have our IFE program is also a tracker program inside the same surveillance system, inside the same reporting system for COVID-19 registry. And we also did a rapid pro based active IFE surveillance tool, and that is our focus for today's discussion. So let's talk about because our presentation is largely focusing on integration. We want to talk to you about how the systems are linking so we have called VAC system which is the registry, it is the DHS2 tracker system, we are using it both on laptops and on on tablets for remote facilities, and when we are doing like campaign work. And we are all the vaccines that we are currently providing in the country like your J&J, Pfizer, Sinopharm, Moderna, they are all in there. And it also captures like detailed demographic data of the clients, like their names, their ID information, their mobile numbers, they are covered in that system. And we have also included like census data so that we can be able to calculate things like your coverage for vaccination in the by district, like for the whole country that is why I'm able to say that we are we have covered like 60% of the 12 plus population, because we have actually just posted our data with the central census data. And in terms of interoperability of the DHS2, once you see COVAX think DHS2, because the COVAX system is the DHS2 system. We have integrated that system with what we call Citizens App, because we were actually using Citizens App for generating vaccine IDs, especially for people who are traveling outside the country. And later we changed to what you call a trusted vaccine for printing those same certificates. And that process requires that we correct and have correct mobile numbers of clients within our system, because otherwise it does not going to generate the certificate if your mobile number is incorrect. So that was an advantage, it was an advantage to our system to improve on the data quality in terms of mobile numbers that we have in the system there. So we, our COVAX system or the surveillance system that we were running in the countries had some limitations in the sense that it was since people have to come back to the facility report. We had limited numbers because it was more like a self-propelled system where people have to come when they feel like they have a side effect and report. You can imagine that if someone felt like it's a minor thing, they will not come back, so we'll never get that. So that was an issue and we ended up in a situation where we have a number stalling for a very long time. And this is like a country, a country where dashboard that shows you like the number of IFE cases reported in the whole country. So the whole country has reported like 1300 cases. And for you to get to this 1300 cases, it has been like a two-year process to get the data. So it's very slow because it depends on people coming back to report the IFE and then the people they find in the facility having the willingness to also function the data into the system. So we learned of RapidPro and we built on that platform to develop a more active open source cell phone-based community-level surveillance tool. And the aim is to complement that our passive system because you remember that we have a very passive system that depends on clients having to come back to report when they have IFE's. And the numbers are not coming in. So we want to be a little bit innovative and try to pull the numbers to come into the system. So we are looking for a way of pulling the data from the people who may have IFE's. So we thought of how if we implement a solution that calls people from the community to report when they have a case, right from where they are. That is where the SMS, we are not using WhatsApp. While RapidPro is able to use WhatsApp, Telegram and other technologies, we prefer SMS because in Lesotho, people don't have, like a lot of people don't have smartphones. So if you are going to go WhatsApp, then you're going to be limited in terms of the people that you're going to be able to reach out to. So that is where we are. SMS is actually a deliberate choice to resolve that kind of problem. So the system, it's automated in the sense that once you are registered, I'll show the workflow, once you are registered in the DHS for your routine immunization, the data gets pulled from the DHS to instance from the tracker into RapidPro. So there's like an automatic process that you have done there to put the data from DHS to instance into the RapidPro which will then manage the workflows of sending SMS to our clients. So the purpose of RapidPro is to manage things like your workflows, when to send SMS, which SMS to send to which client or which day and when should you stop, what is the criteria to exit the workflows and stuff like that. So RapidPro manages all that work. So it was quite important that we integrate DHS with the RapidPro instance to manage that process. So the workflow is like our client will go from the, now I can move from here. So this is our client here coming to the health facility and once they get to the health facility, they get vaccinated and they have to wait for some time. I think you know the process that I mentioned earlier, they have to wait for some time after that and while they are waiting, the data gets captured into the DHS twisters. That is our COVAX system. So then this is our, we have put like a small, I don't want to talk more about that but we have put a little software in between just to help us to pull back data from DHS because when we were pulling directly from DHS we were struggling to pull back records directly from DHS to. So we had to pull like a small plug there. So I don't talk about it a lot because it is not yet like available to everyone to use but it's a small Python tool that we did just to be able to pull back data from the DHS to, and that's all it does, pull back data from DHS to and then send it to RapidPro. So RapidPro will do all the workflows inside there and it will send like introduction messages. I'm trying to be, I can always come back, I'm trying to be careful about the time. So we have our DHS to instance here. This is where everything starts because the lens gets registered here. And then, so that small Python code, what it will do, it will take like all the demographic data that is required and key data elements that are required from the, from the things like a contact number, the vaccine that you actually took, the dose number and things like that. And then it will send that data into, into our, into our RapidPro. And then RapidPro then will use canal. So canal is like your SMS gateway. So where you are able to send the messages now to the, to the, to the clients. And when the clients respond back, they respond back through the canal, and then the SMS gets registered in RapidPro. And then RapidPro then we are, it is no longer using this. RapidPro then state, will, will state send the data into, into DHS to a tracker instance that you have developed. So that it is now available on the, on the dashboard for, for everybody to, to then see it, to see at central, at central level. So that is, that is like a quick depiction of the, of the, of the workflow. So with, with this system. I didn't really want to call this conclusions, but because, because we have just started so we don't have like, like real, tangible stuff to tell you at this stage because we just started, but I thought we can, we can communicate about things that like the potential that we see like the next steps and some of the things that we think we can be, can we can be able to achieve from this. So with this tool. COVID-19 virus, vaccine recipients can, can report any presumed. I feel in a timely manner from the comfort of their homes, because the system actually, after you're vaccinated, the system automatically sends you a message. Hi, how do you feel today. So something like that. And then the plan cannot, cannot is, cannot always respond by, and then it will ask some something like after you're vaccinated. I'm not, I'm not, I'm just talking stuff. Yeah, I'm not talking about the communication science, but something like how do you feel today after your vaccination and the client can be able to say, do you feel like you have any, any, anything. Are you feeling any pain anywhere today after you're vaccinated and that client will then respond to that. Well, once they respond, the system, like the rapid forward flows will then be determining whether this is an if you are not if it's an if you then need to possibly check out. That is how that is how it, it works. So the client receives an SMS that they respond to the SMS, of course we are paying for the SMS so they don't have to worry about the cost of the of the SMS. And so if, if, if the client then feels they have any, any, any IFV that will report back on the SMS. And the, and the, and the, and the system will send it rapid progress and it to the chance to. So integrating rapid pro with a IFV tool to all the DHS vaccine registry can be scaled beyond COVID-19. Obviously, with COVID-19, for example, we have cases of people that have vaccinated. They have not yet received their boosters. We have the mobile numbers, but we are not able to remind them at the moment, we just know that we have this list of guys that have taken the first dose, they're supposed to take the booster dose. So we are, we are seeing potential in this system that we can be able to remind those clients to to come back so that we have like a pulling effect for for our clients to to come back. But even beyond beyond COVID-19 vaccination, we are also looking at scaling this to other other systems like other vaccinations. We have children that are being vaccinated. And sometimes we have like a very high top upgrade of certain vaccines. So we are thinking of actually using this at less scale within that within the EPA program to cover other vaccines. But even beyond that, when you think about things like your bigger HIV program, we have clients who are coming for HIV can treatment services that we need to remind of the appointments for for picking up their medications. We are already thinking of that as we talk so so that we can be able to integrate this system with our email solution so that clients can be reminded of their of the appointments as as things go on. That is where the Ministry of Health is currently at they want the system so this rapid flow to also be used in those other use cases that I've just mentioned now. I think I went really fast. Yeah, I'm here. And one of the issues we are facing is reporting a face to the WHO status and we have been able to be able to integrate with that. And clinicians have to, you know, report manually to the WHO. So did you have similar experience or do you have any plan to integrate this with the WHO VG flow up. And maybe to, I guess the question was able to go to the audience right I don't have to repeat it right. Do I have to repeat the question, or it was able to go through. So I can continue okay. So we are not yet at that level. So at this stage, we were actually answering the question of pulling the data from the clients. And then we can later worry about sending it to, to, to, to how, how to interface with, with, with the digital system, the digital system, but at the moment, our priority was to pull the data from the clients so that we can have the data at least at means of health level. And then we can be able to talk about how they managed to then deposit to other stakeholders. Yeah, but we are not yet there. Since you captured this as plain text data from the, from the clients. So like, say if they have a skin rash or they have fever or something after. Do you have a separate and separate coding step inside this just to kind of assign the correct codes and everything and do the coding part. You're not following a dollar show guidelines for COVID reporting. So do you just capture it at plain plain plain texts, if they have another adverse effects reaction. Yeah, so I mean, when we were using coding system but it's a local, it's a local coding system based on our details to instance so obviously when they report for example I said we will be asking a question like, how do you feel today and that land will be. Yeah, we have tried to limit the so that our answers are structured because you can imagine that the level of literacy of our clients is not the same so you want to avoid situations where they have to type things in. So the, the, the, the data elements that are within DHS to already coded into some into something that they have to respond something like one to something like that, even for years and know they have to respond something like one and a two. And then we have mapped that into the DHS to data elements, so that when the report the SMS will, we know that for this particular question, one response, something like headache, and then headed the code for headache will be the Swiss linked to DHS to like that. Yeah. Oh, yeah. Good afternoon. So, I'm Nilo Kavijekon. I'm a medical epidemiologist working for WHO emergencies program at headquarters in Geneva. And today I'm also joined with my colleague, my cell phone who will also join and present some of the slides. And so I lead a project called he was in a box. You must have heard about this tool. If you have worked in an emergency or in any other countries, you work has gone through emergency. So now you had the first two presentations about more of routine, peaceful times, or things that you can do in a more methodical manner. Where we work. This tool is about early morning alert and response about priority communicable diseases, basically to detect outbreaks. Where we work. There's always huge population displacement, following a natural disaster like tsunami cyclones floods, volcanic eruptions, or in places where the surveillance system is overwhelmed by large outbreaks, like West African Ebola or color in where the reporting mechanisms are quite overwhelmed, or that you also want to know what are the cases in the community through alert mechanisms, or in other places like northern Syria, northeast Nigeria, there is conflict or violence, or there's no state factor. So we have to work with non state. So, so the tool is by double WHO is mandated to make sure during an emergency that we detect outbreaks early. So while we have so many other surveillance principles, the tool and the mandate is in an emergency to detect an outbreak early. This is not about understanding the case burden, the seasonality of the disease. This is entirely to detect an outbreak. Because, as you know, during an emergency the risks are high. So if there's a, there's a threat, it will spread rapidly. So this is a very joyous moment for us because the tool has been in operation for nearly eight years since the West African Ebola. This is the first time we are integrating that with DHS to. So, I will take you through some of the strategic components and my data analyst, Marcel will take you through some of the, the experiences in South Sudan DRC and how we move forward. As we go into production. So, how do I move now. Okay. Okay, so the tool is is really a simple rapid and a flexible data reporting mechanism that can be established in a field setting, which has gone through a disaster very rapidly. Basically, it's an Android mobile app that can collect data, and I'll take you through in any location we are not fixed on primary health care level, or fixed facilities because you can't find them all the time. So it can be primary healthcare clinic it can be a mobile clinic it can be a health post, or it can be a community health worker or a volunteer who's reporting. It can be a community leader religious lead as well. So it's a very simple and a rapidly deployable I'm talking about few days. That's the rapidity because emergency work if you if you know that there are certain standards that we have to be abide by. We cannot set up something train something if it takes one and a half months or three weeks to set up, because that's good enough time for a outbreak to spread. So in emergencies we have in other programs this kid concept, you know the color accurate trauma kids that can be sent to places where where there is earthquake or floods. So we also take this kid concept. That's why it's called it was in a box. So it's an early morning alert and response in a in a in a box basically a kid that can be sent, which will have the mobile phones with the app, and a laptop for the surveillance office. And we most of the time know in these places during any disaster whether it's conflict war or natural disaster there's no electricity. So there are solar charges to to charge the phone as well. And I will go through quickly because so basically what happens is this data collection from various locations from the field, and even from the laboratories, and these laboratories are not the type of sophisticated referral laboratories that you spoke about these are the field labs, which will usually test basic key priority diseases, and also connecting with the response teams. So we manage alerts at the surveillance office the support is given to the surveillance office at the web interface to manage the alert, and we use WHO guidance for this. And these are not just alerts where you send an alert with the phone number. This is the full alert management as per the protocol that people go through the risk characterization, identifying the outcome taking samples etc. So we have features that will allow the surveillance office to to analyze data. So the dashboards. And also if you want to do your own reports, the reports are automated. The basic idea is that you collect data. Rapidly, whether it's through indicator based surveillance system or event based surveillance system, and that you can analyze data rapidly for response. So the epidemiologist and surveillance officers have more time for response activities, not for data crunching. Even there's an outbreak detected the tool also provides support for line listing, so that you can do if there's a core outbreak detected, they can also do the line listing, and they can have outbreak reports dashboards prepared. So that's the kind of the tool that we have. And it has been in operation for quite some time. So basically, so we, as I said we've been in implementing since 2015. Quite number of places and we also support number of Pacific Island countries for pacific syndromic surveillance. But here for the the new feature we've selected South Sudan and DRC. South Sudan has been one of our initial implement in places where we have implemented. In camps, and then, because South Sudan is a protracted crisis for a prolonged period of time, it is one of the operations where we have more than 1000s health facilities reporting, and the RC, we have five provinces reporting to Evo's in a box because it is also in in crisis. And in crisis as well as number of outbreaks happening. So, so, just to highlight why we are integrating he was in a box at these stages, if you look at emergency data. When an emergency happens, you don't have the normal structure there's population displacement to IDP camps, refugee camps, the health clinics are managed by either UN partners, NGOs, NGOs, not the normal surveillance structure. Most of the time the data is lost. Even when there's an outbreak happens it is sits the data sits in Excel files of somebody's hard drive, and that data is not incorporated to the national system. Previously we had a feature that just allows import export of data so anyway, whatever they try system in the country can have data back to the system. And it is also one other. Another thing is with the chance to you collect quite a lot of other sets of data that can be used for response activities in an emergency. In some countries, we may want to know about your wash facilities or nutrition or the bed capacities that can be used for better response. These are the reasons why we wanted to integrate and we will talk about the way forward after Marcel will take over and discuss a little bit about the the experience Marcel over to you. Hello, I hope you can get me okay. Thank you, colleagues. And so I will start with the technology behind the tool. Basically, he was have some few component we have a web interface that is the main interface where we are managing all the features of the tool. From the surveillance office from community health workers. We have what we call the country instance because it is a multiple country tool, such as GHS to it can be used for many countries. We have the stand alone server. Actually, in some places in the previous sheet me look at presented, we have island and in island. It is important to have a system that can collect the data because the access to the internet and access to global connection might be challenging, especially during emergencies. We have a mobile app that can be installed on a mobile phone that will be used for managing data and it can also be used for processing information. This is based on the fact that sometimes the data collected my journey rate alerts that have to be managed by surveillance officers in order to see what are the risks associated to the public health events that are reported. We also have an SMS gateway mobile app which helps for the user. We, the user will be reporting directly to the system, but the system will use the SMS to report because most of the time in emergency. It might be an earthquake or a disease outbreak. The places might not be accessible with the internet so we can use the available to generate work to communicate. So for technicality, the module have been developed using Python, Trian above, the front end module have been developed using React.js and then we are using GHIS to API for communicating with GHIS to. It is also important to note that for the system, there are in-built API to communicate with other systems. So, basically, no matter what the system is, it can be another HIN, HIF, it can be a simple code or a visualization software, it can be used. So, for the integration process, we are using three types of data set. We can have interval-based aggregated data set that are adapted for epidemic-prone diseases. It might be daily, monthly, weekly. We have immediately notifiable data like diseases of immediate notification based on the HIR regulation. We can also have the alert. Actually, it was a system of alert based on the threshold on the national guidelines of countries where when the threshold is reached, an alert is generated into the system that needs to be processed to see how data can be handled. And we can have as well case-based data or line leads. For example, we can have information about the event-based surveillance. So, for the general principle, the data can go to GHIS to aggregated model if it is interval-based, or it can go through the tracker model using event program or tracker program. So, the user have the ability to configure all those elements himself to see where he wants the data to be. And the feature is integrated into the system. We can move to the next slide where we have two key components of the process. The first component is the mapping of reporting locations, given that it is two independent systems, and the second key component is the mapping of reporting variables that will be formed. Next slide. So, what is the integration process? Given that Niliqua explained it earlier, we might have in an emergency a situation where temporary health facilities and IGP camps have been set up and then we, EWAS is collecting data from those places. We have integrated into EWAS a process that will handle all the data and push to GHIS-2 in the right place. So, basically it will be in a five-step process with creating a project, connecting the GHIS-2 instance of the country, mapping location and variable, and then finally pushed. Let us have a quick look at how it works. So, basically, for us, the process has been completed, already completed, and it has been tested in two cases. We have tested it with GR3, with South Sudan. Basically, we are using from WHO guidelines to prevent access to the data for unauthorized personnel. So, this authentication feature shows how easy it can be to just enter the GHIS-2 server URL, the username, the password, and then the system will log automatically to select the data source. And to select the type of program we want to into GHIS-2 and then to see how we can push. Let us have a quick, we can commend this sheet for one minute. So, basically, for the push configuration, given that we know that sometimes it is outbreakable diseases or immediately notifiable diseases, the person, the surveillance officer can configure into the EWAS system how often he wants the data to be pushed. It can be on daily basis, and he can select a time frame, and he can even decide to push the empty form based on the availability of internet and the size of data to be integrated, to be pushed from one system to another. Let us move to the next slide, please. So, what, yes, next slide, please. What can be the gaps on the opportunities of improvement? The first one is the inconsistency of common identifier for location that is in GHIS-2, it's called organizational units and disease variable. In that case, in the EWAS system, it is possible to handle it where we have prepared a CFV mapping, so rather than mapping it from the system variable to variable, we have a CFV mapping that can be done so that it can be used to map the two systems. And the second challenge might be a clear vision on how the country uses, once the GHIS-2 will use alert data for outbreak, it is worth to note that alerts data are dynamic, so it changes very quickly and pushing those data from one system to another might be somehow challenging. Sometimes it might be just analytics on the processing that needs to be pushed. It is also important to explore options for integration process and the risk for local expertise in countries, and we expect to have this participatory approach of communication for the EWAS GHIS-2 integration roadmap. Yes, thanks Marcel. So just three things I would just say, so going forward our plan is, so once we have this activities in process, so once we know that we can integrate EWAS and GHIS-2, the next step is to strategically think whether we can also talk about this as part of readiness and preparedness in countries where the GHIS system as well as it can be affected by emergencies, and also to liaise with the GHIS-2 team for further capacity building and training as we do for emergency surge teams through WHO and other cluster partners and health actors. And we also have plans to integrate the tool with other platforms and tools like contact tracing, Godet, HRAMS, other emergency tools. So thank you very much. I think you can ask her. Yeah, I'm just going to help. So this is great. But my question is related more on the implementation. I don't know if you guys implemented this and have reports emergency outbreaks and so on. So if there was any decision made based on the data coming out of this system, it would be great if you can give us some information. So the other question I have also is like, do you use fully this electronic reporting or do you use hybrid like the manual and also the electronics? So a little bit more about the implementation. Thanks. So they quickly send you the slides. It's Chick Mada with DHS-2. In places where there is no connection. You also have SMS gateway app so that you can have the reporting done in the mobile but it converts the reports to SMS and again other web interfaces. So it's basically we want rapid reporting so we encourage the Android app use. If there is nothing, we have open WHO training so it's free to everybody. There are 16 modules, English, French, Spanish, Portuguese, Polish, Ukrainian in many languages so you can, because we support Ukraine crisis as well. So you can log into your land login and get trained. Thank you very much. Sorry, it's this email, it's that email right there. We get it very rapidly I'm sure. I realized when did the start of my English. And one of my roles is. And all of these materials that we get, we really get to us to dig into them. But we do meet every Friday morning. We've got this running integration team, which one of the things we try to do is to reach out to within the time that we get people to share their important examples with us. So you might find over the next few weeks, we'll grab you again. Right. Martin was asking when these integration meetings happen and how to get involved. What I think it's 10 o'clock every Friday morning. 10 o'clock my time. No 10 o'clock Oslo time. 10 o'clock Oslo time, nine o'clock Dublin time. It's a, it's a very simple format of, we just use Google Meets. The easiest way, just send me an email at Bob at dhs2.org, anyone who's interested, and I'll just put you on a calendar invite, and then you'll get the notification each week. You'll be at dhs2.org. Sorry about that with the. Ready to go. Thanks. Just just a second. Let me be sure that the screen is being shown. Didn't lose too much. You still have 19 minutes. Okay. Well, good afternoon. We like integration so much we have integrated four speakers into our presentation. So my name is Whitney Peterson. I'm the first of the four speakers. I'm going to start with Samaritan's purse. I'm going to start us off and talk a little bit about from a program's perspective. And then my colleague Derek will come on and talk about from a business systems and project management perspective. And then we'll hand it over to Sarah and Adrian to talk a bit more about the technical specifics. So yeah, I'll just briefly start Samaritan's purse is a faith based nonprofit organization were based in the United States. We have 16 country offices and do programming in over 100 countries through local partners as well. We do long term programming through our country offices and then we also have an international disaster response division. And then we also have an emergency response to disasters as well, which is what we're going to talk about a little bit today for the utilization of DHS to in our organization. We have started by targeting emergency medical responses. The focus on emergency medical response will respond to war and conflict to sudden onset disasters, epidemic outbreaks as well as population displacement so little bit of everything. Here's just a sample of some of our responses from large scale full fledged tier three hospitals with multiple ORs and 47 inpatient beds to smaller outpatient clinics mobile clinics will do training programs. As well as a mobile surgical theater here in the bottom center ish some of those end pictures are cut off but is our Ebola treatment center in DR Congo from 2018. And so I show this just to say the breadth of our programming is very broad. The context that we work in are always different sometimes we have connectivity sometimes we don't. And all of our deployment systems are modular so they can build on each other we can increase the scale we can decrease the scale. We really needed a data tool that would that would be able to adapt with that that could operate in any of those contexts and meet a variety of our needs. We also looked at the data stakeholders for us that's everything from the program managers to our team at headquarters and really looking at who we're reporting to who are accountable to so if we have institutional donors or WHO local Ministry of Health partners and wanting a data tool that could incorporate all those components while maintaining a standardized model so we're not having to reinvent the wheel every single time. So that's really what led led us to DHS to and our approach was this looks a bit chaotic. This is where the technical team comes in and makes our chaos, something simple and usable for the field. Here's just a sample of our paper general intake form and that's on the right over here. And this is the form that we created that's actually all encompassing so we use the same form, regardless if it's an outpatient facility if it's a massive field hospital like we just did in the wake of the earthquake in Turkey. And so what we did is we looked at the key indicators that all of our primary stakeholders would want so we looked at our internal indicators. We looked at the WHO minimum data set which includes reporting for the early warning system that we just heard about. And also looking at our, our most frequent institutional funders which for us is USAID and BHA. And so we created a form that encompasses all those possible indicators, and then our DHS team was able to map those out. So that regardless of what type of response we're doing, if we're being funded externally or internally, it's innately in this form somewhere whether we choose to utilize that output or not. The possibility and the data collection on the front end is there, which makes it really scalable for every type of response. Now I'm going to hand it over to Derek to talk a bit more about the business. Speaker number two. Hey everybody I'm Derek play lock on the it program manager for Samaritan's purse for our international ministries of which international projects excuse me ministries of which Whitney is a member. For this particular project was also the project manager. I want to talk a little bit about the business case I really want to get you to the technical solution but I want to talk a little bit about the business case. And really how we how we were able to successfully execute this project you know when you think about an emergency medical response wherever it may be. I'm not telling this room but data management good data management practices are absolutely critical I mean whether we're reporting to internal constituents or our benefactors or even for us to better make decisions about the program itself we need really good data that we can we can depend on to realize that we've had we had a tool that we've been using for quite some time that was based off of spreadsheet and so I think we all know the issues with running a very large organization on spreadsheets we quite honestly we just gotten too large for this, and all the other problems associated with spreadsheet so what we wanted to do is build an industry standard of a heavyweight tool that will allow us to professionally manage this EFH anytime or any emergency medical response anytime we were to deploy. We didn't want them to have to worry about it we can do some customizations and get it done quickly we want to respond with an IT solution as quickly as they want to respond to people who are hurting and so the result was that we build a tool we have a very good partner next and I call them partner they're not a vendor they're very good partner with us. Thanks Siri. And they build a very nice tool for us that that is supported on on our corporate hardware standards. Who's becoming a bigger standard as well and says it's been a really nice tool that we've been able to deploy coming up next. We've already deployed for for emergency medical also for cholera response. We got an Ebola template that we're going to plug into here to this this core platform, and then we've already done medical training tracking airlift distributions medical supplies. And then we do have a cliff lip palette and cataract program that's going to be using actually DHS to this platform plus the Android app pretty soon here so we're pretty excited about that. You know, the reason we were successful is kind of this classic project methodology of people process and technology and in this case it truly was a good partnership we got. And I think from senior management that we needed to help solve this problem. We connected with our international health unit of which Whitney was our lead SME had a great team was involved in the requirements definition. Testing, piloting and in the eventual go live and of course, our partner who's coming up next EST built this thing for so it truly was a good group of people working. In the process we really had nothing and that's where Whitney and her team came into play with good governance, really great standard operating procedures, and most importantly enforcing those things in the field. You know, whether it be common indicators or a standard form, they did a really great job and the technology becomes easy when we have that. And lastly, yeah, you know, it's this is based on DHS to surprise surprise. And I think, although support may not necessarily be considered technology for us that was most important so we have them, or as important so we had a multi tier support strategy from our internal ICT country it folks in the field, coming up into our headquarters team where we triage issues, and then third level support being our vendor. And we did that through a help desk ticketing system that so that we have some some consistency and some standardization and so that's it from the it, the project side I'm going to bring up Sarah to talk about the technology solution. The speaker here. Don't worry, Adriana is not going to speak to, he will be here but just to answer some questions if you have them so I will be like the last speaker. So, Adriana is the one leading the development of this application so it's a pity he cannot be here but well I will do my best. First of all, I wanted just to introduce you briefly the company behind the app we are ICT and we work mostly in cooperation and health and research projects. We work a lot with the IS to but not only. And one thing that it's like very important for us is that almost all of our work is open source. So, focusing on the application that we are going to talk about today. I just wanted to tell you that this application is part of a bigger context of emergency responses that we are building for some items. So, we have like other apps that some of them are already working to I mean they are being used for instance, the one for cholera in Malawi, and we have another one that it's been tested for Ebola, and two more in development. In the case of the emergency field hospital, we are pretty happy because it's being used already by Samaritans into very important emergency responses such as the Ukraine war and the Turkey earthquake and I think it went pretty well so we are very happy with the results. And I wanted to highlight briefly the requirements of an app such as such as us. So as we may was saying before, some of the requirements imply that sometimes when you are in the field you don't have connection or you don't have a reliable connection so one of the most important things was that it had an offline mode that it only not permitted you to capture data but only to have already some visualizations tools that could enable you to analyze that data and take like better decisions in the field. So, apart from that, as I was saying, we also wanted like a tool that it's very easy to use and very easy to install and configure and something very quick. And also I'd like to highlight that although it's a DHS2 app, it's not only that, it's more kind of an integral solution so you get a laptop with Linux with Wuntu and you can turn it into a complete solution that you can take with you into the field and it's a DHS2 server that you can use to pull data from the headquarters and also use it to push data. We will go into more detail in a little bit, but first I wanted to show you the application so you have a better feeling of what it is. So, when the user goes into the app, this is what she or he sees, it's like mainly eight functionalities and I wanted also to highlight that this is a mix of when we could use like a standard DHS2 functionalities we have used that, I mean no need to reinvent the wheel. So, but sometimes we need it or either custom sections or we need to use some of our generic apps to go to the last part of the process but some of the features are just taken from the DHS2 like the tracker or the visualization tools like dashboards and so on. So we have here one very basic feature that is just like to configure like some kind of maintenance data and then you have the intake form just to capture the data and we have as I was saying before three tools for visualization and this is like one of the complex part to turn all this data into a report that is completely customized and it's pretty complex. And the user just have to click on a button and the magic happens. So I think this is this is pretty great. And this is done thanks to a generic app that we have that it's open source that it's called metadata sync. Then we have also another module that is the training app to teach the user, even if it's very easy to use if he or he needs to know how to use it so we have this app that by the way it's also a generic app and won the contest here like two years ago so you can take a look if you want. And as I was saying before you have to have the ability to pull the metadata from the headquarters and to push the data and this is just two buttons for the user so this is pretty very easy to use. And then just to finish I wanted to explain a little bit about the whole process, how do you go from a Linux computer to a DHIS to machine that you can use in the field. So you have several steps but they are, as I was saying before, very easy and very quick. So first of all, you have to get your machine and you run a script, it's just one line script. And this launch a website so you have already an interface and a graphical interface that is curse no one. And from there you can choose from several dockers. So for instance here in this case you could have like the color outbreak app, but here we have like the emergency field hospital one so you choose your docker, you click. And you have already like the skeleton for the app. And now you need the metadata, because that's what can distinguish between, for instance, I mean you can be in Turkey in the earthquake and you can need some kind of metadata, or you can be in Ukraine and you can need another kind of metadata. So you have the skeleton of the app but when you pull the data you have like a customized solution. And then you have already your app running and you can use it to capture data and to visualize data. Here I'm presenting the two of the places where has been already used as I was saying I think that pretty successful. And, and all these flow of data, as I was saying for the end user is just two buttons, just pull and push. So that's pretty cool. And then to finish, I wanted to add that we have also like develop these home page like this landing page. So when a user goes into this website, depending on the permissions that that's here he has, maybe she can access all the five apps or maybe just one of them. And this is done thanks to another generic app that natural is going to present on Thursday, and it's in the context so take a look and if you like it, vote for it. And so yeah, I think it's also like a very nice way to present like the quick links to different components. And that's it. Thank you so much for your interest. Thank you, Sarah, Sarah. I'm going to ask one way. I said at the start, it wasn't. That's the bit that you left out. What about this. Squeezes of red lines between the mapping. We really need better open source mapping tools. So I think in all of the projects you you're mapping rapid pro flow results. In all integration project you have that mapping thing it sometimes really really complicated. We've got some tool chains that we develop index. So we're looking for a good mapping solution to make the job easier. Quick questions to the presenters before we. Unfortunately goes. So you're on the. I'm alone. Come on. Oh, sorry, thank you. I just have one more question. Thank you. I have some Ebola responses and I know that. There's often. Like MSF is setting up hospitals and if Samaritan's purses, I wonder, is any collaboration across for developing some or collaborating on some of these. Modules and then also thinking, you know, at the other side for the case management side, and the Guinea outbreak that was happening. There was a module that was built on tracker that was used for the, you know, in DRC. And I'm just wondering if there's any sort of way to kind of collaborate across partners so that everybody's not developing their own system. Thanks. And do you want to? Hi. Hi. Hi. Oh. Hi. Hi. Hi. Hi. Hi. Hi. Hi. Hi. Hi. Hi. Hi. Hi. Hi. Hi. Hi. Hi. Hi. Hi. Hi. Hi. Hi. Hi. Hi. Hi. You know, we try to do a lot of preparation so that typically when we respond, it's within the first like 72 hours of an emergency. And so for us, we have to have that stuff cued up and ready to go. Once we're in the field, we hit up all of the health cluster or the cluster meetings, right? Which is typically how we collaborate. And like, happy to coordinate on the ground once that's there, you know. And so that would be like my initial response. I think you had a second question as well, right? Well, it's just more about if there's coordination with the partners, we're developing a key standpoint by the top major people with treatment. And then on the surveillance, are they sharing information surveillance, or are they developing their own system? Yeah, and especially with regards to surveillance, like specifically that's why they were able to build that the MDS form because the surveillance team, the WHO who leads the surveillance wanted that form submitted in a very particular way through an Excel document. So they were actually able to make that as part of an app and enhance our like immediate collaboration on the ground so that there's no delay in that reporting and that coordination is there. But I mean, to speak to your point, I know that you're with MSF, I assume. No, I'm actually with the city. We also need to bring myself in the jail. Yeah, and I think that we're so happy to start collaborating. I know they've been using DHS for a lot longer than we have. And for us historically speaking, our Ebola response in the past in DRC in Liberia, we weren't utilizing DHS too yet at that point. So we're looking for it as we're newer in the DHS2 realm to see what the collaboration looks like in real time among actors on the ground. We all have to go and I was gonna close down, but I'll give you the last words. I'll give all we thank the four presenters. I think there is a lot of way of collaboration and coordination across different partners in the countries. And so I'm from WHOs and I'm working on the health data analytics and also part of that was the health data collaborative that we also as a secretariat. And I think this is an area, especially surveillance and country sort of integration of different data systems and how we support the country's strengths and the capacity and data use. This is one of the area maybe we, there is a health data collaborative different working groups and surveillance is one of those working groups. But also we are trying to see how we can work together in the countries. And maybe this is something we can touch outside and have discussed further. But this is one of the area really important when we don't go in and set up a new systems and then we leave or the system collapse or having a multiple system in the countries that nobody can use and share the data. And I think this is one of the reason why we started to work together a lot so across different programs in WHO and on that concept. So maybe we can touch outside later. Thanks. We're starting a fight now. Exactly. You know what I love about where this has got to and I know we all have to go. But a lot of people tend to think that integration is about getting computers to talk to computers. Generally, that's the least difficult problem and integration in terms of human beings and partners and organizations as where it all starts. And that's the theme for next year. Right. Thanks very much. Big hand for the four presenters.