 up Maria Vector Control? Okay, sorry. Okay, perfect. So the Maria Vector Control is a component of a more comprehensive Maria program that can be carried out by the Ministry of Health on different countries. In the HIS-2, in the WHO health data toolkit, we actually have two different types of packages you can see. We have a great one where you can find information and keep that information used on the malaria burden reduction and elimination strategies, the surveillance part, so two different tracker programs supporting case notification, case investigation, investigation classification workflow, and then the logistic part, so supply chain readiness and traceability of Maria drug test and long-lasting insecticide net down to point of service delivery. As well, very soon, three different packages will come and specifically on the vector control part. So the breeding sites monitoring will be a tracker program, the indoor residual spraying as an event and insecticide treated nets as events as well. So this will come in sooner on the June release. So what we are going to talk about today, so is the innovative use of the HIS-2 for malaria vector control intervention that is shared from malaria in any country. So we'll have experience for different countries alongside lesson learned implementation, control customization, and data use. So we will actually, for an example, so the HIS-2 Android application has been used to scale up entomological monitoring data collection in Ghana, Nigeria, Côte d'Ivoire, and Mali, and to optimize in the case of Zimbabwe, so to optimize the U.S. campaign performance. And as well, Tanzania will share, the East Tanzania will share his experience using the HIS-2 dashboard to integrate logistic and health service delivery data to optimize continued distribution of long-lasting insecticide nets to premium women and children under five. So I don't want to take more time. Just introduce the first speaker, Otias. Thank you very much. Okay. Thank you, Stefano, for the introduction, for the presentation. Can you hear me well? Okay. Thank you. So I'm going to make a presentation on a Zimbabwe scale study. I'm Otias Tafmani, responsible for data management within the National Malaria Control Program, funded by global funds, mostly. That's where we get most of our funding from. So as a country, Zimbabwe is in Southern Africa. This is just an introductory part. It's in Southern Africa, close to South Africa, Botswana, Malawi, and Mozambique, for those who might actually need to know the place. So that's the country in green, for those who are actually seeing. That's Zimbabwe. So our aim or objective was to integrate indoor residual spraying, that is IRS reporting into the DHIS-2, and improve quality and the data availability for decision making. So basically, generally, initially, as a country, we had a challenge in terms of accessing our data. So we had opted to use DHIS-2 as our source or as our data storage, since we are mostly relying on the paper-based system. Therefore, we had to make sure we integrate it into DHIS-2 for ease of accessibility and also for decision making by the decision makers. In terms of the description, use of DHIS-2 system for indoor residual spray, that's what we actually targeted to do. And in terms of the platform that we used, we had to use DHIS-2 web based, which is Android. The version that we used is 2.356. I think this should be the first country, we just done that. So we started this in 2020 as a pilot in one of our provinces, and it had about nine districts. So those nine districts started to report using the DHIS-2. Then we had to scale it up in 2021 to 30 more districts, these are high bedding districts, that is an incidence rate of five and above per thousand population. So generally, in terms of our data, it's reported as aggregates, meaning to say it's not tracked per entity, but it's consolidated for the district. The intervention that we're actually focusing on is indoor residual spraying. In terms of the indicators that we're actually tracking, they also included population protected, that's the coverages. Also we've got the room coverages. Then we're also even looking at the spray operator efficiency that we could actually follow the efficiency of each and every spray operator, the name spray operator commodity management, which include the fuel, insect size, and other things that could actually be tracked. So DHIS-2 was introduced to improve our reporting since we're actually relying on the paper-based system. So we had to introduce the electronic platform so as to enhance better reporting for the country. So previously we used to rely on the Excel WhatsApp at times to reporting data. Then after that we had to consolidate to come up with the report of which it was actually a bit of a challenge, but with the digitization now, it was actually easy for us as a country to improve in terms of our reporting. This is that's the Excel part of it which we used to try and actually consolidate our data that was reported by districts. So as you can actually see here to put some formulas and so forth so that you could actually give a better insight in terms of what they've been reported, but with the coming in of DHIS-2 now, we improved in terms of the palette of data that was being reported by the districts. So this is one of the examples of what could be produced through the electronic platform that we use. That's the DHIS-2. In terms of the completeness, those who were reporting the data on a weekly basis, but for those who were submitting the data from the district, it was being reported on a daily basis, meaning to say on each and every day, we're getting updates that were coming from the districts. Therefore, in terms of decision making and for management, it was easy to make decisions based on the data that is submitted on a daily basis. Therefore, in terms of planning and other things, you could actually rely on the data that was reported on a daily basis, but on a weekly basis would also come up with the reports, which could be used by the districts, by national level, by the province. They would rely on the reports that we're actually producing. So this is the architecture in terms of the design and in terms of how the data flowed from the campsites through to the nationals or the central server. So basically, we have the spray operator who is the one who is responsible in terms of making sure they collected data as they spray the rooms. So during the process, as they collected data, they will be able to call the spray operator notebooks. That's the one that they were actually collecting the data through. Then from there, they would go to the data manager or the DAO who is responsible for consoled dating at the end of the day. So each spray operator would submit data to the data manager at the campsite. Then the data manager would consoled date any capture in the DHIS-2 tracker capture. From there, the data manager would submit the data to the central server whereby all those who have got the access rights, a district level, that is at national level, a provincial level, from there they could actually make decisions based on the data that is submitted. So in terms of consoled dating and submission, it was done on a daily basis and would expect those reports every day. That is for the period that we're doing the spraying. Meaning to say, in terms of supporting supervision and core banks, we would actually rely on the data that is submitted on a daily basis. Therefore, what it means is for each and every action that was being taken, it was based on data that has been submitted. If for example, the targeted population or the targeted rooms to be sprayed is 2000 and they would spray let's say 500 or 700 or a thousand and the other thousand is not sprayed. A core bank would be required. This would be determined by the data that would be submitted. It was already for each word you'd be having the targets already set in the system of which that would determine in terms of your performance. Therefore, the data that would have been submitted will actually determine should we give a core bank or not. Meaning to say, our coverages would definitely improve. Therefore, in terms of the core banks that were being done, they actually improved in terms of the number of rooms sprayed and even in terms of the population protected for each word. And also again, the data would also determine in terms of where should actually support go or supervision should be done. And even again, in terms of resources, you could actually see the fuel that has been consumed, the mileage that has been traveled and in terms of logistically, you could actually rely on the data that was being collected and submitted through the central server. Then even again in terms of feedback, we also try to make sure the reports that we produced would send them back again to the districts and also to the province so that the managers would want to make decisions, would actually rely on the data that is being collected or submitted through DHS so that they could actually easily make decisions based on the data that is collected, of which that was one of the greatest strength that I would say the digitization of IRS in Zimbabwe actually helped us a lot in terms of improving the quality of spray and even also the quality of management and also even the quality of resource utilization of which that was actually an advantage to the country. In terms of the quality or the performance that would get at the end of the day, these are some of the results which we actually try to automate this way, the dashboard that we're actually using, the greens, that's where we're actually having a good IRS coverage, the yellow, that's where we would have a low IRS coverage. So in terms of decision making, it was actually easy to say where should we put more resources, where should we put more support, which districts would require more support. Then even in terms of performance, you could actually see where you've got population protected that was low or where it was actually high. Then you also have got the dashboard again in terms of the impact, that's for the targets, where rooms protected, rooms sprayed, the population protected, it was actually easy through the dashboard you could see on a daily basis. So it improved in terms of how we do things or in terms of how we're actually analyzing our data because it was done on a daily basis. So those who did failed to report you could actually easily follow up to see if they could actually submit their data. I'm actually being told my time is actually up. Then in terms of the benefits, one of the benefits was actually increased data completeness and timeliness resulting in better visibility by management and by those who are responsible for vector control. Then improved data quality is one of the greatest advantages. Then also again, there was also capacity building in the districts and even at MOHCC. Then in-country technical capacity and support in-country capacity in terms of using the DHS2 and also in terms of customization and everything else, it was done locally. There was no external support that was actually sourced. Then one of the challenges in terms of using the system, of course as we know most of our African countries, we've got challenges in terms of internet connectivity. So it didn't spare us as a country also. Then the other issue was also to do with the staff rotation. Then in terms of the future plans as a country, we're actually looking forward to take advantage of the digital decision that we have already done. So we are going to tap in in terms of the experiences other countries have gone through in terms of digitizing most of their reportings. Therefore as a country, we're also even looking forward to include the endomological data. Then we're also looking forward to also include the length, long lastingness data into DHS2. I hope I'm still within the time frame. Thank you so much. But I was too having a lot to say, but due to time I've tried to compress so that it fits within my time frame. Thank you so much. Back to you, Stefan. Thank you, Stefano. So we're moving from IRS to entomological data. I will be talking about how we used, can everyone hear me all right? Okay. How we've used Android capture app to collect entomological data in four countries. I am a monitoring and evaluation specialist for Apped Associates and I work with the U.S. President's Malaria Initiative funded project, PMI Vector Link. In the project we work in 25 countries. We have three main aims. We equip countries to have a plan for vector control interventions, which include bed net distribution and indoor residual spray campaigns. And then we also conduct robust entomological monitoring. So what is the mosquito doing? Is it transmitting malaria? And then all of that with the main goal to build capacity in how to use all of that data. Among the countries we use a global DHIs to dashboard called Vector Link Collect. In 2018, we started using, we built an event program for indoor residual spray data. We're now up to 17 countries. And then a year later, we rolled out five event entomological programs. And now we're up to 19 countries. In both of these datasets, we use mobile data collection and paper data collection for IRS. We use ODK. And for entomological data, we use the DHIs to Android capture. So some of the things we had to think about when we were considering mobile as a solution for entomological data. It's super technical. We, I work with PhD people all the time now. And it requires the actual data collector to know WHO recommended and SOPs to be able to effectively collect the data onto the phone. It's also, we have five different programs. Each program has very distinct data workflows. So some data is collected hourly, some data is collected monthly, some annually. And all of them require multiple data points of collection and different people at different physical locations. We also, each country, as we all know, this is very different. So there are different security concerns or internet connectivity concerns to take into account when you're planning out a mobile data collection system. And then also a lot of the Sentinel sites, so where the actual mosquitoes are collected, can be super rural. So we have to be able to have the capacity to collect data offline. So we, because we had in-house experience with ODK, we did do a landscape assessment of other softwares. But we primarily compared DHI through Android Capture and ODK as our two main solutions. Primarily, the biggest advantage was that we wouldn't have to recreate our programs onto the phones if we use the DHI through Android Capture, as well as we have the ability to sync automatically into the system without using an ETL, which has proved to be challenging in the past. We are using 2.4.3 right now. We're hoping to upgrade in the near future. These are some of the other criteria we took into account. It wasn't everything. It was a very comprehensive list, but these were some of the things that we considered essential for mobile data collection. So last year, we did, we piloted DHI through Android Capture in Ghana. We did a very comprehensive, well-documented pilot of three phases going through all the programs. And based on the successes and lessons learned in Ghana, we moved pretty rapidly into Nigeria, starting of March of this year, and then Cote d'Ivoire and Mali. Nigeria was one of our newest country add-ons into our system, so they went straight into mobile. They never had to do paper data collection. Cote d'Ivoire did the full transition from paper to mobile, and then in Mali they've done a smaller rollout. So we've been able to really have great success in rolling out the application, and we are confident we can move forward with several other countries. So now we'll go through the four countries in a little bit more detail, just highlighting some lessons learned that we've had. In Ghana, we did have existing, we had in-house experience with mobile data collection, so it was fairly easy to roll it out in Ghana. We did discover that users have the ability to edit categories in using the DHI through Android application. Our event programs have a lot of categories, and so this discovery actually made our database managers life a lot easier in terms of data quality management, because they are able to use the phones to catch those errors sooner in the data workflow. Because Ghana was the first country that we introduced mobile, we've actually already seen improvements in terms of team's efficiency. They're able to decentralize their mosquito selection, so now within two hours of synchronization, the database manager can see which mosquitoes have been collected, communicate with the entomologist, and then disperse information back to the Sentinel site teams and tell them this is the mosquitoes that need to be processed for molecular analysis, and they all love mosquitoes. I just saw this process not too long ago. It is infectious, the passion that entomologists have for mosquitoes, so it was a lot of fun to see. Then we moved on to Nigeria, and with Nigeria, we learned really about how unique each country is. The process of mapping out your workflow should spend a lot of time on this process. We developed some tools that help us facilitate what questions to ask our teams to really understand, okay, where does the mobile phone need to be if this person needs to collect this information and then sync it at this point for technical review. We're lucky that we work directly with the field team so they know the reality of what happens in the field, and they can inform us, okay, this is what happens here, so then we can say, okay, well, then the phone needs to be in this location or transferred. One of the things that has helped us greatly to be able to roll this out in different countries is the ability to edit, open and edit saved and synced events. One thing that DHIs to Android capture offers is this ability to sync at multiple times, so it minimizes your data loss. This is not something that's available with ODK because of the ETL, so we're able to have multiple points of syncing to save data, but then also multiple points of technical review of the data. So with Code2Var, we recognize some connectivity issues that made it necessary to do a more comprehensive connectivity assessment in all of the sites. This really pushed us to look more into what happens when we restrict the programs per device and what happens when we restrict the organization units per device. Minimizing the metadata that's needed to be synced on each device actually lowers the need to have fast internet and can greatly improve the ability of the user to make configuration differences on their device. All right, and then Molly, last one, is when we really recognize the need to have hands on multiple practical training sessions. One of the interesting takeaways was that the field technicians were afraid that by using the mobile device, all of a sudden they were going to be collecting different information, and so it was necessary for them to see, okay, here's the paper form, it's going to be the same information collected on the tablet or the smartphone, and having them get really used to all the validations because yes, they know maybe the SOP, but they don't know maximum number of mosquitoes that can be collected at a certain point, whereas the person entering the paper data collection into the browser would know that really well. It was repeating this in the office and then repeating it in the field really helped roll out mobile. Some challenges that we've had. One of the things that we're working on is ideally we would want for a supervisor to go into the field and use their device and open up whatever events to be able to to review the data. Right now we're not able to pull up events on multiple different devices. Working theory right now is that we have so many categories that might be inhibiting this action, and then we also we're not able to start with the 2.5 version of the application at the beginning of this year, because we do have one of our programs has about 1,400 data elements, so it was causing delays in terms of data registration when they were going through the program, so that's why we opted for 2.4.3. And then some of the some of the successes or the reasons why it's been such a success, we've really been able to streamline the the data workflow across all these five distinct workflows. It's you know it's increasing the availability of data in in our system. Let me go through all these here. Here we go. And it goes straight into our dashboards. So our dashboards are pretty pretty standard right now, so people are very used to how the data works in the how it looks in the system. And so there we have technologists that are able to review the dashboards and make ad hoc recommendations to the National Malaria Control Programs and getting this data faster and cleaner into the system allows them to do that. And then I think this is my last slide. I just wanted to round it out with giving maybe the the audience a little bit of info of why we use entomological data because maybe not everyone knows, but we do a lot of data collection on what does the mosquito do, how does it behave, and what hour of the day. And this informs all this informs how to make vector control control decisions in the country. We also collect data on resistance. So if there's resistance in a certain area, does that insecticide need to be rotated out into a different a different type of insecticide or in a different location? And then all of that, we also monitor whether vector control interventions actually reach the demographic area demographic people population targeted population. So all of this informs National Malaria Control Programs in deciding what is the best vector control decision for a certain location and a given in a certain time and taking seasonality into account. And all this is available on our dashboards. We do share regularly with PMI National Malaria Control Programs as data is cleaned and ready for for use. Thank you. Service delivery data to optimize continuous distribution of all long lasting CPC bennets through entomological care and immunization. Thank you. Thank you, Stefano. I hope I'm audible. Yeah. Good afternoon, everyone. So as introduced, I am Ismael Colleni from East Tanzania. I'm going to be talking about more on the data use side of things as far as the LLIN distribution is concerned in Tanzania. So East Tanzania has been working with Tanzania Vectoral Control Activity. This is a PMI USAID funded project, which is mainly targeting to improve evidence based vector control approaches in Tanzania. So it focuses on a couple of things, but for the purpose of this presentation, I'm going to be talking more about the insecticide treated nets ITNs. So this activity covers both Tanzania mainland and Zanzibar so it involves working very closely with the government and all the implementation is done through the National Malaria Control Programme for the case of mainland and Zanzibar Malaria Elimination Programme because these two parts of the country are in different stages of malaria control. So for the purpose of distribution of ITNs, the target of the focus is to pregnant women who are attending the ANC first visit. So that's one of the core first distribution area. And then the second one is the children who are receiving measles one dose of vaccination. And then the third is community distribution which is done through coupons and they go to the health facility to receive the alliance. And of course on demand there is some campaigns which are done from here and there depending on the needs and on specific interventions in different areas. So all of this data across distribution, all the data is collected and managed through different systems by the government. The core one is DHS2. So there's HMIS which collects data, all the service delivery data and part of the data that is being collected at these two programs. They also collect the distribution or the services as far as the ITN is concerned. So we have supported building up this dashboard mainly to focus on the again as the keyword is evidence-based decision-making is to focus on providing the necessary information so that they can be able to act or to perform different and monitor the performance of the program across the board. But also another keyword is accountability. From the previous implementation there was a lot of bed nets being lost and you would find them being sold in the shops. So one of the key also focus as part of the design of this dashboard is to make sure that it fosters on transparency and accountability so that you can easily be able to detect whenever there is lost or potential theft of bed nets and they end up being sold in the market from the private sector. Another key crucial component as far as this dashboard is concerned is it combines data from two main data sources. So the HMIS service delivery data which is already in DHIS2 but also logistics data which is from LMIS system and this is based on an open LMIS in Tanzania we call it ELMIS platform which is not DHIS2. So we managed to do this by having the two systems DHIS2 and ELMIS interoperable but of course all the data analysis visualizations and all the outputs as far as this dashboard is concerned are done and accessed through DHIS2 dashboard interface. So again the key aspect of this dashboard is to be able to support the program implementation and performance monitoring and this is of course giving access to the required data to different levels as you know the program is implemented by different levels and each level looks at different levels of data so the district level would have some specific information which targets them but also it goes higher level up all the way up to the national program and of course there is a lot of data driven action decision making and strategies across the way. So this is one of the example of the visualization which shows the trend of distribution of bandwidths over different months. So the blue bar is the ANC visits for that particular month and the red bar is the ANC visits that were actually received L Alliance and then the green line is more of the expected number of pregnancies which is derived from the population. So you can clearly see that from the ANC distribution point the bars are pretty close to each other but on the other hand this is another chart which looks at the another level of distribution point which is distribution through measles one vaccination and here you will see the gap is a little bit higher meaning that there is a lot of children who are actually vaccinated and they are supposed to receive bed nets but they don't actually do that. So this is one of the things that through the use of this dashboard it was kind of noticed this is the one of the district but if you go to the high level national level you see the actually gap is quite high so the national malaria control program had to find the best strategies to actually intervene and make sure that the children are also receiving L Alliance and you can clearly see of recent months this gap has been slowly reducing over time. So this shows how also the use of dashboard can actually make the program managers, the implementers, re-strategize and over time you cannot expect in one month the gap to be clear but you can clearly see over time that the gap closes and hopefully soon it will be at the same part as ANC distribution point. As I said earlier on one of the key interesting functionality is it also provides an opportunity for the system or program users to triangulate data across two different data sources that is HMI service delivery based data but logistics data which is coming from another system and there are a couple of visualizations that we made that to compare but one interesting one is this one so this looks it's a scored manner but it kind of compares and calculates the difference and actually there are some color coding and this helps the program to by the use of colors of course like red and then they can prioritize it means that there is a big gap between what is shown on the LMS versus what is shown on on HMIS and that triggers a question of why because ideally if you provide of the facility distribute 20 LLINs in a month to maybe 20 pregnant women and then the logistics the stock status and the data should show the same so through the triangulation of this kind of data and then it also able to provide opportunity for the program people to pick up these triggers and actually either give feedback or intervene because it's also a potential that there could be some some leakages of LLIN over time and of course the last column which is blue this is a month of stock it's more of a logistics space indicator which predicts how many months the facility would be able to last with with that stock that they have yeah so this is also a bar graph which kinds of compare again the the distribution of LLIN from from HMIS source and then from logistic source you can see there are some months that they are very close to each other but you can also identify that there are some months which there is a big gap so this also because again ideally this this numbers should match very closely to each other of course we do understand that there are some some data quality issues that can lead into into discrepancies but this kind of very high discrepancies rises a lot of questions that the program people will need to find ways to identify what is the problem and and and actually act on them yeah so challenges are part of the experiences or challenges they they are quite diverse amount of challenges the biggest one has been through the interoperability process and and it's because of the mismatch of reporting period so ELMIS is designed and done in Tanzania on a quarterly basis currently it's in the transition to from quarterly to bi-monthly and the HMIS based service delivery data is monthly so now you're kind of looking at data monthly level data versus quarterly level data so that's one of the biggest challenges that really lead into into some gaps in terms of the use of the data another challenge is maintaining the interoperability as these two systems are supported implemented by two different departments within the Ministry of Health but also two different technical teams often the interoperability breaks and at the end of the day DHIS2 has a receiving end and at the dashboard as the program people uses the dashboard there is quite a lot of gap in terms of the data is not coming as expected for so many reasons it could be some upgrades and there are changes in in the systems across each other and oftentimes some some aspects that are happening in either of the system and they do not inform the other one and then of course one of the other biggest challenges data quality issues still this is a very cross-cutting generic problem not only for malaria or in the land but it's still causing a lot of challenges as far as the use of data because you would set up these triggers and color coding but you get reds and you realize that it's a lot of data quality issues and of course there is still some distribution of bed nets that are happening from non-facility based in other in other distribution points and this data is not properly recorded within the health facility system and that again causes a little bit of leakages as far as when you're looking at the holistic distribution information so thank you very much I hope I am on time and I'll pause for me thank you so thank you very much we are very on time for 20 so just a logistical information there will be from 4 30 the expert lounge just here where there was the pizza yesterday and if there is any burning question maybe we could take a bunch of two minutes yeah okay so we repeat the question for the ones online also I was asking if the data from the from a massive distribution was always also captured as well on this dashboard or not no no did you respond to this thank you so yes the system design massive mass distribution or any of the community distribution is supposed to be done and the data is supposed to be collected by specific facility because at the end of the day the health workers that are distributing these airlines are actually coming from the facility so there is a like by system design it's supposed to be captured under the community distribution within the health facility but in reality that does not happen it's very controlled within the facility but once they go out and then they would hand over the nets and they don't report back to the facility so that's the challenge but yeah it's more of an operational issue than than the system design but yeah perfect thank you very much so now we run out of time in case you have any other question please say you can write on the community of practice and for the ones on lives if you have any question you can just catch our presenter in the coming hours coming days thank you very much