 We are now live welcome everybody and I know there's people still coming in but I'd like to welcome you all to another webinar in the DHIS to immunization webinar series, which is organized with the support of Gavi. This webinar is going to focus on the highlight and share the DHIS groups approaches to obtaining population denominators data for use in immunization programs. We'll go to start with a systematic scoping review that Dr Wilfrid Signoni, Kofi Siljadin and I conducted to look at what denominator data is used in practice. And that'll be our first presentation by Kofi Siljadin from the HISP West and Central Africa HISP team. Then we will look at two of the areas that are not as highlighted in the literature that we reviewed. And this will be an example from Rwanda, looking at the work of linking the CVRS with the DHIS to for vaccination Rwanda. And Jean Paul had to give Kimana from the HISP Rwanda office will present that. And the third presentation will be again one that's not as documented in the literature as we would like. So I would like you to hand over now to Kofi, who will give the first presentation. And can I remind you that you can ask questions within the chat here, but also as you see what's posted in the chat. What's posted in the chat is our community of practice. So please answer questions, ask questions on either forum. But if you put them on the community of practice, we'll be able to address those as an ongoing question. So it might be in terms of kind of continuing debate to post them on the community of practice. We'll try to get to the ones in the chat here. If not, we'll post them at the end to the community of practice as well. So over to you, Kofi. Thanks, Alain. Good morning and everyone. Good evening, depending on where you are. I'm happy to share with you a summary of the coping review that we did together with Alain and we freed regarding the approaches to calculating the population denominators. And a little bit of backgrounds that everybody knows that the lack of the denominator data is identified as an obstacle in targeting populations that are not reached by services, essentially zero doses and excluded. A bunch of alternative domain to that data has been used across the programs and at across the levels of the health pyramid and the differences are based either on numerators or denominator data definition. And sometimes you may find differences on the guidelines that are out there and what is being done on the field. You can have some disputed of a census data and have some divisions in the countries. And there are challenges to estimating the dominators for catchment areas when it comes to sub-level, sub-level at the sub-district levels. And casking behaviors can be very different across the catchment areas. For example, for nomad population or for people that are living in villages that are close to borders, country borders. So when we want to solve this denominator gap, you have a number of choices or options that are available to you. If you're from the national census data or for the server-delivered data, you can take community or household sources. For example, the sources that come from mass campaigns, you know that sometimes usually when you want to do those bed net campaigns who undergo animations of population, you can use civil registration data. And also you can use geospatial data. So what do you choose between all those options that you have? So your choice may depend certainly on the availability of the data source that you have. And on the context, you want to use it and the appropriateness usually refers to how purposeful is the denominator looking for. And are you looking for something that relates to service coverage? Or are you trying to map specific population coverage to look into zero doses or uncovered populations? Or are you trying to calculate a global indicator? So all these things will help you make a choice between the options that are available to you. And remember that sometimes the denominator may be suitable for the regional or provincial level, but that same denominator may not be the most suitable at the sub-district level as shown in certain of the articles that we have read. So how did we proceed? Our review question was how the denominator population is calculated in routine health management information system and at the sub-district level, we specifically looked at the sub-district level and we used the Prisma approach to inform or to run our process. We had retrieved articles from Ovid Medline, all the whole database from NBase and from the Web of Sciences. And we in total retrieved over 300 articles and we searched from the terms of denominator and health information system. NNOTE was used to remove duplicates and then what we have retrieved was uploaded to COVID-19 for the review of the summaries. And each article was reviewed by two authors, we were three of us. So when there is a dispute, the third one was coming in to solve the dispute. So by the end of the day, we have included nine articles. For those nine articles, we did a full review of the article. And we found that the mentors were more in global health and health service journals, but also in some specific programs such as EPI. And the publications we have included are mostly from 2017 to 2021. Mainly the articles were regarding Eastern Southern Africa with three studies conducted in Uganda and one in Bangladesh, Nepal and Tanzania. Mainly the program was maternal and child health. What have we found? Regarding routine health information system, the most common source of data was from maternal and child health services such as internet or care or DTP vaccine. And the most common challenge is where the data quality assessment and adjustment that were needed regarding the coverage, the travel time and the quality. We found that in several articles. And near the universal coverage intervention need to be selected because that was the wills described by minor and some written data were better than others. For example, if you compare the agreement between demographic health survey and census, you may have more agreement between ANC and BCG than between vitamin A or measles in Uganda. And census based coverage were better, the estimate were better at the national level and routine data coverage were better at the sub-national level. Similarly for the routine health information system, different sources and data can be used in between census to get reliable estimates such as surveys and routine health information system. So minor and all provided a four step approach to adjusting routine health information system. First of all, you should obtain data from different sources and then you access them and adjust them at the facility reported data so that you can get adjusted numerators. You should improve your denominator by computing a lot of target population based on health facility data. Then after you have gotten your adjusted numerator and your improved denominator you can calculate more accurate indicators. Regarding the census, we have noted that there were two approaches when trying to estimate the adjusted population from census data. It might be simple projection from the baseline assuming that there is a linear growth rate, no adjustments needed. But sometimes it can be more complex methods that help adjust for age, for geography, for mortality, etc. And that is normally or generally done by the national statistical offices and that is done in between census so that you can have adjusted numerator over the years. The challenges are that these kind of calculations are not generally undertaken for smaller geographic areas at a sub-district level and internal migrations or functionality are not correctly accounted for. Another challenge is that those national census are not being done at a regular basis because you know that normally it should be done each 10 years. It requests a lot of resources in terms of mail, in terms of operation, in terms of finance, so people or countries are failing in doing regularly those 10-year census. And finally, the population estimates, the problem is that the consequences that come from the population estimates are so huge that political interference comes in usually and it can be a bias to the result of the census. Regarding due special data, we have found that bottom-up gridded population use and facility return data are more reliable in some areas, and particularly when it comes to small geographic areas. And those due special data are more useful and very useful when the other sources are either incomplete or unreliable or not available at all. And you have to know that it's not a universal result when used by some voters and essentially in terms of reliability of the coverages. And we have concluded that projections from census data will be more accurate when we use them along with geospatial data. So how can we summarize our overall finding? The first thing is that more than one source is used. And then people tend to use multiple sources either to assess the quality and the reliability of the denominator or because some sources may be more accurate or more appropriate regarding certain programs or purposes. The most common sources of the meter, population meter are either projection from national census or estimate from health service data. Demographic health survey can also be used as a common source, but it is used in conjunction with other sources. There has been a need to develop guidelines on how to adjust, to choose different options of denominator or how to adjust the denominator for calculating more accurate indicators. And the example was provided by Meinand and Al. We showed that example on slide seven. And we have very few examples of the use of geospatial data and CRVs registry at the national scale. So by the upcoming presentation we'll have us look on later on those two less documented approaches to obtaining population data for the delivery of routine immunization. So this is what we found from the scoping review. I'm going to turn back to my colleague John Paul to continue on the experience of CRVs in Rwanda. Jean-Paul, it's up to you. Thank you very much. Thank you very much. Thank you for this presentation, after its kind of continuation. Allow me to share your this principle and continue. Yes. Good morning, good afternoon dear participants. My name is Jean-Paul from Rwanda. As my colleague was presenting, I'm going to share with you specifically the example of how his Rwanda is collaborating with Minnesota Health and the other partners to put together data and technology to come up with the solutions which can enable data use and investment decisions. So I'm going to take you this example just showing you how we are enabling the API data use and investment decisions by supporting systems integrations and also use of data use applications here, I mean mainly SCOCADs. So Rwanda is a background here in Rwanda. All public and private facilities are collecting data. They are using different systems. Mainly all facilities are using electronic immunization registry for collecting data related to vaccination. And also they are using a monthly data set in HMIS, doing the collection of vaccination data. All facilities above private and public are using a module of CRVS where they are correcting birth and death. So that's about Rwanda Bonta correction, but also we have the existing data sessions here in the country where facilities are discussing about data. So these sessions also are opportunity and through these sessions people are discussing about data to measure the progress by comparing to their targets. So these sessions also are really helping for data use, but also we can mainly highlight three kind of data use sessions. Here it includes monthly data coordination meetings. These are the meetings at hospital where all head centers are meeting and their supervisors from hospital, then they share the current key indicators of their choice. Then they discuss about the progress, about the coverages, but also they discuss about the data quality. And also we have a data quota meeting, which is DHMC, district health management team, where all stakeholders of the district who are involved in health information management and health sector, they are together on a quota basis to discuss the key indicators of the district. And then the third one is Kotari API data review. This is a specific meeting focusing on the review of API immunization data. So they discuss, they first check the quality, but also they discuss about the status of the indicators. So all those mentioned discussions, they are to promote data discussions and evidence-based decisions. And then what are the opportunities here in Duanda? Just here we mean opportunities which can help to strengthen data use and promote evidence-based decisions. There are three main opportunities and strength. The first one is that the top leadership here promotes the country of using data for each and every decision making and planning sessions. Second one is that there is an existing ICT, porous and luteinous information management SOPs. Actually these are the guidelines because without guidelines you can't be able to promote the availability of quality data and data use. The third one is the existence of data review and discussion sessions in the press as mentioned above. Then what are the challenges behind this process of promoting data use? I categorized these challenges into categories. First category is related to denominator related issues, the other ones related to the schemes. Then denominator related here in Duanda, we have national coverage for under one year. Just to calculate the under one year target population, facilities we are using the same under one year targets and we have by census, we have different under one year targets by district. So that was the challenge about denominator. The second one was that denominators were calculated using the forecasting census based. Then, for example, the existing types of nominators here, we mentioned the census based nominators, the BCD based nominators. That was a kind of a challenge. Then about the schemes, we have different analytics tools in DHIS too, but we used to observe that users preferred to use to extract the data from DHIS too and manipulate data using Excel. So we noticed that they were a kind of lack of skills about how to use these analytics tools without dealing with Excel. So visualization of coverage and dropouts was an issue. Then it wasn't easy to use tabular reports, but not the issue of technical side, but the issue of scarce gap. And what's now being done to overcome data use challenges, including denominator, both passivity and national review. By collaborating with the API program, we decided to import to calculate growth, the target population under one year by district and import those populations so that each and every facility within the same district will be able now to use the real under one year target using, calculating the under one year using the national growth data, which was at some point giving some facilities, high coverage, others low coverages. Second one was to create integration between CLRVS and immunization registry so that now CLRVS goes as I said before CLRVS now all facilities are registering in real time or birth. So those birth registered in CLRVS, we thought that it's a good idea to integrate CLRVS and immunization registry that CLRVS would be pushing birth in real time so that those birth can be used as the real denominators. Then we are now developing the report builder, which will also use to improve data use and real time reporting. The main steps now, during the integration of CLRVS, but also the implementation of data use apps as we said. First, the implementation of data use apps was done in the following summarized steps. We, it was first we did an assessment, second we configured the API indicators, third we configured the MCCCH scorecard, then we trained the central remote team, then we documented the process, then we did the TOC, then we did the end users training. We talked about the implementation of data use apps, scorecard, botonic analysis and action tracker, and those apps are now being used at all facilities in all districts, and they are being used to promote data discussion sessions, data use, and also they are helping to in taking real time decisions. Then for system integrations, we did integration between CLRVS and electronic immunization registry. Then we did the TOC, but currently jointly with the API program, we are conducting different mentorship sessions at facility to reinforce the implementation of this integration after this integration. Then, this is the schema, if I can say, or the drawing of the integration. Let me explain a little bit about this integration. Actually, the CLRVS is a module from the national population registry. So, only the two modules for birth and death is the one being recorded at facility level. At facility level, after each and every delivery of death, data manager at facility level records birth. Let's start by birth. Then after recording the head of the institution is the one to approve. Then after approving this birth or birth, I post in this program, we created, if I can call it an immediate program, where all birth are posted from the CLRVS. After being posted from the CLRVS, then the user, that's the one who is going to vaccinate. In this birth registration, and he looks, he search, or just there is a search option to search a newborn who has been pushed from the CLRVS, and he send this birth to immunization registry for adding different events. The event here means different vaccines. And then after adding those vaccines, after registering, after allowing the child's new birth in the immunization registry and adding the events, they give them the appointment for the coming visit, and so on, until the child gets fully vaccinated. So there are other highlighted sources of information here. For example, once a child is recorded in CLRVS, a mother is going to get the birth registration from the Irembo. Irembo is an e-government online platform where most of the services are provided here, and it's an integrated platform where most of government services are provided. So once registered, a mother or parents of this newborn are going to this Irembo in government service to get the birth certificates. Then you can see that here we have also another source of information is my friend Kofi highlighted. We also have other sources of information which can be considered or being used as denominators. As we have said, you have national population census, we have other NISRA, this is the National Institute of Statistics. So from here we are getting different denominators from the surveys. So that was a short discussion on the integration. And then, okay, okay, then the next slide is about CLRVS, what was the integration after integrating, what was the challenges after integrating CLRVS and immunization. One is that immunization, some records were not updated in your time in immunization. Because after integration, some, we still, we kept observing that even though we created this integration, but it's not updated in your time. Secondly, after assessing the success of the integration, we noticed that some of birth were not being pushed immediately from the CLRVS to immunization registry. Then what was the mitigation strategies or practices, what we did. About the records which were missing, we conducted training at hospital, API supervisors. These are people who are overseeing on their basis the coordination of vaccination activities. They are coordinating vaccination in their respective health centers. Then after training, also we trained the head of health centers, because the head of health centers are the ones overseeing and coordinating the activities at the health center. And for the issues of the, some of those birth, which are not being pushed from CLRVS to registry, immunization registry. In collaboration with concerned entities, we are updating the API so that now it can push all birth from CLRVS to immunization registry. And now, jointly, we are conducting different mentorship sessions at PASIT in collaboration with the MSOVS and RBC to improve the leadership engagement. That's a kind of awareness, but also we are collaborating with implementers at PASIT-EVO to correct their any puts to see whether something is in there to improve the system functionalities. And the next is what are now the next steps to keep promoting data use to strengthen this integration, of course, by promoting evidence-based decisions. One is that HIST-1 in collaboration with different partners will keep participating data review sessions at both national and sub-national review. But also in collaboration with different partners, we are planning to organize a webinar for data use with HIST-PASIT so that we can promote a culture of peer learning. And through that, we will be receiving such stories, but also for cities we will be learning from others for different approaches being used for more data use and evidence-based decisions at country review. Then we will keep implementing continuous documentation and data use initiatives. Then systems integration and denominator aspiration for data integration, conduct co-tariff review, visit to assess and correct different user feedback. Then review the existing scorecards. Actually it's kind of strengthening the implementation of data use apps. And also we are now going to strengthen the bottleneck analysis implementation and action tracker at the district and PASIT-EVO to strengthen planning sessions by using data. Because I swear we can strengthen data use because we found that we need to strengthen these planning sessions by using this BNA and action tracker, which are the apps really which can help to plan using data, available data. Then this is the end of the presentation. I think you have tried to put your comments, questions in the specified channels. I can stop by here and maybe give a default to my colleague, Sam, who is going to proceed with the other presentation. Thank you very much. Hello and Sam, I'm just going to jump in there before you start. So I'm Paul, I thought that we've got some time and I just wanted to ask some of the questions that are coming up in the chat specific to this. Some of the other questions around issues around migration seasonality, over 100% coverage, workload on health professionals with these kind of new apps. But some of the questions specific to this were around how widely used is the CRVS and a related question in terms of the registration of home deliveries. So what kind of percentage of births are registered within the CRVS and what about home deliveries. Yes, yes. Thank you very much, Erin. Starting from the home deliveries, those birth facilities are being those being registered by data manager to facility and being approved by the head of the facility. Because those are actual, let me start by saying that the home delivery is, we do have a few number of home deliveries, but those home deliveries are registered at cell review. The head of the cell is the one responsible birth registration for those few cases in community. And once they are registered there, they also push in CRVS. So the first thing about the registration process, you see it's CRVS in a facility review. Once a new bone is registered is pushed to the program called birth registration, then the vaccination provider will be looking. They will request new bones in birth registration and allow them immunization. It takes not more than one minute to move a new bone from the birth registration to immunization for adding different events. So considering the workload, those in charge of vaccination are the ones who are also recording. And for information now from this October, we are going for paper race. Since the 1st of October now, we will not be using the parallel recording, because currently they were recording first registers and secondly in immunization and immunization, but now we will be only using the electronic and immunization registry for recording all related vaccination data. Thank you very much. Okay, thank you. And maybe Sean Paul, you can address some of the questions in the chat there. I'm going to hand over now to Sam. I know there's a hand up and some questions in the chat. But if Sam can present and then we have time at the end, we can get back to some of those broader questions. Thank you very much, Sean Paul, and over to you, Sam. If you could stop sharing so that Sam can share. Thank you. Okay, hi. So, thank you, Ellen and Sean Paul, just to make sure you can see my screen, right? Yes. Thank you. Thank you. So, good morning, good afternoon, and so good evening to everyone. So my name is Samekam Bupau, you can call me Sam. So I work for Ips Vietnam, but based in the Vien Tien Lao PDR. So I'm happy to present our work to support the Ministry of Health also work with development partners on this to identify the excluded. So and how we use the health information system to achieving a better representation for the population. Okay, so before I'm going into the details so just a brief generic definition of the exclusions. So, in short, for our work here the exclusions we mean that those vulnerable people that do not have access to health and social services. So in this presentation that I will explore the subset of that which is the how we use the HIS to identify the zero dose children. So the zero dose children so that are the children that are missing one or more vaccine doses. So, that part of the questions also come from the country needs to improve its vaccination rates and also our own research questions, because we think that to identify these excluded, we need to use the HIS and the HIS can make those visible. And then if we have the visibility we can mobilize the resources and also the services to assess the situation. And also just keep a big background on the loud PDR and maybe the HMIS. So for loud PDR in terms of the vaccination rates are poor. So it is estimated that less than 90% of all children have been vaccinated for any childhood vaccines. But throughout the year, various strategies have been developed to address this gap and improve coverage such as reaching every child. So this has led to a number of community-based interventions such as the door-to-door vaccination, delivering vaccines beyond infancy, focusing on marginalized populations, but often the process is quite labor intensive. Additionally, identifying the zero dose children can be difficult to calculate. If the target population denominators are not accurate and importantly, it has to be agreed across programs and level of health system. But despite the increased number of different community-based information systems, the majority of routine national HIS have data from health facilities as the lowest level of data collected. As a result, those not accessing formal health services is not captured. So even for HIS that partially include community source data, the quality and coverage can mean that data is still missing and marginalized are still not represented in the formal HIS. And in terms of the low HMIS, so low HMIS, low MOS endorsed the HSU in early 2014 as the National Health Management Information System. And since that, the aggregated data from NIE programs and all the public health facilities collected nationally by the end of 2017 and up to now we have almost more than 12 aggregated programs as well as the tracker which I will explore later. But for the EPI, the EPI as I mentioned and also as reported from the Gavi, the law has the EPI coverage among the lowest in the region. So, and because the EPI data mainly being used for reporting purposes, later than for action or decision making. And for the EPI programs in from the beginning until 2018 they were using the Excel based reporting in parallel with the HSU. And the reason they are using Excel because in the HSU we can only calculate the service health facility service coverage where EPI they also need to see the population coverage. So there are some problem or there are some requirement from the EPI that they also would like to monitor the stock and so also campaign data which was not indicated at that time. As you can see from the image, because there are many logbooks so data quality problems are coming from that and health center before making the report they need to consult different logbooks to generate indicators. And those are just the context and the reason why we would like to explore if we can use SNRIS to reduce this one and also to help to improve the vaccination rate. And in terms of the solution we work with MOS and as I mentioned development partners to introduce two solutions. So first of all we included family information in DHS too, but family information is the project of MOS since 2008. But at that time they were collecting using pen and paper. In terms of family information, it's just the work that each year the health facility health workers, they will visit each village and then each household and to check the family book and also record the number of people in the family and also order information such as latrine sanitation, also water, same drinking water, etc. So once they record in the paper then they will enter into DHSS too. So once we include this activity, we made some changes in DHSS too. So with these activities, because each health center they will have the assigned villages that they need to provide services. With this activity then we work with other ministries like Ministry of Home Affairs to have the standardized village list. And with that then we also include that village to down off unit as the for data entry. So once we have the list of all the official villages in DHSS too then we can enter population data for each village. And then we use this data as the base for the immunization which is the electronic immunization register that I will also talk about next. So like I mentioned with family information, not only we capture the population, we also have the other information that linked to the household regarding the water or sanitation, etc. But with these activities the consequences that we now we have the denominator or the catchment population available at the village by year, because in the past we only rely on the census and the census can keep us only data at district level. So with these activities not only we get the data at the village, we also use this data to compare with the census if there are some missing here. So now that we have population at the village we can calculate according to the EPI needs. So I mentioned previously that EPI they also need to see the population base or geographical base coverage. With this one we can also generate indicators based on the service provision or the population base. And as the outcome now that this program can answer all the requests from the EPI program we stop the parallel reporting. So that's for family information and just a bit to discuss on the family information as you can see, we compare with the census. The latest one we conducted in 2019 before COVID as we compare with the census we can see that almost 93% matching with data from from last statistic bureaus and also to show you by the provinces. So on average it's about 93% and some provinces that reaching as high as almost 100%. So by visiting this one then we can know that the population in each village and also including the children. So that's the first part. The second part is about digital register. Now that we have at least the list of children in each village. The second part we also introduced is we call the digital registers. So the digital register is just the way we capture information in using DHS to now we capture information. Then we can compare it with the data from family information. But for digital register we are trying to capture children from because in the Laos we have villages in different zones. So we have zone one, zone two and zone three, but zone two and zone three are the ones that are hard to reach. Before we had this one is really challenges to calculate and also to capture data in those areas. So to capture those data we also have the offline that people at the health facility can visit those villages and enter data and send it to DHS too. So then coming back once data is captured and then the data capture is more than just aggregate. We have data bird, we have sex, we have names, we have the village. So with that one then we also can link to the village and also using family folder we can project and also matching those children. So just the conclusion from these two activities so with family information and also the digital registers we can address or able to identify this zero dose that the children that missing one or more doses because we combine these two for for the digital register we can get the number of immunized children. And then the family folder we know we can have the accurate estimated number of children down to the village level. So by combining these two we can calculate the areas of low immunization coverage. We can also use this to, like I mentioned to identify the zero dose children that we also have the more accurate drop out rates by village. But however there are some challenges that we are facing. So the first one is also the cultural, social and health service barriers that need to be overcome. So have to be have the demand side as well. Also addressing the non-serve is not as simple as just identifying them and then providing the services. So needed action would involve incoming overcoming cultural barriers by active health promotion and education among the groups with low participation as well as engaging community leaders and political levels. And the transformative aspect of digital inclusion for social justice, therefore not only about digital representation, but needs to be understood through the wider aspect of automating and implementing dynamics addressing the social levels. And those are just my presentation on the world that we help the Ministry of Health to identify the missing children or the missing, the student that missing the vaccines using the family information with the digital register using DHS2. Okay, thank you. And there's any questions and welcome to answer them in the chat and also in the community of practice. Okay, thank you. Thank you. Thank you. Thanks very much, Sam. And we have time just for two questions. I'm going to leave some of the specific questions to address in the stream, but some of the interesting ones I think coming up that cover all the presentations. Or how do we deal with migration and seasonality in in both the both systems. And also, what do we do when we've got 100% coverage I know. And coffee you responded within the over 100% coverage the whole I issue of kind of care seeking behavior and some people kind of going to different facilities. And in terms of either better or perceived better services or just easier access. And, but I'd like to open up to Jean Paul and Sam as well, in terms of migration and seasonality how you manage it. And also, you know, is there a process you go through when you see this over 100% service. Now we only have five minutes that I can only give you two minutes each. So maybe, Sam, we haven't heard your response. In terms of migration seasonality how'd you capture that. Sorry, Sam or maybe Jean Paul, maybe if you can come in on, you know, have your particular process for migration and seasonality. Sure. Thank you, Erin. I can see that there are different questions and maybe curiosity but then you start by that seasonality. It depends on the type of indicator that we're just measuring. For example, when it comes to malaria in that case you can think about seasonality cause the malaria depends on seasons. And when you analyze it's a good also to understand the area. And here, if it's a country, if it's a province, if it's a district, for example, for example, let me give the example of Rwanda. When you analyze in the malaria in Northern province, it depends from malaria cases in Eastern province in between January and March or April, June. So that a kind of seasonal where you need to understand the type of the indicator comparing to the season. And from there you can come up with an analysis saying that this increase is normal in this season, but this increase is not normal in this season, depending on the type of the disease of type of program you are measuring. I think that's how I can respond to the way you can deal with denominators and seasonal variations. The key message here is to understand the type of the indicator we're measuring, comparing to its nature in terms of variability according to decision to season. Thank you. And then you had gone through Jean Paul, the various mechanisms and processes that you already have those meetings you have. So sorry, Sam. Yep, sorry, I forgot I was muted. I can cover the migration part. So for that for us because we are using tracker in DHS to so there are two parts. So in terms of the system, once the people migrated or moving from one, let's say one district to another. They moved that using DHS to transfer. But in terms of the paper, because once they, they move, they need to carry paper with them and register and also to report in terms of the service they need to also visit the health facility and that facility will notify and also to to update on the system. But in terms of the over 100% because we also capture the address. So once the one people move, we also update the address and using that we can also calculate the percentage. That's why if that's clear. Okay, thanks. Maybe. Sorry. Okay. Quickly then, Jean Paul, we're at the end. But yes, last word, Jean Paul. Yeah, last one just about the migration. The migration here in Rwanda. Once a newborn is registered in Sierra Vista descent to the immunization for continuation of service in Rwanda. A newborn is allowed to get vaccinated everywhere in the country, but where you have received the first vaccine, you can move from that place to other place to get the second, but we have given the right to all users who are accessing the immunization registry to be able to search a child from wherever he has been vaccinated or received the previous vaccines. So wherever I can go to look for next vaccine, he can be searched and found, and they can update that event. So for during the analysis, they can analyze and be able to see how many children vaccinated at this facility. And maybe for over 100 coverages, at Subnational River, they may have the over 100 coverages depending on those caused by those movements, but at National River still will be having the normal coverages. Thank you very much. Okay, that's great. Thank you very much. And I'm afraid we're not getting to all of those questions that are in the chat but interesting ones around kind of the importance of leadership and support. And looking at, you know, questions around length of time, you know, the systems have been in use and doing some pre and post testing. So I think we'll kind of post those questions to the community of practice and the conversation can continue then. So just to thank you all for joining us and listening and participating. And a big warm thank you to Kofi, Jean Paul and Sam for presenting and a reminder that you can also join the French session next week on the same topic. So thanks again for everyone and sorry to the question on presentations yes they will be at base available later on. So, thank you all. Thank you for joining you can continue posting your questions on the COP, and we'll be addressing it them there. Thanks. Thank you very much. Thank you for listening and your questions.