 this meeting will be recorded. I also remind all the participants to post their questions on the community of practice. The link is or will be posted in the chat group. The first presenter we have today is Beatrice. Beatrice is from Kenya. She's a data manager and information systems specialist and has several years of working as a software developer, implementer of free and open source information systems, especially in the HRV in our MNCH space. Beatrice, please. Thank you very much Mahima and thank you very much everyone. Hello everyone. Mahima has indicated my name is Beatrice Akayo, a data manager at FHI 360. I'm happy to present to you today on the topic of how we're using DHIS 200 app for rapid result initiatives, information gathering, for monitoring to increase fourth ANC, scale birth attendance and PNC attendance. Okay, so we'll go today through the topic. So I'll give you a little introduction about AFI OZAZI. I'll give a background, our objectives, based on system architecture, the methods we use, our results, conclusions and recommendations. So AFI OZAZI project is a USID funded five-year project that will be operational from October 2016 to September 2021. It's a consortium of two organizations, that is FHI 360 and Goals per Kenya. The project aims to increase family planning, reproductive maternal, newborn, child and adolescent health impact by increasing access to and demand for these health care services in the project coverage area. So we operate in two major regions in Kenya, that is a Baringo County and Nakhuru County. So for in Nakhuru County, we operate in two sub counties called Kuresoi North and Kuresoi South. And in Baringo County, we operate in four sub counties called Baringo Central, Baringo North, Marigat and Bogota. So I'll give a little background of how we came about adopting the DHIS2 Data Capture app system for data collection. So through the implementation of our project, we noted that in our operational regions, most of the pregnant women are attending first-in-sea visits very well. However, they're still finding it a challenge to complete four-in-sea visits to deliver to health facility by a skilled but attendant and to access postnatal care services within three days following a delivery. And we can see that in 2018, only 49% of the pregnant women nationwide completed four-in-sea visits. Only 65 had a delivery at a health facility by a skilled but attendant and 21% accessed PNC within three days after delivery. So in Kuresoi North and Kuresoi South, only 22% of the pregnant women completed four-in-sea visits and only 41% of the mother's deliver data health facility by a skilled but attendant. 43% accessed PNC services within three days after delivery. So these indicators were low and we wanted to find a way that we can be able to improve these indicators in our operational areas that is Kuresoi North and Kuresoi South. So that prompted us to partner with the Department of Nakuru County to implement a rapid-result initiative, which we call SAHG, so that we can be able to improve the health outcomes of the mothers in Kuresoi North and Kuresoi South counties. So because it was an other way, we needed to have data more frequently than what our national system was providing because our national system was providing data on a monthly basis, but we wanted to have data on a weekly basis so that we can be able to use it for implementation, progress monitoring, and also for decision-making. So we adopted the use of DHS-2 data capture, which is a mobile based application for weekly data submission by the healthcare workers. So our objective was to provide timely data using the DHS-2 data capture app for collaborative review and decision-making. So this is the architecture of our system. It involves the use of mobile phones, which we provide to the healthcare workers. So the mobile phones are Android-based and they're installed with the DHS-2 data capture application, which the healthcare workers use to submit data over the Internet. So the data they are submitting is just the number of pregnant women they are seeing at their health facilities who are attending first-aid visits, who are completing four-aid visits, delivering at the facility, and those who are accessing PNC services within three days after delivery. So this data is transmitted over the Internet to a central server which is hosting DHS-2 system. The data that is collected here is used to generate Excel-based dashboards, which is disseminated through WhatsApp groups and also used during project technical review meetings and quarterly advisory review meetings with the regional teams. So the WhatsApp groups we created consist of regional health management teams and project technical teams and also the healthcare workers. So these are the methods we went through to accomplish this implementation. First of all, we started orientation and sensitization meetings with the Nakuro County Department of Health, where we selected 15 health facilities in Kurosue South and 14 health facilities in Kurosue North that were contributing 80% of indicator performance, that is for ANC-1, ANC-4, SBN, PNC. And then we developed health facility microplans for each of the facilities based on the service delivery challenges that each facility was facing. And then we designed paper-based data collection tools that now could be used by the health facility to submit data on a weekly basis. After that, we went into the system development phase where we developed DHS-2 data sets from the paper-based forms that we designed in the orientation phase. We customized the DHS-2 data sets for mobile data collection using the DHS-2 data capture app. And then we created WhatsApp groups for each of the subcounties, which consisted of the county health management team, sub-county health management team, healthcare workers and the project technical teams. We moved then to the rollout phase. In the rollout phase, we procured smartphones, SIM cards and data bundles to be used by the healthcare workers. We trained 29 healthcare workers from the selected 29 health facilities during the orientation stage on how to use the DHS-2 data capture app for weekly data submission. And then we provided each facility with a smartphone that they could use to submit data. So from there, we went into a dashboard generation and dissemination phase. So here we designed Excel-based graphs and pivot tables for implementation progress reviews. And we disseminated the dashboards to WhatsApp groups and also used them in technical review meetings on a weekly basis. And then during quarterly advisory meetings with Nakoro Department of Health, we also used the dashboards. So what our results are, we have noted a couple of things using this DHS-2 data capture system. What we've noted is that we've been able to receive data on a timely basis for decision making. We've been able to identify service delivery and app tech challenges in a timely manner and provide resolutions. And we've also been able to design other interventions and also to identify intervention implementation gaps to increase impact of interventions to increase maternal and newborn health service app tech. So because of this, we've been able to note significant increases in the number of in the maternal and newborn health service app tech. So if we compare the period when we did not have this implementation, that is October 2018 to January 2019, and the period when we had this implementation, that is October 2019, January 2020, we noticed significant increase in the number of pregnant women completing four-in-seam reasons, which increased from 1,440 to 1,998. Those delivering at our health facility increased from 2,399 to 2,850. And infants receiving maternal care within two to three days of delivery increased from 2,314 to 2,371. So our conclusion is that through the use of DHIs to mobile app, we were able to enhance efficiencies in data connection, which led to more frequent access to data for rapid feedback and assistance. And we are recommending that the Departments of Health account should adopt these strategies so that they can enhance efficiencies in data connection, timely reporting and decision making so that they can improve for the in-seam reasons, delivery, and access to PLC health services. So these are the authors of the abstract and that is Bechizakeo from Mafia Uzazi. Maureen Chebe is a reproductive coordinator at the University of New York. Silla Kimanzi is our USID AOR. Eryko Dipo is from Afia Uzazi Program and Bernard Yauchiz from Afia Uzazi Program. So thank you very much. That is all I had for today and I hand it back over to you, Mahima. Hello, Mahima. I was handed it back. I was not able to unmute myself, sorry. Okay, thank you, Beatrice. This was really interesting and especially because you have managed to work across the continuum from antenatal care through birth care at birth through postnatal care. It was very interesting. Please remember to post your questions. I've posted one question. We can come back to that if we have time at the end of this session. The next presenter will be Brian O'Donnell. He's a senior implementation advisor at the Norwegian Institute of Public Health. He facilitates e-digestries development or tracker data collection at an individual level. And Brian is a colleague. We work together on the development and implementation of metham and child health in Palestine and Bangladesh as well. He also works with the University of Oslo to configure standardized DHIs to metadata packages that follow the WHO guidelines. What do you, Brian? Thanks a lot, Mahima. It's a real pleasure to be here and join us for a fantastic discussion today. So today, I'm going to talk about how we have analyzed data from DHIs to for RMNCH. But going about it in a slightly different way, I will share how we can use some passively generated data from DHIs to trackers to analyze some patterns of client movement within the maternal care system in Bangladesh. And then also explore some possibilities for how we can better communicate and visualize these data most effectively. So to start, I'm going to give some background on the context of this case study, and then share some methods that we've used for data extraction analysis. Then we'll get into some preliminary results of this analysis. And I hope to encourage a discussion about how we can better leverage data on the location of different events within DHIs to to share different analyses for other health areas in context. So to start with this work is a part of a very large, one small part of the very large RCT in Bangladesh on the effectiveness of tracker as a pregnancy registration system to support clinical quality improvement interventions. So there are a number of interesting features of this system, those of you who have been to conference before may have heard of your registries model. And of course, we don't have time to get into all of them, including feedback dashboards to care providers and SMS to clients. But what's really important is, for our discussion today is two key features of the system, which is first here that it's a shared register across the community and facility based cadres of health workers. The Bangladesh MCH system is really complex. And there are four different cadres in two different directorates under the MOH that provides services to pregnant women. At the community level FWA visit women of childbearing age in the community to offer family planning and counseling, while health assistance have a satellite post health post to provide immunization. But all clients can access other services that community centers and family welfare centers. And in the registry client records could be shared by all maternal care cadres in the public health system. And we hope that this really improves efficiency and quality of care. Secondly, this this project included a system to use biometric identification of clients was called the e Palm app here you can see a representation of it. It was developed by element Inc. And it greatly improved identification workflows and really minimized duplication within this registry system. I think within the first guess of taking a picture of a palm it was about 85% matching and then over three times it was 99% in our in our tests of this tool. So the system is really designed to approach the goal of many tracker systems, which is one record per client. And so because it's capturing all the different public health services that are provided to pregnant women, and there's very little duplication at the community and facility levels, this really allows us to understand more about how a individual pregnant woman moves through the public health care system in this one context. But we run into a bit of a problem here, which is that there are no metrics within the HIV to do that multidimensional analysis. So we need to get data out of the HIV to I can share a bit with you about how we perform that analysis. But here we have some of our are shortly our priors about how we would expect patients to move through the system. Probably the family welfare assistance would register the pregnancies first and then send them to the facilities. And then there's limited movement between the other partners. So we extracted these data from a sequel view on ANC events by the different TI included some information on the stage, the event organization, the gestational age of that event, as well as the enrollment organization as well. And then you know, extract some other data to the API just to get a better sense of contextual analysis. And then we loaded all these data into R to process them and come up with different ways that we might be able to explore them. So in a very basic representation would be a histogram of event counts by the organization type. And you can see that most of the pregnancy identification at the unit level took place around 10 to 12 weeks gestation, while others peaked around 20 to 23 weeks. So there's a link to the code here as well to where we explore a bit more of the different visuals. And at the end of this, I want to talk a little bit more from my CHI through analytics perspective about how we want to represent these these movements data through different visualization channels. So getting into the literature of visualization best practice for that type of data. So first off, the first question we might ask is, which types of org units actually enroll clients? And then where are they followed up? So in this faceted grid of different charts, on the top axis, you can see the event organization types. And on the right, you can see where they're enrolled. So really here is where a lot of the action happens. It's the event happens at the unit and the enrollment happens at the org unit. And then here we can see the different colors to represent if this is a first event or a follow up event. And you see that there's really not a lot of crossover between the different org units, this green section here. In fact, just 8% of all the events after enrollment occurred at a different organization unit than where they were initially enrolled. So while we had developed a system that allowed for sharing of this health data across different clinics, we actually didn't see that that was the case in the vast majority of cases. So if we actually zero in on these 8% of events that occurred at different org units, we can see that, sorry, someone's, there's a lot of background, someone can mute, please. Thanks. So if we actually zero in on these, on these org units and events where there was a, that were at a different location than the enrollment, we can see that the majority of these are actually moving from the enrollment at a FWA unit, and then migrating to a family welfare center. And so if we merge the enrollment and event data outside of DHIs to that's the type of insight that might be available. One thing that I would like to explore a bit more would be about using time as a variable with different animations as well. So in the, so one way that we could do this is be a cascade animation, where each dot represents a unique client. And you can actually see the flow of data flow of patients through the health system. So here we can actually see that actually get a sense that this is a health system with flows of clients between the different levels. And maybe that there are other ways that we can do this and start talking about how we can represent these data. But it's not just for getting a sense of the overall health system, but we could also do some basic outlier analysis as well, and try and find certain cases where individuals might be bouncing between different org units. And maybe there are good reasons for doing that that we can look into a bit more. So if we take some of these pathway generated outside of DHH2, that's one thing that we might be able to do. So I think I have another slide here that seems to be skipping over. But it's okay, I can move directly into the final slides. So for the insights for the MATLAB context, first, I would say that we've seen that most of these clients actually stay within the same organization unit at community or client levels, but few of them get past their third visit. And so there are a variety of reasons that we can think about why that might be the case. So is that because there's a leakage to private facilities? Are they simply discontinuing their service? And so that's something worth investigating that these types of analyses can offer. For those that do crossover, most of the events actually happen right around when we would want them to see a clinician, which is at 18 to 22 weeks of age. So that's a positive thing. But maybe we can dig a bit deeper into the different gain and loss or units that we see here and understand why that's happening. Is that related to clinic quality? Maybe some clinics aren't performing that well? Or is it just that there's a strong referral system linking different clinics that we're unaware of? And finally, this is a this is a step that I would like to discuss a bit more with the DHH2 developers there's time. But after we get some feedback from the MATLAB team and see like which ones they really prefer, and we classify them based on like best best practice principles of visual encoding. What what would be the next step for how we would want to visualize these crossover events in in DHH2? Because as a tracker really takes off and is the common register for many different health systems, we really need to start looking at how we can best use these crossover event data to understand how clients move through the health system. So maybe there's something worth exploring here that we can take into future discussions. I think I had another slide after this more slide as well. And we can talk a bit more maybe after break or in question to go into other interactive visualizations that we've also done as well. But that's all I had for now. Thank you, Ryan, really interesting visualizations there. I have a question for you as well, which have posted in the community of practice. But before that, I think we move on to our third and final presenter for the session. The last presenter is Praveen Katka from Nepal. He is working as a monitoring and evaluation officer in a nonprofit organization called Family Planning Association of Nepal. Today we'll present his work on the integration of a clinic management information system with DHIS too. Please. Praveen, you have to unmute yourself. Okay, I hope you hear me here, right? Yes, we can hear you. Okay, thank you so much, Mahima. I also like to thank to the organizers for giving this platform to share our experiences. And thanks to all my previous presenters, they have really comprehensive presentation. Though I don't have that much comprehensive presentation, but it's more about like our experiences, what you have done by integration between the DHIS and CMS. So my presentation is more on my experience theory of our organizations. Yeah, it's like integration between the CMS. We call it clinic management information system. And with DHIS too, that has really improved in data management in family planning issues in Nepal. So first, I would like to talk about the diagram of my organization. A plan is like established in 1950 and it is one of the oldest organization in Nepal. So it is working for the reproductive health rights, sexual reproductive health and rights. So later on, it gets like a memorization of IPPF, which is called International Planned Parentheses in 1969. So we cover around like 34 districts out of 77 districts in Nepal. So it's a big organization working in Nepal. And we have 22 family health clinics and 56 community clinics, and 151 mobile teams with 68 associated clinics. We also have a network of 349 APRF health volunteers who distribute all those family planning contraceptives in the community. And we have like 255 active groups working to aware all the sessions among the youth people. Yeah, so this is how we integrate because we've been using the two systems simultaneously. That's like in the left-hand side, you can see the open AMR, we call it CMS Clinical Management Information System, and that has been integrated to the DEATIS2, which is local DEATIS2. So you can see the screen and the left-hand side of your screen. So before that, you know, I just want to give us some backgrounds, you know, about like the journey of EPAN, you know, before they know like a EPAN, you know, working before having this kind of systems, until 2007, we only had a fully paper-based system. And believe me, that makes our job, especially job for our search provider very tedious, and it's very hard for them to generate the reports to compile all those data and generate the reports. In 2008, we introduced manual CMS, which is called Client Card also, and also implemented ECMS in. We started from the 10 Family Health Clinics. Actually, Family Health Clinic series, it means Doctor Bates Clinic, and CC means it's a Star Bates Clinic. And then we scale up to the 26th clinics by 2018. How they till 2016, we were continuously using Excel Sits for reporting of remaining STBs. In 2007, we roll out a local level with the DHH2 in all our STBs. And after a short while of having both the DHH2 and ECMS in clinics, we arrived at the bridges and we understand like taking the system simultaneously parallel. It makes our life very difficult for the search provider. So we decided to integrate it and to make our jobs more effective and easy and make our data more data-driven decision-making process. So then actually, for the people who really are not aware with DHH2 as a skip away, we know like silent features of DHH2 and CMS. Actually, DHH2 is more automated application tools for collection, validation, analysis, and presentation of aggregate data. And it's like open source flexible data warehouse which serves as a repository to take data from multiple sources. That is the silent feature of DHH2. Whereas the CMS is more, it's a client focused clinical managed system. So like we entered the data of each and every client in the system. It's like a client flow and it generates the report also. But it's totally, it's a client-based system. It's called CDR. So these things, because we are like running simultaneously, it makes a little bit difficult for us to get to gender reports, you know, at once. So that's why we thought like, why we integrate the system, these two systems, because it has like, because there are lots of duplication of reports because people, our search provider has to fill out many registers, you know, like, and duplication of reports is going on. And there is always a chance of constancy in service data. And there's a lot of chances of data use. And it has also made like reporting delay in reporting. And then there's no option for collision of reports of different communities and our centralized data. And it's a very difficult for reporting, especially we need to report to the government administration, they have their specific, you know, this tailor format. So we need to fill up that makes quite tedious for us to generate reports, acting to be, you know, like a standard format given by the government. So then we start like kind of bridging the system. So actually what we did, you know, like we use the, you know, like ECM as we export the file, because you can see there are like three ECM systems going on in different, you know, STPs, self-settling points. So from there, we export the data. It's a client-based, you know, like it's a total individual client's data, it's been exported to the, you know, CSV file. And then that has been imported to the DHS, local DHS too. This is at one level in the local level DHS too. And then there's another level like a kind of integration between the local DHS to the global DHS, because we have to report to our, you know, like region office and to our central office, which is in London. So you can see here the screens like of two different countries, like one is from like our country, Nepal and one is from like India. So generally, the data we enter in our local DHS too has been exported and that has been imported to the global DHS. So generally, we'll be doing this for the like a yearly, all those data has been exported to the global DHS too in the early basis. But in our context, in the context of Nepal, we've been doing, we've been generating reports in a monthly basis. So though the DHS has had the absence of doing it weekly basis also, but as per our requirements, we've been doing in a monthly basis. And the main thing, you know, like the benefits, the results, what we get by integration of the system, these two systems CMS and DHS too, it has makes us provider or more time to provide more services to the clients because they're most of like the 50% you know, like a pressures of like entering all those, you know, register manual register has been reduced so that their spare time they can provide the service to the clients. It's very good for the organization. It has also reduced the cost because it's like a one time investment. And then after that, it's like ongoing process. So it has also reduced the cost also. And then it's a single source information, everything, every you know, like reports can generate from the DHS too. So it's a very single source of information can generate it from it. So which is also very benefited for the organization like like Nepal and so that we can respond it as per our requirement to the donor to our you know, like regions to our central office also. And then it's a significant it has significantly improved the data management and analysis and since like the SS to have like, there's lots of options of like a data visualizations and then like a infographic systems is there. So it makes very easy for us for the data management for the data analysis thing. And it also makes our office and very efficient. And no doubt, you know, like, because that's always improved and enhance the data quality also because it has we have to go through different data validation checks also. So once we go through it, then data quality that it has increased and enhanced the data quality too. And and and and as I already mentioned you that like it had ease and speed the coalition and reporting because otherwise, you know, like a search provider or special admin is they have a hard time like they're getting all those report compiling and reporting today, you know, like a constant department, so it takes like a two, three days for them to just only collect it or just to gather and analyze the, you know, all this data. But now with less than five minutes, they can do the job very easily and then generate the reports. And it is very transparent also. And it's very efficient also. And that has also make them to have a like a very effective like data driven decisions in their programs also, which is very good for them, you know, to, to, to, to be accountable to the donors and to the, you know, like to the, to the government also. So it has totally, you know, minimize and reduce the, you know, like duplication of reports, you know, at all levels, you know, from the region levels to the, you know, like central levels, as that is one of the benefits we have getting from the, you know, like, because of the integration of the system. And it has also added in data entry at the time of coalition, coalition. So that's like, that I just saw how the system, the two system when it's integrated, how it has improved our, you know, like the if sense in working style, because it has increased the speed, the speed of like a service providers, like a giving the services, it has reduced the time, it has increased the quality because the client, the service providers can give more time to their service to the clients. And it has also reduced the waiting time for the clients. So it has enhanced the quality of care in the different service delivery points. And it has reduced the cost also, because I already mentioned you, it's like a one time, you know, investments. And it has reduced all those like printing costs and all those, you know, all those extra costs after integration with the system. And then here come like, and then I've also gathered some of the quotes from the, you know, in users, those who are using the system like integrated system. So as from the service provider, you know, like real points, like they said that before the integration, they had many ratios and it was difficult and took time. Now, we have more time to serve clients, which is the good things for them. And from the branch managers, you know, like from the managers level, they said that it has become easier to prepare and generate monthly reports in a short period of time. That is the quotes I've gathered from, from the in users. And that's, though, it has also set challenges while integrating the system, it's not easy, you know, because there's like a DHS and there's it has its own system. And there's like a CMS, it has a system, the service courses. It's quite different. I mean, like quotes are quite different. So we need to map according to the DHS too. So what we did like we all the services in the CMS has been mapped according to the local DHS too. And during the initial time, there's also a lack of appropriate skill, because like first time we're been doing this kind of integration. So we have lots of, you know, challenges at the initial time. And then there's always a question of training and interaction. This system is new for our service providers. And we are at the initial time, they're not so much friendly with using the computers and you know, like getting friendly with the system also. So in the initial time, it's quite challenging for us, but they overcome all the challenges. And generally, you know, like, because the organization like, like us, like it's, we are like a nonprofit organization. So our budget is we don't have like a soft sin budge is to invest in the infrastructure like computers and, and others, you know, like a logistics support. So during that time, there was also challenges to, to, to, to manage to arrange all those kind of things because they lack of financial leadership. And our next step, you know, so because as I already mentioned, from my you know, early presenters, there are lots of like options in the modules in the DHS too. So where, where you are planning, you know, like, because we still have these like CVDs, like our volunteers are still collecting the data in the company label. So those, they are doing it manually. So what do we have tries like providing the, you know, like mobile to them, and we can augment, we can mitigate that, you know, like the data to the, you know, like a DHS to throw the mobile. So that's like we're planning to do that thing. And we are trying to scale up more comprehensive service, service delivery data management system itself in the DHS to still there are some of the, you know, like services, we are not able to tell us especially advocacy thing and other, you know, like comprehensive sexuality education thing that has to be still need to be, you know, like a map. So we're, we're doing and that, making more comprehensive. And also, you know, like, by the, by the, by the enough, you know, like, we want to, we want to more enhance the data driven decision making cultures that need to be, you know, like a practice in our organizations. So with the help of DHS to, we are looking forward to improving that in that part. So this is the system what we are using, like transformation, like they're using in the, you know, like, left-hand side, it's like an open MR, we call it CMI. And in the, you know, right-hand side, those all those data has been exported or imported to the, you know, the DHS to. So this is just one figure, the pictures that I want to show that, you know, before, you know, like using all those manuals, you know, it's more like a service-based recurrence system. So we have a hard time to start out to retrieve all those data as the clients, you know, so it's, it's real strong to do that thing. But later now, we've been using like, it's like client-based recurrence system. So it makes life more easier for our service provider and to the managers. Yeah, that's, you know, that's like a experience. That's a, that's the experience that what you have, you know, like, educating the systems. Thank you, Maima. Yeah, I give it to Hanover to the Maima. I think, yeah, Maima, some muted now. It's just some difficulty on muting. Thank you. This last picture in the last slide, I think it speaks to the magnitude of difference between the system you have before and what you have now. So I think we have about 10 minutes considering we have to end the session five minutes before four. So please feel free to post your questions on the community of practice. I have a few questions and I think we can start with them now. So let's just take it to the same order of the presentations today. So I have a question to Beatrice. So you had mentioned about making microplans for each facility. So I was wondering if you could tell us about some of the specific service delivery challenges that health facilities had. Okay, thank you very much, Maima. Yeah, so some of the health facility challenges that were faced at the health facilities were in some health facilities, they didn't have the lab laboratory equipment. So they had to take the lab tests to other facilities. And they are finding some challenges like transporting the lab results from one facility to another. So that sometimes caused delays in getting lab results. So because of that, some of the mothers preferred to go to other facilities which had the full lab equipment. And that and sometimes these facilities were not very close to where they are. And that caused some challenges in uptake. So that is just one example of some of the challenges faced. Others had some other challenges like maybe they might not be having like the full maternity equipment and things like that. And sometimes we provided some of the facilities with some maternity equipment. And so where we noted that the one facility had more equipment than others, we were rotating the equipment so that the women in other areas within the region can also benefit. So we were looking into those kind of challenges per facility and developed coming up with a plan so that you can be able to to resolve some of the issues that were faced. Interesting. Thank you. We're waiting on questions in the community of practice, but I had one for Brian. I think you had really nice visuals. And I was just wondering about, if you could say something about the implementation considerations, how should you implement a system if you were to do these kinds of analysis with the tracker data? Yeah, thanks, that's a really, really great question. Fortunately, it's quite difficult in many contexts because of the first, the upstream technical challenges of making sure that all of the clinicians have access to the clients who may come to their health, to their health service post, right? So if someone is enrolled at a very far away in a district, different districts, for example, should that clinician have the right to access that health record is a balance between privacy of the record and completeness of the record that you have to strike. Secondly, identification as well is still quite challenging in many contexts. If you don't have unique identifiers and national levels and you have to fall back on incomplete or non-ideal attributes such as full name and birth date, which can be challenging. And finally, also having encouragement and incentives for creating a record for, sorry, for building on an existing record rather than generating a brand new client. I think for the idea of searching for a client that exists in a system is still a barrier for many people, they don't quite see the benefits of it. And so we have to design systems that encourage people to do a search, then open the record and then add on rather than skipping that initial search step. So I think there's a combination of those three things. Thank you. I was just typing the question to the last speaker, but I can read it out as I type. So, Robin, I think I wasn't completely sure if you had mentioned what kinds of data you are including in your system. I think you had a brief presentation about it, but then I was wondering if you could explain a little bit more about the type of data you collect or the extent of data, what kind of target groups that are. Yeah. Okay, thank you. You know, like actually, you know, like since like we are providing the services and reproductive health, you know, all those family planning and all those, all those like, like services, you know, related to the sexual health, you know, so, so we are, you know, like getting all the services, like we are entering all those services, you know, so all those integrated packets, essential services, you know, all those family planning, you know, abortions, like everything, you know, like related to the sexual reproductive health system. So we make, you know, like clients, we make like records of all the systems and all those services in the, you know, like CMS, and that all those services are integrated, I mean, uploaded to the DHS too. So, so it's like generally all kind of services related to the sexual reproductive health has been, you know, like a collected follow up question to that would be then do you have is it set up in such a way that it follows the client? So there is one client and then all the services are in that together? Yeah, yeah, it's like client based recording, you know, so like doing in the, you know, like CMS, we capture the client based, you know, the services, you know, and with like, in while uploading it to the, you know, DHS is more like aggregated, you know, like it's clients are not entering that DHS to only services of those CMS is uploaded in the DHS too. Okay. And the last question, I think you'd mentioned the resource constraints in terms of finances is one of the important limitations. But if you had to name a technical limitation that you thought was the most important or most challenging, what would that be? Yeah, for that, you know, like we're lucky enough, like our, you know, we have the memorized system of IP international plan for a good federation. So technically, we've been like a supporting, we get support from day, you know, like we have the experts like working in the regional office and working in the central office, we've been getting all the supports from them. But then also, it meets like other financial sources for, you know, like getting all those, you know, logistic supports, you know, but so far, we are getting the, you know, like supports from them, we have like special projects in like the projects like from anonymous donor help invest a lot for, you know, like for setting up all this kind of, you know, like a systems, you know, but it's still technical, suppose in the local, you know, like in the local context, we have lack of those kind of technical sound people in the local context, you know, so because we've been working in the lots of in rural areas, you know, in different parts of the country, it's very mountainous countries, because in Nepalese, so we don't have that much technical people are like available in that context. So that is always a very challenging for us when it gets crass or when it's like a problem. So, you know, to, you know, like fix it up, you know, so it's, it's, they have to rely upon the, you know, like in the capital cities, like where we are located. So in that way, you know, there's lots of challenges to kind of overcome all those, you know, problems.