 This session, it's on data triangulation. I am not of course here to teach you what is that a triangulation I'm assuming that everyone knows and everyone is interested on like on how to use the system to actually improve their analysis. So, in general, just as data triangulation here, we are just assuming that like we need to use multiple sources to have a more comprehensive and only steep perspective to generate information for our actions. In general, of course, we know that there are quite a lot of added value. It is a complex process that requires a lot of work and requires the understatement that in theory you should have all your information at hand, hopefully all in one national repository that allows you to actually triangulate this information. Of course, by triangulating you are trying to also reduce biases and errors so your data become more accurate. You're trying to also having multiple sources to avoid any kind of distortions that could actually raise by just using just one single source of data. You are actually might might have as our results and like increase your effectiveness because having your a deeper understanding and a deeper amount also having a large amount of data you might want to have like a more nuanced analysis of your information. Another added value, of course, is the engagement of the different stakeholders because in a lot of cases in a lot of countries, different stakeholders are also collecting part of the data that will then feed into your national system and also increases transparency of your actions. But most importantly, the added value probably triangulation is definitely building on all these reliability validity all your extra insights to support your decision making process and to actually do programmatic decisions for all the different programs that you are supporting. Of course in health programs, there could be different kind of triangulation that you want that you might want to do like you can do like I don't know program outcomes across different periods so you might want to triangulate okay what happened this year and last year together with other type of information. You might want to do, for example, other different age groups and subgroups and then you might want to do, I don't know, like written here different program outcomes, among age groups or also very important comparing data from different type of surveys and maybe triangulating this data also on top of your routine information that you're getting like let's say for example coverage, let's say, like I don't know any kind of results of activities that you might want to do for example either in facilities or communities. And of course, like you can also check start checking your triangulation about your outcome for example from your different health programs in different geographic areas, therefore you might start to check also a little bit of a pattern if there is any kind of geographic patterns for for outcomes. I mean the type of triangulation that you can do are immense and almost infinite, let's say, and of course we know quite well the information that you can triangulate about surveillance about your stock and about your EPI so any kind of like immunization data. There is also other other examples that and some of them today will be covered also from our presenters so like for example, entomological and that back to control data together with your, with your surveillance data, you might want to triangulate your routine. And I mean in strategy data together with your survey data. Very important also for validating your population data in some cases you might want to triangulate your information that is coming from estimates together with the information that you might want to collect from the community, and also from the the pharmacist that you're having probably some countries doing more often so countries doing less often that's even more important in this case. And of course it can also be also very important actually triangulations between the information that you're getting from your HMIS and the information that you're getting from your CHIS, therefore maybe triangulating the kind of information that you're getting from community and the triangulation and the information that you're getting from your health facilities that's also very important when when we're talking about surveillance and of course API. So, as you can see these are just like a couple of examples out there but the examples are infinite and it's very dependent on the availability of data that you have in in your national repository. Again, for example, here I mean these are things that we get on a daily basis also from from our use cases out there. Here we are looking at BCG coverage, like here in the in the little dot we're having like the information that you collect from from data and here in the in the actual in the actual area we are having the the information that have been extrapolated from the facilities. You can see that in some cases it matches in some cases it doesn't, but how will you ever figure it out if you never actually bother to triangulate the different sources that you are presented with. Same story with the API and IDSR. Here you have for example your meso cases that are confirmed and here you have your MR coverage. In some cases, the cases are very high could be due to the fact that the coverage is very low but it can also be due to the fact that for example, some some coverages although very high might be targeting like too late to the children and therefore the children will be exposed to the disease that for which they should in theory be protected instead. So there are quite a lot of things that you can be done and a lot of interpretations that are normally ignored unless you actually start digging a little bit further. I don't want to take more time because today we have four presentations that are very interesting and show us different use cases. We have the two first presentations that talk about data triangulations also with immunization data. One is more to try to find zero doses children and it's a use case coming from Rwanda. Then we have Vincent who is going to talk to us about trying to find in Kenya missed opportunities which is an incredibly important activity to be done. And then we have two other use cases that are more about surveillance of and vector control. So one, it's about like also the way that they have done a micro stratification process to support their outreach activities. And then another one on the strengthening of a vector surveillance based also on their vector control in in in Pakistan. So actually without further ado, if, if, if Samuel unfortunately is not Dr Adeline who is going to present he was supposed to be in S but unfortunately she's also unavailable to present so we have Samuel Juan Garnira, I hope I haven't but church your surname, who is the coordinator for the DHS to data triangulation and for afternet. And yeah, if, if they are online and they are co host, I will stop sharing my screen. I'm getting there. And then I have to put the different speaker. Is it these one the right speakers. Yes. And, and Samuel, if you're online, can you please try to share your screen. Hi, Victoria. Hi everyone. Good, we can hear you very well. Can you try to share your screen. Absolutely. Let me share. Do I have to share the presentation or you will upload it. You can, you can share your presentation from Iran, if you want, if you prefer. Thank you. I'm trying to find the sharing button here. It's at the bottom of your zoom screen and it's a little square with a with a upward arrow. Thank you. You managing. Oh, there you are. We see your folder access denied. Oh, sorry. It's more for you. I'm sorry for you. Can you see my screen. I still see unique permission to perform this action. You might have shared a wrong window. Oh, I mean, I'm saying it for you. I might have seen behind these something interesting. Yeah, if it's difficult for you, we can share the presentation for you and that's not a problem. Sorry about that. No, no worries. Just stop sharing and try sharing again the right, the right window that has your present PowerPoint. Okay. Right. Usually there should be multiple windows that you can try to present. There you go. Perfect. Perfect. We see it. The floor is yours. All right. Thank you so much. Good afternoon everyone. I'm very happy to be here and to share our work on. The DHS to data translation for immunization and vaccine preventable diseases. We will share with you our implementation roadmap and some of the findings. So my name is a woman somewhere. I work with affinities as the project coordinator for the DHS to Randa. So this is our outline. So to start the, I want to say that the high quality public health data is key to timely and informed the decision making. And in Randa, we have electronic health information systems and robot data sets, especially within the immunization and surveillance programs under the DHS to system. So there are challenges that are associated with the use of these systems for programmatic decision making. And they are for new methods are needed to integrate the systems and find ways to critically analyze and summarize this data. That's why we designed the DHS to data triangulation dashboard with the aim to synthesize and automate vaccine preventable diseases and routine immunization indicators. And to identify data quality issues and program performance gaps that can guide problematic decisions. And to illustrate a bit further, I would say that the systems that are being used in Randa. Excuse me are the integrated diseases of balance and response, the expanded immunization registry, which comprises aggregate and individual data. There's the civil registration and vital statistics. There is also the vaccine logistics information management system. And you can imagine that with all the 30 districts and around the fire 56 hospitals, 500 health centers that are using these systems to report indicators at various frequencies, whether daily weekly monthly or quarterly. It creates a huge amount of data that is not easy for an individual to really summarize and and provide good insights. So, to address these challenges, therefore, the concept of what the translation was seen as a solution. And of course, this approach will provide more time and opportunities for discussions and actions. And that can be taken to allow deeper exploration for more complex questions. So we use the global data triangulation guidance to explore ways we can compare multiple sources to identify immunization and program performance gaps. And for instance, I can say that the vaccination coverage may appear very high, and yet the high number of cases can still occur in a particular geographical area, which may raise questions as to whether the reported coverage is accurate. So therefore the data triangulation is seen here as a way to identify these gaps in vaccination coverage. So we to start we identified we developed a roadmap of activities and timeline for this for the dashboard development and implementation. So the first activity we developed is the data mapping, which essentially was able to identify or map data elements, map also systems, first of all, also metadata among the systems. So the next activity that we we identified is the integration module is to develop an integration module that will essentially help these multiple systems to share and communicate with each other to share data so that dashboard can be developed. So the other activity that we we conducted that would say is the customization and visualization of indicators. So we customize indicators in the systems and integrated programmatically informative dashboard visualization for national and sub national levels. So this is a picture of the, or an architecture of the integration module, you can see in the middle. The whole essence of the idea is to make sure that the integration integration module helps these multiple systems to communicate with each other says dashboard can be that data can be accessed and the dashboard can be also created. Let me invite you to the results section where we, this is, this is a list of indicators that we developed for AFP measles neonatal tetanus and HPV. So these indicators for immunity gaps can be grouped in, I would say three categories, but there are multiple indicators that were developed but it's a huge list but we need to also kind of prioritize some which are useful for program monitoring and decision making. So the priority ones that are indicators that help us assess the vaccination status or coverage among VPD cases and we try to do this by edge group or time or by map. So the other types of indicators that we do is to is our dropout rates for MR1 to MR2, Poly1 to Poly3 and DPT to DPT3 and others are zero dose and under immunized rates. So for program performance, we tried to, we have indicators that help us assess the access and utilization of immunization services. Here we try to compare the coverage and dropout rates. We also have indicators that help us identify the quality or discrepancy issues in surveillance and immunization where we, for instance, we try to compare like measles cases in both aggregate and case-based surveillance systems. So the other kind of indicators we have is the surveillance performance system, for instance, the sensitivity of measles detection, the representativeness of measles by geographical area, the timeliness of lab results and also indicators that are associated to the lab. For instance, the lab distilled sample adequacy and we also have some other indicators like the detection of measles to identify the source of transmission, whether it's endemic, important or both. And tried also to distribute outbreaks of measles by geographical area and also assess the completeness of investigation of suspected cases. So for purposes of time, I won't go through all the visualizations that we developed, but I will try to illustrate a few. For instance, here we see on the left hand side an indicator for measles confirmed cases by age group and in multiple timelines by year. And this indicator tried to replicate that in our system. However, we still also have to make some improvement. I would say that the dashboard is now has now been developed and is under review. Feedbacks are being provided to our technical is being provided by our technical team so that our technicians can finalize this dashboard. The other type of indicators you can see on the left hand side and the global guidance we have. We try to compare zero doses and DPT coverage. So by year and by region. On the right hand side it's in our case what we try to replicate. But of course it has also some things to improve on the visualizations. On the program performance. You can see on the left hand side, the global guidance. For instance, they show that you can compare measles cases suspected cases in ideas are versus the CBS system, and we also try to replicate that in our system. The same applies to the dropout rates on the left hand side, the global guidance, what the global guidance provides, and we try to replicate that in our system. One thing I can add is that we've been using testing data, but soon we will migrate to use actual data. And we hope to have some really good insights in developed indicators. So we have some of course challenges, but which I would say not which are minor. The first one is related to data accuracy of using the, the civil registration we want to use the civil registration and better statistics as denominators for births. But the accuracy of data hasn't reached a good level. So we are still using the census data, which provides a projections. But we hope that in the future we will be able to use the CRBS data. The same applies to the VLMIS, whereby the system has been developed and is being used, but we need to monitor the use and data quality so that indicators that were created can really be deployed and be made available to the program users. But right now we can't say that we are going to use the VLMIS since it has to improve its accuracy. So in conclusion, we would say that being able to finalize the dashboard, we've been able to develop the dashboard and being able to integrate these systems, it's a really good milestone. We hope that once we have finalized well the dashboard, it's going to make a lot of improvement in terms of monitoring and decision making. So we are planning to have a rollout training for the national level and district level, actually next month. And we, I would say that one more thing that we have been able to achieve is being able to pull together stakeholders to really discuss these kind of issues is also a huge milestone. For instance, the server lens and immunization program, the HISP and all stakeholders, CDC, WTO, it's a really huge milestone and it's being appreciated. So also the fact that we will be able to identify these issues in areas, it will help us to provide supportive supervision. And obviously in the future, we hope that it will have an impact on the incidence of BPD and obviously deaths and as well as in efficiency and management of vaccine and services which are provided. Of course in the coverage and yeah, it has multiple, multiple benefits. I wouldn't be able to illustrate all of that, those right here. So as next steps, we are planning to provide or to conduct the first national rollout training and also the district for the national level staff and also for the district level staff. After that we will be, we conduct, we'll monitor the dashboard use and also document lessons and obviously sustain this kind of collaboration and coordination to make sure that we stay together and can work together towards the impact. Thank you very much. This is the list. This is a picture of us in a workshop and these are the few among all the people I can acknowledge. Thank you very much. Thank you very much Samuel. And, and I witnessed it myself last time that I was in Kigali like the amount of work that has been put in these, it's humongous and hopefully you'll be able to bring in also like the BCRBS data soon. It's a, it's a big endeavor that you have started. Thank you again. You can stop sharing your screen because next we will have unfortunately we weren't very lucky with these presenters. Everyone had quite a lot of problems with their visa we only have one person in, in, in the room but the next one will, will be able to present online as well. So we have Vincent Omondi, who is the program officer for vaccine and immunization program at CHI. So, yeah, if, if, if you want to, to actually share your screen and unmute yourself, that would be great. Thank you very much, Victoria, and hi everyone. I'm happy to be here. So let me just see my screen. Can you see my screen? Perfect. Thank you. Thank you very much. Hi everyone. I'm happy to be here to present to you our finding in Kenya on how we live with data is to strengthen health systems and improve HIV vaccination in Kenya. This was through mapping missed opportunities and tracking them and finding the best strategies to identify and vaccinating on site. My name is Vincent Omondi as you've heard from Clinton have access. So I'll just give a little of a background statement and why we're doing this. So in Kenya, in 2019, the government successfully rolled out HP vaccination into the routine immunization schedule that was targeting 10 to 14 year old girls. And since then, a lot of strategies have been employed to accelerate and sustain HPV vaccination. Health facility based vaccination has been the mainstay approach supported by school based vaccination approaches. However, school based vaccination has dominated despite despite financial strain at the sub national level. Our target population for HP vaccination here in Kenya is approximately 2.1 million girls. And as at the end of 2022, our current vaccination rate was at 63% with a completion rate of 32%. There have been a significant sub national variation in these performances. During the introduction, the government had targeted to achieve at least 70% vaccination through facility based vaccination alone. But this only achieved 13% coverage. Part of this high coverage was through school outreaches and only 13% vaccination was attained during activities that did not include school outreaches. We continue to experience financial strains, especially when it comes to school outreaches because a lot of resources required to carry health care workers to vaccinate the girls. Again, part of the challenges that we've experienced with low HPV vaccination in Kenya is the fact that there's a lot of financial constraints in terms of doing those outreaches. And also the fact that the very low health facility visits among the target population. So there has been a notion that girls aged into 14 year old do not visit health facilities and so health care workers tend to conduct school outreaches to find girls in schools. So we've experienced very low health facility uptake. In this work, a data driven approach was now employed, do some data triangulation and was implemented to look at the frequency of health facility visits among the target population to optimize HPV vaccination during these visits. So we looked at did some sort of the analysis and looked at what is the frequency of girls that aged into 14, how frequently they visit health facilities and how can we optimize these visits to optimize HPV vaccination at the health facility. So objective and methodology for this triangulation were clear. So we, our main objective was just to leverage data as to platform to examine the frequency of health facility visits among girls aged into 14 essential services and evaluate the potential to enhance HPV vaccination during this visit. We employed a few methodologies to meet this objective. Number one, we utilize our processing data from the data to platform. The main data set that data elements that we use were the MOH-710, which contain the immunization data, the MOH-515, which contains the workload data, and also the population from Ken Vies that enables get the target population of 10, 14 year old girls. We then combined the triangulation outputs and selected part of this in a few health facilities. So comparing these visits, comparing these data findings to the total number of HPV vaccination administered, we were able to quantify the extent of missed opportunities for HPV vaccination here in Kenya. So our data triangulation pipeline followed some few steps. The first was to gather the elements from DHS-2. This means we put together the data sets that is MOH-710, for immunization workload, and also MOH-204 for outpatient visit data, and DHS-2 data sets that is really data triangulated already using Ken Vies data from 2019. Our triangulation approaches was leveraging, looking at approximating total visits during the entire year. Our target year was 2021, and then adjusting these visits per person per visit. Remember, if you look at in the next slide for the results, you will see that a lot of visits was experienced. But of course, this visit could be one person can have more than one visit, and so we tried just so that we can be able to get approximate per person per visit during that period. And then we conducted now descriptive analysis where we explored the total visits nationally and also looked at them sub-nationally, and then finally quantifying the missed opportunities and then used these to make some data driven decision. We looked at the results, the graph, the first graph represents the total population that we are targeting in Kenya, 3.1 million. So we expect that we should vaccinate at least 90% of this population. So in the year 2021, looking at the outpatient visits from DHS-2, total of 4.7 million girls visited health facility. Those were registered visits. Of course, I've said in initial statements that one person can visit health facility even more than once, or one person can visit health facilities, more than one health facilities. Those will be registered as visiting the first, will be registered as first visit. If you look at 4.7 million visits, and in the country here we have at least 13,000 health facilities. It means that we are having at least 30 visits per month. If we try and let this and adjust this per person per visit, we are having at least 1.1 million visits of girls aged 10 to 14. So these were approximate adjusted per person per visit. So it means that in the entire period of 2021, from January to December, we are able to experience 1.1 visitations of girls aged 10 to 14 years. These approximate were at least seven visits per month in all the 13,000 health facilities. If you look at the uptake during this period, we had HP vaccination for the first dose in that year was 879,000. This was uptake that combined both outreaches, school visitations, and health facility outreaches. So if we adjust for 13% which is only health facility visits, we only remain with 114,000 outputs that come from health positive visits only. So if you try to compare the total visits from GHAs that were recorded as visited health facilities in 2021, we had 1.1 million and only 114,000 girls managed to get vaccinated. So that means we had a very big gap of more than 900,000, 1 million girls that had potential to be screened for vaccination but were not vaccinated. So this is how we were able to quantify the millions of girls that had health facilities without any opportunity to screen and vaccinate. So looking at the map, we can see areas that have very high potential girls visiting health facilities, a lot of high workload and a lot of visitations also vary sub-nationally. Looking at the results that we're able to get out of this, the total missed opportunity to screen for eligibility for eligible girls increased six times from 300,000, 1.6 million when we adjust this for health positive visits only. We could see on the previous graph that if we only look at uptake in the health facility, the numbers goes down. So it means that there's a lot of opportunity to screen and vaccinate at the health facility. So out of 1.2 million eligible girls visiting health facility for essential services, at least 696,000 were potentially eligible for vaccination at health facilities but only 19% were vaccinated. So it means we had an opportunity to vaccinate a lot of girls who brought themselves to the health facility but did not get a chance to be screened and be vaccinated at the health facility. So having looked at all this quantification, we came to a total of 581,000 girls who were potentially missed and were eligible for vaccination and were not vaccinated and they attended health facility in various vaccinated health facilities in the country. And this was five times more than the national uptake during that time. So the screening intervention was piloted in 14 public health facilities and resulted in a 49% increase in HP vaccination. After doing the triangulation, we went out there in 14 public health facilities and implemented this and piloted this in facilities and after our post assessment, we got an increase in 49% vaccinated in that health facilities and that is displayed in the graph that you can see below. So if you look at the graph on the right, we have 1.1 million girls attending visiting health facilities, potential girls visiting health facilities and eligible at 696,000 based on the coverage at that point and only 114,000 girls were vaccinated at the health facility. So which means we missed the potential to vaccinate 581,000 girls who brought themselves to the health facility for other routine activities. So these findings highlighted the magnitude of missed opportunities for vaccination and emphasize the need for targeted intervention to optimize vaccination during health facility visits. So by addressing all these missed opportunities and maximizing the potential for school-based outreaches, the aim is to significantly enhance the coverage and effectiveness of HP vaccination program. So the figure on this screen is an example of a screening tool that was introduced after successfully piloting this intervention. And this brought up a vaccination screening tool in form of a vaccination system that the government currently is preparing training manuals, training materials to train all health, all vaccinating health facilities to have a screening tool for vaccination. And this has been scaled even not only for HPV, but also for COVID vaccines and routine immunization vaccine to track default has and even to track zero dose children in the health facilities. So in this way, all children who attend health facilities through the outpatient visit will be able to be screened, be identified, and if they are eligible for any vaccination, they will be referred to the MCH for vaccination. So this has been piloted, has been tested and currently the government is in the process to train and offer capacity reading for the same to be scaled. A few recommendations here and there, by leveraging technology like HIS to the decision-making process in public health can be enhanced, leading to more effective interventions and policies and also regular exploration and utilization of the HIS to data enables informed and insightful decision to improve public health outcome. These findings provided evidence to support the implementation of the screening strategy, which when combined with other approaches can contribute to achieving the WHO's ambitious target of at least eliminating cervical cancer by 2030, ensuring at least 90% of girls are vaccinated by the age of 15. By implementing these strategies, healthcare system can maximize the potential of health facility visits, leveraging data for targeting intervention and improve the effectiveness of HP vaccination in programs in Kenya. Some of the key takeaways include enhanced health facility based screening for outpatient, for optimal HP vaccination opportunities, use data driven targeted approach with digital health solution for vaccination programs like HIS through platforms and also improve monitoring and reporting for HP vaccination data, including missed opportunities. And this brings me to the end of our presentation today and thank you very much to people who supported this work. Very great technology mentioned the means of health, the Clinton health and also the data esteem. Thank you. Thank you very much Vincent is actually super cool you had really nice, really nice numbers and always gets me very excited to see a low p value. The next presenter is actually in person. Yes, that's really nice to see in person people. He was one of the lucky ones who got a visa in time. And yeah, he, our next presenter is Rajab Mkoa, and, and he's a senior system developer at the University of Jerusalem. So, just a second that I share the screen. Let me open first, because I'm a disaster here. It's more complex than you think you are number three. I also share my screen. No, it's not that one. Yes. No, it's not that one. Let me, let me try. No, it's not that one. Okay. It's very complex. Okay, you do it. Presentation. It's really open, like it just need to stop sharing this one. You see it. Okay. Thank you. Okay, is it evening or afternoon. It gets confusing here. So I will say both good evening and afternoon. Okay, I have prepared. So my name is Rajab Mkoa system developer from University of Jerusalem, which is to lab. On behalf of my colleague here presenting one of the work that we have, we did, we did. We call it micro stratification. So just to kick it off. Yes, so a little bit of background is actually this work is around malaria and due to resurgence of malaria, despite many other effort. Now WT recommends moving from the direction of all size fits all to a little bit of more tailored or targeted interventions. And most in certain ways, it is focusing on categorizing the intervention based on the risk aspect. And for example in Tanzania, they have categorized into level of disease risk from very low, low moderate to high. So, in a way, there are different intervention around that category. And at the moment for, for, for the while, that categorization was only focusing on sub national level district and region, also called the macro stratification. So they created some stratification or maps to show the malaria risk around region and district and over years they have been using malaria parasite prevalence data, most specifically school malaria parasite data collected from schools and household and they have been using, or they have been using this data for after every two, two years and all three from the household. So this approach sort of seemed a little bit lacking, because it covered a little bit of small area. They actually sampling hundreds of houses in every survey that they're doing and use the data to actually categorize the malaria risk maps. And of course, now, also if you see the, the collection interval that is in two years, it a little bit not give you that final fine resolution on the, the categorization that you need. So, national malaria control program in Tanzania in recent year sort of came about seeing how best this can be done and with the availability of data, thanks to DHS to platform that is from HMIS. So, NMCP sees like this could be an opportunity to more like use the routine base the data that is collected in monthly basis to more like create potential risky maps around stratification, but all together with the increasing empowerment in the decentralization health sector in the government. So the NMCP also so they needed to more like you go towards the granular level for stratification and enhance forth. Now we, we come into the micro stratification or Wadi stratification. So in our country, the hierarchy level is from the country, region, district, ward, and then villages and facilities. NMCP sort to now goes into what stratification. So now, how this stratification or risk information can can be realized. So as initially I said, they had used the malaria parasite prevalence, they established sort of they can say threshold around what can be very low, low moderate and etc. But again now going towards using the data, the sort of establish the process towards using the HMIS data now to identify which words are to what categories. So, based on the DHS to and you need some DHS to lab in collaboration with NMCP and towards elimination of malaria project worked on more like this certification, or by designing an automated now process that can help creating those different categories. So a little bit of what has been done is sort of to take the data from HMIS, clean them up at least you can have a clean the data, and then more likely create some some scores or strata around, and then create the risky maps that can be used it for micro planning. So major outputs around the stratification where the risky map for planning. This is the one that goes to the government to more likely be used during the planning, which we call micro planning, but the other output is for monitoring. That's what NMCP use regularly to more likely look into how the risks are changing over years so planning is is generated after every three years but for monitoring is generated after every one year. So a little bit of how this process is is actually first they identified different indicators from which the stratification can be done. So during the identification of the indicator, the sets the criteria sets for more like processing the data. One way is to remove the duplicate data missing elements based on reporting and completeness checking the consistency assuring that the data that can be used is a is a little bit clean, etc. And the next for those data that seemed a little bit not consistent. They are sent as a feedback for NMCP to more like look them across. Also, in order for stratification, which I said is what is stratification, the other process for mapping health facility to respective word had to be done and this had to use another system which is health facility registry that has registered for facility and words which HMIS does not have. The next process was the stratification process in terms of what is being used. There are some data that need that are coming from survey and the population from sensors. So there's also a malaria composite database that is essentially is just basically a database that comprises of survey data for malaria. All of this data are now grouped together within this database and the process for micro stratification that is automated is now running using this data collected from HMIS from health facility registry and within the composite which include sensors that are like what population, etc. So the result is actually with all this together. And with the automation automated process, the malaria risk map is hence generated that can be used for planning and this now is sent to another system which is called a plan rate. A PRO system that deals with the planning so the map from DHS to now goes as the input into the planning system. So just a little bit of what I have said these are criteria for data processing. Like checking for so the data that was used these lab and the ANC data and of course the cleaning in terms of reporting where we were looking at the missing reporting reporting periods. Also we are using we're looking at the empty elements across the data. Also we are all only targeting the facility that are performing MRD test for the data that we we used. Also we checked for consistency. And also for reporting rate in order to remove the data that may give a little bit of wrong result around the certification. So the data that had been used no indicators included testing positivity rate. Annual parasite index which also uses the population data as denominator, but also the testing for positivity rate from ANC. And of course for mapping as I said these were the facility that mapped to the word and the criteria that had already been created from the malaria prevalence data is in this way. So for each category of indicator that are different it is on how you can say this is low, very low, etc. And again, as we are calculating the score. So this is how from the criteria you establish a score for different indicator and then you can have a total score around those indicator. So with this just to mention all this process from now using the criteria to establishing a final strata had been using the HS to predictors. Thanks to predictors we were able to more likely calculate the total score and also in the end, calculate the final strata to what now you are seeing in the stratification map. So, as you may see, a little bit of achievement on this is now this stratification map is now available in the malaria composite database or they call malaria composing management information system that more likes have contents data that normal HMIS data do not have. And then this report is now sent to plan rep as I highlighted earlier for planning. So a little bit of achievement is that we managed to customize this and of course this tool is automatic. And every time of course every year it runs to provide necessary strata for plan for monitoring and every three years it runs to establish a strata for planning. Future work for this is actually from DHS to to plan rep now we are sharing the data manually so we are looking to integrating with the plan rep to allow the automation of sending the data. But also we will be also integrating with health facility registry more like to ensure the continuous update of the data. But now this process is now having some gaps because there are some words that do not have facilities, and you cannot create a strata without having the facility data. So we are looking into different situation on how we could do this. One way it could be to use the intersecting word around using the strata around other words in order to establish a strata or more like other approaches like a geospatial modeling and etc. So these are things that we are hoping to do in the near future for those words that do not have facility. Thank you. So, mostly, I need to acknowledge my colleague from the University of Jerusalem. Tempty program and also from the government, especially in MCP in the Swiss embassy. So, thank you. Thank you very much. Also, like, I also also got very excited with with maps. So I mean, I'm very, very easily excitable in all fairness. The next presentation. I'm just leaving it to you as a default, because I'm terrified is from Dr was each of it. And who is the health advisor for the HR strengthening project in Pakistan for the UK. Unfortunately, he also didn't make it in person, but he's presenting online. Maybe if I can, if I can manage everything. Can you. Okay, he started sharing already fantastic. But should I stop sharing these. No, it's already sharing. Okay. Okay. Okay. Can you can you try to talk about that. I haven't heard you yet. Yes. Yes, we see your screen perfectly. Thank you very much. Thank you very much. So, hello everyone. And good afternoon. It's almost prayer time here in Pakistan. So I'm Dr say it was a job it currently I'm working as health advisor with UK health security agency here in Pakistan, and we're looking after the implementation of ideas are I HR work in in next few minutes I'm going to produce the introduction of DHS to system to strengthen the dengue vector surveillance in the northwest western province in Pakistan, which is known as habit. So we knew. We knew that vector surveillance is basically an ongoing systematic process. Can I ask you if you can just like I put it in like a presentation mode so it's bigger for us also to see. Okay. There's fine now. Perfect. Thank you very much. Sorry. Okay, so as we know that vector surveillance is an ongoing systematic process of collection analysis interpretation and dissemination of vector related information to guide the response activities are evidence based interventions. The rationale to establish a vector surveillance is to establish the presence of factor, identify the mosquito breeding sites both indoor and outdoor sites, monitor the high risk areas and population and guide prevention control strategies. The vectors surveillance also help us to evaluate the effectiveness of different vector control activities. Dengue fever if you're talking about Pakistan so Dengue fever is now endemic disease in Pakistan the first case of Dengue was reported way back in 70s and then the separated cases in the 80s and 90s. But it is no endemic disease with multiple outbreaks and epidemic of Dengue fever reported each year. And across multiple geographic locations in Pakistan. As of now, in 2023 the risk of Dengue fever is assessed as high due to predisposing risk factors in the country and also the conducive environment for Dengue vector to reproduce into adult at ease. Also, you must have an idea about the catastrophic historic floods of Pakistan due to the climate change. So it has increased the risk of Dengue fever outbreaks and epidemics in the flood affected areas. Unfortunately, as I speak with you guys so the coastal areas in Pakistan is facing a huge challenge. Due to a cyclone bepper joy the name it bepper joy so we are expecting heavy rainfall cyclone and potential flooding in the interior. Southern province of Pakistan so again the risk of Dengue will be huge in those areas. Pakistan talking about the Pakistan is a federating state with four provinces and two region. The Northwestern province where we established that system is is known as Khabar to Khunha province, which is well known for his for its tribal areas bordering one is done. Historically the province is the gateway to Central Asia. KP province experience a huge Dengue fever epidemic in 2017 and 2018 when 25,000 lap informed cases got reported with more than 100 times. So before the Dengue epidemic of 2017 and 18 the government of KP initiated the integrated vector management program for effective prevention control of vector bond diseases. And with special focus on malaria, Dengue fever and cutaneous Lishminiasis. Under IBM program the vector surveillance methods are broadly based on OVR larvae traps. So you guys have an idea that Pakistan is a resource constrained country so due to the limited resources, not only the financial resources but the technical, limited technical capacity. The main method used by the IBM program is the pupil or the larvae survey here in Pakistan. The activity is performed by IBM program for vector surveillance method is characterized into three phases, which starts from the mobilization of outreach vector surveillance team. Then in the second phase the inspection of houses and mosquito breeding sites and it ends with the lab confirmation of samples which the teams collected from the community. The data collected in the field and the lab is there's the program calculate. We call it the vector indices, basically, and these indices basically the program, which, which program calculates basically characterize the risk of Dengue fever as high, medium or low across different geographic locations. So for a little bit of context on the Dengue fever so Dengue fever is a priority infectious notifiable disease in Pakistan. UKL security agency along with other partners is working with the Department of Health here in Pakistan since 2017 to strengthen the priority infectious disease surveillance in Pakistan by implementing integrated disease surveillance system by focusing on three pillar system strengthening approach. This is basically the snapshot of the Dengue fever cases reported by the KP province in last three years. Through IDSR DHIS2 system, which you can just say established in collaboration with National Institute of Health and Provincial Health Department. This was the first time when DHIS2 digitalization and digitalized disease surveillance for the Dengue fever was established in any area of the Pakistan. So in my previous slide I mentioned that IVM program was struggling with technical and logistical resources to achieve multiple program indicators which also includes the digitalization of surveillance method. So UKL just say after establishing the human Dengue fever surveillance in the province work with the IVM program to digitalize the vector surveillance by introducing DHIS2 vector surveillance system. The layout on your screen is the data collection and flow mechanism which starts at the community level right at the community level where outreach vector surveillance team basically visits houses, indoor areas, houses or outdoor areas, the mosquito breeding sites like ponds, tire shops and the potential high risk container site outside. So in the second phase the team records the response against each variable on a survey questionnaire. The survey form is then uploaded on DHIS2 digital platform in real time. The data is visible at the district and the provincial level. So district is the starting administrative unit here in Pakistan and provincial is the headquarter above the district level. So at the same time the data is being visible at the district and provincial level and the data analyst at both level basically can analyze the data by using the DHIS2 system analytics. After the analysis the information is interpreted and shared with the relevant stakeholders for informed decision making or the information can be used to guide the response activities where there is high risk indices based on the data collected by the team. The whole process of data collection and flow mechanism is on the weekly basis currently but it has the provision we can switch the frequency to the daily basis if there is any outbreak or epidemic of then if fever is reported. So here is the DHIS2 output on which you can see the responses against different variables. This draft and live table basically gives you the snapshot of number of houses and container examined by the vector surveillance teams along with the number of houses and container which were found with the presence of dengue larva which they recorded as positive houses are containers. The system automatically calculates the different indices which I mentioned in my previous slide and the relevant people can easily identify the risk of dengue fever where there is a high indices the container household are the Brito index. The program knew that high index area through DHIS2 system and they can basically guide the response in that specific area of the population. You can see that output currently for the whole province but the system has the ability to calculate these indices and the data can be shown in different districts are right at the community level at the UC or the CSC level. So the next output is the spot mapping. This is basically the whole of the KP province showing different districts. So that data basically using the DHIS2 advanced analytics can transform into the spot maps and gives the district with high vector indices which is the direct indication of high risk for the dengue vector. So it basically had the program to direct resources and response to these area are prioritized for the intervention very easily by referring to these spot map maps. This is another output which gives you the indication of houses where dengue larva was found by vector teams across different epidemiological weeks. Also, the system basically tells you the presence of containers. These are we call them the high risk containers for the dengue. So the data can be analyzed and can be presented in the presence of high risk containers where dengue larva was positive across different epidemiological weeks. Also, we added into the system to record the capacity building or training of the workshop and gives a snapshot to the program manager and partners that how many trained male or the female staff is available at the community level or at the district level and how many were trained in different weeks. We know that community the engagement of community is very important for the prevention control of dengue fever. So the DHIS2 also captures the engagement of local communities for effective prevention control of dengue fever. The system offers to record community sessions conducted by the program social mobilizers right in the community level. So in the end, how the DHIS2 system or the vector surveillance system basically help Pakistan and the province, because it was the first initial phase of the vector surveillance onto DHIS2 system. So after the digitalization of vector surveillance method the program captured real time and timely information on the presence of dengue vector which helped them to prioritize the areas and population of high risk. The DHIS2 vector surveillance dashboard is also accessible to other sectors which ultimately improve the multi sector coordination and collaboration, which is vital for the dengue fever prevention and control. Since the DHIS2 vector surveillance system was established as I have told you very recently in 2022, the quality data generated by the system helped the IVM program to revise the dengue prevention control action plan for this year, 2023. And added few more variables to the system which will make it more helpful to early detect the dengue vector and ultimately prevent the spread of dengue fever in this summer. Basically just couple of days before we receive an official request from IVM program in the province to basically add few more indicators to the system and build the capacity of their staff onto the system. With this I want to acknowledge the support of these people for making this system possible in a very complex political and administrative environment here in Pakistan. Then I'm thankful to the DHIS2 conference organizers, scientific committee for giving me the chance to present my work. It's just the start of the DHIS2 in Pakistan and I'm looking forward to work with University of Oslo for the rollout of DHIS2 in context of one health and IHR climate change, a lot to work in the coming years. Taking the opportunity I wanted to say a special thanks to Victoria for all her support with regards to my visa problems, although I didn't get at this time but sooner or later I will manage to meet all the people in Oslo. Thanks again. Thank you very much, Wasif. I hope this is actually like an inspirational presentation as normally vector control and entomology tends to be a little bit forgotten especially in national systems and you can see like how important it can become when you're trying for vector control in general vector related diseases. So thank you so much, Dr. Wasif. I am very proud of my presenters because they've been unbelievably on time. So I was wondering like if you had any questions that you would like to ask them about about their work or pretty clear. Yay. Okay, we have one. I'm really excited about the stratification work. My question was around, I know in Tanzania, it's not the Ministry of Health is managing the implementation. It's P. O. Rouch, right? So my question is really around access to DHIS2 and whether the Ministry of Health is the one who were sort of the primary users. What does Tami Sami also have access to DHIS2 and how do you build the capacity of sort of non-health workers in some ways to use these maps, these stratification maps? My other question is around sort of indicators for stratification coverage. I know there's a mixed package of interventions depending on the strata. Are you going to maybe have indicators with color coded maps which show then how you're implementing the stratification interventions? Thank you. I hope I could be able to answer those as a system developer, but I will try. So in Tanzania actually, P. O. Rouch in most cases at national level are the main implementers, but they are working hand in hand with the Ministry of Health. Just quickly to say is all primary level health facility belong to P. O. Rouch. Only district, region and national hospital are the ones that are being managed by the Ministry of Health. But Ministry of Health is actually creating some policies and they live up to P. O. Rouch to more implement everything. So they are really collaborating. And for this work stratification, the outputs that comes in from DHIS2 are being used in the plan rep system, which is being owned by P. O. Rouch. So there's that collaboration and the users are using this data. So as for non-health, that could be tricky for me, but as I know in CHMTs or in the RHMTs section there are health officers around who uses this data to more likely prepare some different plans. For example, there are very specific in malaria planning is meetings that are being conducted within P. O. Rouch that uses this data. So I may not be really satient, but hope that answer your first question. As for the second question, yes, I only showed a stratification map, the final result, but in the same system there, we put it another visualization like a table that shows the data for those indicators that I highlighted and show how this data brings in this strata. So I just only showed what was cool to present, but there are more that can be viewed within the dashboard. Hope that answered your questions. Okay, thank you. I don't know if you have any questions, if there's someone else with questions. No, but then I declare this session closed. And yeah, thank you very much and thank you for everyone online and hope this session clarified some of your doubts and inspired you to use further the HIS. So thank you very much. And if you have any other question to like our presenters, please feel free to let me know and I can put you in touch. Cheers.