 So welcome everyone to this DHAs2UscaseBazaar for the DHAs2 annual conference 2020, the digital one. I'm the unofficial host. Alice is the official host. She's wrapping up a session. So I'm starting on her behalf and she will obviously join us as soon as she can. So you've been waiting a bit now. So what I suggest is that we start right away. We have 12 presentations. So I will ask each of the presenter just let me remind the rules to present each of the presenters who present for five minutes and then we go to the next. So I'm inviting Lucia from WHO who's presenting Metadata Sync, multiple DHAs2 instances, interacting to timely serve decision making at all levels. And the next to come will be Fanatenan from Madagascar who will present integration of private sector data into the National HMIS approaches and experiences from Madagascar. But first, WHO, the floor is yours. Hello, good afternoon everyone or good morning. Since we have only five minutes, I'll start right away. We are here today to present to you an application, the DHAs2 application that we have developed to facilitate the exchange of metadata and data from one DHAs2 instance to one or many others. We were consuming, we were spending a lot of time doing this process in a fairly manual way. We were tired of it and we wanted to find a solution to automatize direct change of data and metadata. I'm Lucia Fernández-Mondoya from WHO and I have with me today Ignacio Fosche Perez, which is representing the developer ICT of this application. And we'll be able to respond to your questions. And with us, we have also Ryan Williams from WHO who has been very involved in the development of this application. So what is this application doing? So this application can help us compare a map, one DHAs2 instance against one or many other ones, and then send data or metadata from one to one or many others, applying all the necessary transformations on the flight. So based on the mapping we've done, based on the equivalence of the different UIDs or data elements, indicators or units, the application will transform the data and metadata from the origin instance and send it to the destination or remote instance. And what is best, that is my favorite, you can schedule these processes to happen on a regular basis. So if you have to send metadata or data from your development instances to your production instances, or from a national instance to a regional instance or from a national instance to a global instance, you can schedule these to happen once a week, once a month, twice a week in whatever frequency you want. So I wanted to show you a bit of the screenshots because we have respected the look and feel of DHAs2. So the application is easy to use and familiar to all the DHAs2 users. And what you see here is how from one DHAs2 instance where you have this application installed, you can configure a lot of other remote instances where you want to send your data or metadata. And then here at the bottom you see how you can map the metadata, in this case the originates, of your local instance to the IDs or elements of the destination instance that you want to send data or metadata to. So that mapping is then saved under each of the configurations of each of the instances. And then it can be used to send data and metadata. So here you see the process to set up a synchronization rule. So the process of sending data or metadata to the destination instance. So what kind of things can you define? You can select what kind of data elements, what subset of indicators, or what subsets of metadata you want to send from your local instance to a destination instance. You can select if you are sending data for which period you want to send the data, for which organizational units you want to send the data, maybe you don't want to send for all of the originates, and you can aggregate it. So if one instance is collecting data with a much more granular detail and you want to send that data to another instance that has a broader scope, then you can aggregate the data that you are sending from, for example, to a facility to province level or to national level. And as you see in the step number eight, you can schedule how often this process of sending data or metadata should happen. Once you schedule, you set up this process and you have selected what kind of things you want to send from one instance to another one. You have a list of what we call synchronization rules, and you can literally just monitor how they are being executed automatically without you doing anything, or you can execute them manually or you can download the JSON file that is going to be sent from this synchronization rule, etc. You can also, as usually in the HSS2, you can modify the sharing settings so that only a few subset of users have access to executing these synchronization rules so you can control access to them quite easily. And then once you have synchronized... Once the data is synchronized, once these synchronization rules are run, then you are able to see whether they are successful, whether they have failed and why they have failed if there was any changes in the metadata of the origin instance in between, etc. So you have a nice dashboard where you can see these processes either succeeding or failing. And I just wanted to quickly end the super short presentation saying that we have broadened the scope of this application very recently, and it not only can send data on metadata from one instance to many others very easily, but now it is able as well to package metadata, to create metadata packages and to offer them to users. So basically we have a couple of widget apps that will be integrated in the DHS2 dashboard very easily, and these widgets can provide packages that are contained in the reference instance or reference instance where the widget is installed. So we are using this to distribute the WHO standard metadata packages to allow users to explore the packages, to explore the content of the packages and to pull them into their national or whatever instances. So this application since recently not only pushes data out from one instance to many others, but it can pull as well, you can use it to pull data from an instance that you are exploring, visiting, consulting into your own instance. So we are using this application to manage our own DHS2 instance at WHO, transferring data and metadata from DEF to pre-prod, to production. We are using it to support the countries to do the same in their national environments to exchanging data and metadata from DEF to pre-prod, etc. And we are starting to use it as well to facilitate data reporting so that member states can report data to WHO. There is a subset of data, only a few variables and indicators they want, easily from their national instances to our global DHS2 instance and finally to distribute the WHO standard packages. So this is it for my presentation, I have a bunch of links there so that you can get all the documentation you want. There is a video, there is written documentation, you have a link to the GitHub report to them, the app, an announcement in the community of practice and our email accounts in case you want to contact us directly with any questions. So thanks to everyone that has participated, as I said, we have Raya and Nacho and myself here, but it is a lot of other people that contributed to it, a lot of developers behind this application and I wanted to thank them before finishing the presentation. Ova, to you. Awesome, thanks a lot Lucia, you nailed it in six minutes, that's really, really good. Thanks for the presentation, I have two good news, the first one is that you can ask questions on the COP, I put the link in the chat, we'll post it again in a couple of minutes for the people who had not joined while I was sending the first link, I sent it a second time and the second good news is that you were all waiting for Alice to be on board and now she's on the floor so she will take over Alice, the floor is yours and the next to come is from Madagascar. Exactly. Hi everyone, thank you Mathieu. So we're going to listen to the next presentation which will focus on integration of private sector data into the National HMIS approaches and experiences from Madagascar. Hi everyone. Hi, welcome. I'm Fana Tinan, I'm from Madagascar as you said before and I'm from the Shops Plus project and today we'll talk about the integration of a private sector that ties into the National HMIS in Madagascar. So to start let's talk a little bit about Madagascar, actually here we are still transitioning to the exclusive rules of the DHS2. The Ministry of Health is now finalizing the process and the tools and the ministry also seek inclusion of private sector data because in 290% of the private facility reported to district level. Actually the facts are that the private facilities are not in the master facility list. The reporting tools are cumbersome and confusing because it's composed of 10 to 12 pages of tables and indicators should be calculated manually so they don't find any motivation to report. So what is Shops Plus? Actually Shops Plus is a new site deflux ship initiative in private sector health. It's a 5-year project so in Madagascar it will end next year. One of the objectives of the project is to improve the routine reporting by private providers and international HMIS for opportunities were used. Let's see the first one. It's about the public-private dialogue. For that we established a private sector platform to advocate and for peer learning. And we developed a roadmap with the Ministry of Health and private sector. The main objective is to improve the reporting from the private sector but also to include them into the HMIS. So by now we facilitated five meetings between the two entities. The second approach is about the sensors that we conducted for the private sector. So we discovered that more than about half of private facilities in Madagascar are located in the capital region which is a little bit normal. 1100 private facilities are in the capital region and 36% of them have access to a computer and the internet. The third approach is about technical support we provided to private sector. So from the roadmap we identified that the main problem for the private sector to report is the tool is confusing cumbersome. So we developed an Excel-based reporting form so the calculation will be automatic and it will be easier to be sent but because they just have to send it by email. We also provided training on Excel and paper-based form for private providers. By now 107 private facilities from the capital received training. About 70% of the submissions from the trained providers are the Excel-based form and 60% of trained private providers have reported at least once after that training. It's really low but it shows that we still have barriers and challenges for reporting from the private sector. The last approach is about capacity building of public sector. For that we established two formal agreements with the Ministry of Health from the district level to the national level. This is to integrate the private sector data into the national HMAS but also we bolster the capacity of the district to accept on the process Excel and paper-based report from the private providers. For that we provided them equipment, trainings and technical support. So these are the activities that we have done but what are next steps then for Shops Plus? So by now we plan to produce and disseminate informational and promotional videos for reporting. This is to make the two level label at any time, anywhere at any level and from the public sector and the private sector. We'll continue trainings. We continue supporting the health district with data processing. We'll ensure that all the information we gathered from the sensors will be entered into the DHAS too and we'll ensure that all activities will be sustainable once Shops Plus ends. So these are our realization, our perspective that we wanted to share with you. Thank you for your attention. Thank you so much Fana. Now we're going to hear to the third presentation from Rosario Martinez Vega, MSF Spain. Rosario, you have the floor. Thank you. Okay, can you all hear? Can you all see my presentation now? Yes, we can. Okay, so good afternoon everyone. I'm Rosario Martinez. I'm the health epidemiologist in MSF Spain and we are going to present our use case which is in service mobile data collection for automatic integration of individual data in MSF HMIS. So why did we do this pilot? Just to give you some background. So in MSF Spain we developed our HMIS back in 2014 based in DHAS too and we did the full implementation in 2015. So in 2018, four years later, we decided to do an exercise of evaluation in collaboration with the University of Oslo trying to understand which were the main challenges faced by the field user regarding data collection and data analysis. So the main challenge we found was that data collection process was identified as repetitive, time-consuming and prone to human errors. So this affects a lot the user acceptance and the data quality and then basically the users have very little trust in data and therefore they didn't use the data much. So what did we do? So the solution in response to this challenge, so we designed and piloted a system trying to assess the use of tablets for improving in-service data collection. So the system consistent of the DHAS to Android app for data collection for individual registries and a custom web app that integrated those records into the DHAS has aggregated data. In order to do that, we need to map all existing registry book to be sure that all the data that the users need to collect, they were integrated into the app. But actually what did the user could see in the hospital? So this is the normal data flow that you all know. We have the clinical files, then we enter every patient in the registry book, then we do a weekly tally of the data, then we enter into HMIS and then we synchronize with the server. And all of these, the three first steps happen a facility level and the last two steps approach a level. So what we do with our proposal was we keep the first two steps but then we enter directly the data from the registry book into the tablet and then we synchronize with the server. So we basically skip these two steps who were identified already as like very time-consuming and very prompt wearers. We could also have a skip this step of the registry book but we decided not to do it because it was a pilot and the people in the field really trust the registry book. So we decided to keep it but it was truly duplicating the work. So how did we pilot it? So we pilot the solution in seven inpatient services in Malacal Town and POC hospitals in South Sudan for four months. We went to the lab of the tablet in the different words where you can see we leave the tablet in the words and these are the picture from the two different hospitals. We went to the field and we did a hands-on training for two weeks. We trained the usual staff that they were working before and this pilot with the data who in case of Malacal was the national staff, they were the next supervisors and then we left after two weeks. So the remote monitoring phases started, we worked independently and we provide remote support. At the end of the four months of the pilot we did an evaluation to say the mobile data collection has an alternative to the previous manual paper-based approach. So we did focus group, we did surveys and we also did like a qualitative analysis with HMI and Zata. What did we learn from it? So we learned that mobile technology to support data collection seemed to be feasible and acceptable in IPD facilities. One of the participants mentioned that it was easy to use the app, it was easy to use the tablet and they felt comfortable having the tablet in the word in the everyday today. What did we learn about efficiency? So collecting data in the tablet and running automatic aggregation was faster than the previous manual process. Regarding effectiveness, the easy of data collection process improved mainly due to the replacement of the manual aggregation. The data did not change substantially even though all the users have the feeling that the data was more complete but it didn't happen in the evaluation but the timely report really improved. And finally regarding quality, data accuracy improved significantly. So very important indicators for monitoring the activities in the field such as better occupancy rate, impatient length of stay, a total of admission really, really improved. The data was better and more accurate. In terms of processes, they pre-data interface had a great impact on data quality. So the conclusions of this pilot was like bringing technology closer to where the data is generated appears to simplify data collection and improve data quality. And that the automatic aggregation of details to reducing manual steps seems to improve user motivations toward data collection. But also, of course, we also have challenges during the pilot. One was the provision of technical support and one was the hardware management. So we have to bring the tablet every day to the hospital and bring it back every evening. The digital skill of the user which arrives a lot from person to person and finally the custom web app for goodness. And which are the next steps? The first is the potential of analyzing the individual data. We were trying to pilot with the solution was the aggregation of data. So the user only have access to aggregated data. But in fact, we have individual data and we have to explore the potential of that and we know that is a lot. So that would be the next step. Second was leveraging mobile device presence in the facility for other health services. So we piloted in IPD but there is other services like for example emergency room that for sure will benefit a lot from the solution. And the last one, so further use of the application. So this is a sentence that we get from one of the participants in the interviews and he said, I've worked with data for many years and I've been working with MSF and working with Taliesheets to me was like going back to the 80s when in a society with data collection tools where Taliesheets are not a tool anymore. So this was a very important lesson for us. I think as MSF we're always afraid to bring technology to the field because people are very busy. We think that they may prefer to use the old ways but what we saw is no. I mean the staff in the field, they ask us for tools to help them to go through the day today and to make their job more reliable. So I think we should be more courageous with this. And just to give a thank you to all the co-authors and all the participants in the pilot but especially the people who participate from Malacal project. And thank you to all of you. Dan. Thank you so much. So as mentioned earlier, once again any questions, please go on to the COP page and you can directly ask, I mean directly write your questions there. We'll make sure to reply. Now we're going to hear the next presentation from Romain Roland-Tourry on Can Technology, Presenting Sari, the following subject. Can technology turn the data use dream into reality, case of Malaria scorecard and dashboard mobile apps. Romain. You can go ahead. Okay. I was trying to find the mic. Okay. Good morning, everyone. So we don't have too much time, but so quickly, what is this about is we decided to bring some the use of Malaria data closer to the users. And we let's go this way. Yeah. So we try to tackle some user use cases that, for instance, at the district level, then the district staff want to do some supervision at the facility level. They need some they need some facility data and the stuff that they can use to do some data review and other things. And facility staff as well, they sometimes need to have the data review to do some performance assessment assessment or some planning and this kind of stuff. And we started from the assumption that the users at the decentralized level they are usually not comfortable building some visualization themselves themselves and they are also not very comfortable using the native GHIS to visualization tools or sometimes they don't even have access to these tools to build their own dashboard and things like that. So in most of the case we don't have permanent access to internet and when we go in supervision they just maybe don't have access to internet at all. So we are targeting users from district level and facility level and even community level. And for that we did was first to have our TMI measure evaluation data use expect put together a series of indicators data set and domain and try to see what are the most relevant indicators. So we have a list of indicators here since in the interest of time you can check that later on the presentation. So we did some list of indicators for the first tool that is called the dashboard it's generating a set of graphs that's a dashboard and we have a second list of indicator that is specifically for a scorecard. And so for the scorecard the difference is that in addition to the indicator we also have some target like a key in progress or not in track. So we have this set of indicators and now we use that to build a specific DHS2 app. So we actually built two DHS2 app one that is generating a dashboard for malaria data to be used and one that is developing a scorecard. I will start with malaria one it's composed on a web base and a mobile app. So for the web base we use React and you can see and as usual you can start it like this and then you have a first screen that allows you to match the set of indicators that I presented earlier with what you have in your DHS2 system and then you can select the relative period or fixed period as usual and your org in it and then it generates a predefined dashboard for you and you can use it on the web and each of the graphs you can have some detail you can play with them and then we developed a mobile app that can go with that web app. So the web app is also acting as back end to configure the mobile app. So with the mobile app you it's this is how it's present itself. I mean it's an app that allows you to connect any DHS2 system that already has the web app installed and it can also manage several servers and you have access to all the data set here and then you can select your org in it and when you submit the data will actually actually log design the dashboard. And one cool feature with the app is that you can share your data your graph or your dashboard using social media installed on your mobile device and the app has offline mode as well so we don't want to overload your phone so it's only the data that you are interested in that is uploaded in your mobile phone if you are not connected to internet you can access it. So and we have the same also for the scorecard application a web app version that you can use to first configure match your org in it and then match your your indicators and then set up the and then from there you can generate you can generate your scorecard based on the org in it that you want to see and when you have it you can use your mobile phone as well and install the app on it and you can see the matching of the different of those different indicators and from there you can select your org in it and the period you want and then it's you can generate the scorecard on your mobile phone and you can also share that scorecard with other followers on social media and yeah so I went very quickly because I'm interested every time but this is our team and we have some videos on YouTube and things like that as well thank you thank you for this presentation now we're going to hear the next presenter who is Mauro Tauban Mauro is going to present on using middleware to connect the HIS2 for TB and AMR in country implementation experiences Mauro you have the floor okay thank you very much hello everybody let me set my starter and try and get through this in five minutes Mauro Tauban I'm a senior technical officer at Find Foundation for innovative new diagnostics and what I really want to talk about is how we've put interoperability in the heart of our DHRs to implementations for TB and antimicrobial resistance in our country implementations so very quickly if I can get this to change just find at a glance Find is an NGO headquartered in Geneva, Switzerland and is principally focused on diagnostic developments and infectious disease I work in the connectivity and interoperability area and our focus is on mobilising health data in order really to fight disease what we have done is really challenged ourselves with these implementations to do a number of things principally is to avoid the burden of manual data collection everyone is data hungry issues like antimicrobial resistance demand enormous amounts of data from many different areas and we are focused on automation we do not want to add burden to already incredibly busy healthcare professionals so that is one of a couple of mantras that really drive our work including simplifying implementation and creating solutions that are integrated across multiple components but are manageable our two use cases for our projects have been antimicrobial resistance surveillance looking at pathogens drug susceptibility testing and also looking at TB management so they seem quite different disease issues but from a technology perspective they are really about working with diverse data sets and trying to map them into services that allow you to manage the various conditions so for AMR it is very much about this complicated isolate drug susceptibility testing results coming from multiple data sources whether they be human health animal health antibiotic usage on the one hand and then in the TB area very much about connecting fleets of diagnostic devices and being able to run that data source into multiple endpoints which may be focused on network management surveillance so our technology solution has put interoperability in the centre of the picture so that we can map from these multiple input sources into DHIS2 and maintain a tight integration between what is effectively an end-to-end data model that allows data from these different sources to flow through and then integrate into things like the WHOTB apps into our own bespoke dashboards and also into existing services in-country analytics products the interoperability also supports kind of machine to machine data exchange so we can look at things like integrating into products such as in-cap or open-clinica as well as into other health management systems so what is all this kind of taught us what we've end up doing is building a effectively a reusable platform out of these major open source components with these end-to-end metadata packages that sit across those components that we have found that with our work in Zambia and in Senegal that we have been able to develop innovations in those countries and then those countries can then inherit those innovations so we're starting to compile a reusable deployable service that is built on the innovations and learnings from the individual in-country projects and the big thing we learn really from the AMR work is that we managed in our first country we spent 12 months developing indicators dashboards and integrations the next country was 12 weeks and we think we can bring that down to 12 days so in all this work we have developed quite a capability it's quite far-ranging in what it can do with connectivity and property and we hope to be able to get a lot more countries a lot more use cases and get a lot more time so I can come and tell you in more detail about how this works the final thing is to say thank you and acknowledgements to our donors country partners and the project group eShift partners particularly and software for health foundation thank you very much thank you Mo so next presenter from East South Africa so Nora is going to present a UNICEF landscape analysis of routine nutrition data in Eastern and Southern Africa the final outcome Nora you have the floor thank you so much thank you are we good to go yes we are maybe the presenter mode and then good to go I'm like Arthur picked it but okay so good afternoon everybody I presented the interim results at last year's conference and I thought that I would wrap up with what happened at the end the final outcome and I'd like to acknowledge Nora and Maria so this was the background it's the normal things of having a look and how can UNICEF do better and also to look at the global and regional standard set of indicators because nutrition is still a little bit in the weeds this is the map of Eastern and Southern Africa for the UNICEF region 18 countries are using DHI to have subsequently migrated to that and it just leaves us with one lone little country that does not have aggregated data and they understand that they're in problem and of course there's also the issues with three languages so in order to do this assessment of looking at routine nutrition data I needed access to DHIs too and then I also looked at what policy documents, what indicators did they use, aspects related to data quality how did they use it in other information systems so here we just have a list of all the countries who was using what was using, what sort of rights and what sort of access did I have to DHIs too to do an assessment so some of the key findings were the main one was that there was supplementary nutrition information systems and one of the reason for this was that they were easier to access easier to get changed easier to adapt to local needs and everybody worked together and a supplementary is not a parallel it contained different sets of data there was use of information with review meetings and I ever looking at DHIs to substantial problems data was collected not used and I was looking at what was the data quality and the other bad news was that so many people did not have access to basic apps the data quality app is not available to them the map app is not available to them and lots of people wanted training the other thing that we identified was when did you last update your nutrition indicators or when are you revising what's the sort of timeline because we know data needs change over time and 13 countries did not have a date to do this and 7 countries have a set timeline in other words they have a policy that says we will update our whole set of indicators, data elements and everything else access to DHIs to dashboards, only 10 countries provided access the others we had to do all sorts of funery things and very few countries had a functional nutrition dashboard and so there's a lot of work that needs to be done about how do you get data out in a format that people can use people were using the data that was very that came across everywhere very strongly the other scary thing was that the version of DHIs to being used in when we did the final sort of counting November less than 2.3 there were 5 countries 2.27 2.28 those are very old there was no upkeep for them but countries hadn't changed yet this was a big issue that not updating the bills and if you want to get this report online here it is thank you that's all from me thank you so much Nora we're going to hear now the next presentation which is from Caroline Bain from PAF who will be presenting trace to treat tracking breast cancer patients in Peru with DHIs Caroline you have the floor thank you so much hello and I guess it's good afternoon to everyone it's good morning from me I am Caroline Bain as she said I'm a senior program officer at PAF and focused on detecting breast cancer in low resources areas so in this point we're working in Peru and I'm happy to tell you about breast cancer patients with DHIs too we've just started this year and here's the poster on the left and I think everyone knows that Peru is in South America on the west coast but I'll show you where Trujillo is Lima is there with a star and then Trujillo is about one hour flight or an eight hour bus ride up to the north on the pacific coast and we've been scaling up this breast cancer model for the past four years since 2016 and the model is four places where women do not have regular access to mammograms so there are a lot of places in the world where there might be mammograms in the main capital city they're not necessarily in the regional area so we're focused this model on bringing breast cancer detection to women in those situations through along the red access community education teaching women about the risks and signs and symptoms of breast cancer encouraging them to go for a clinical breast exam every year and that is available in their neighborhood clinic with their midwives that have been trained if the midwife was to find a suspicious mass they would be sent on for an ultrasound triage which could detect whether it was just a simple test and then could be just trained and not followed up any further but if they still see a suspicious mass they do a fine needle biopsy and then that slide is sent on to the regional cancer center for evaluation by the pathologist and they would have definitive diagnosis and treatment also at the regional cancer center so this model of using clinical breast exam triage and fine needle aspiration biopsy has been resource appropriate and accepted by women and providers we do see about 6,000 women in Trujillo per year age 40 to 69 evaluated for breast cancer and so far everything has been done just in paper based forms which is a recurrent theme I've seen here in the DHIS2 conference and that's what we're addressing as well is trying to get that digitized for real time follow up we've had an excellent collaboration between PATH and the Peruvian Regional Ministry of Health which is the city of Trujillo's health network and EJAS in Spain has been building the system so we have weekly meetings and I just can't highlight that enough it's an excellent way forward the system design is set up to address these three different levels of the clinic the government health system from the first level where the clinical breast exam is done the midwives will be trained to input the data about the clinical breast exam at that neighborhood clinic and then if they're sent on for a follow up the secondary level hospital would also input the Tultra Sound Triage and the fine needle aspiration data and finally the third level where the diagnosis and treatment would be input we will have that information as well this is very exciting for us because we have never seen everything in real time and connected to show all this information about the woman in her breast cancer detection pathway so it's a very exciting moment we have had to pivot from what was going to be in hands on training in person to virtual because of COVID-19 and so we've done Zoom and Moodle and we'll continue to do that in fact next week we have another training starting up and the DHIS2 capture app can be used for clinics that lack the stable internet here is what we'll be collecting as far as data and charts these are just placeholders because due to COVID-19 we haven't had the breast cancer screening going on I think most of the world has put their cancer screening on hold as COVID-19 has been the priority but they are starting up now soon and we hope in the end of 2020 in the beginning of 2021 we will have data on all these areas clinical breast exams that will just sound the biopsies diagnosis treatment type and initiation etc and then some of the highlights here too is this is one of the few DHIS2 implementations in Spanish and one few in Latin American facts it's been a great iterative process sustainable and we hope will be used in the public sector has a user friendly approach and we are piloting in 14 health centers and the Regional Cancer Institute to engage at least 800 women but the whole city of Trujillo has 58 health centers so with a successful evaluation we hope to reach those other 44 health centers and that the regional government would be in charge of that and the health system so we are hopeful for that I want to thank all my collaborators at TAP and in Peru and in Spain and also acknowledge the Pfizer Foundation support thank you so much so now we are going to hear Uri Ba from PATH on the following presentation creating integrated surveillance across human and animal health sectors using DHIS2 the OHMIS system Uri, please go ahead thank you yes we can do the same ice cream please yes okay thank you everyone good afternoon maybe you want to put the presenter mode sorry slide show at the top okay thank you so I'm Uri Ba health informatics specialist I work for IDDS project IDDS and for infectious disease detection and surveillance project funded by USAID we are going to present the OHMIS system this OHMIS system is the system that we implement in Senegal and in other countries supported by IDDS project so the goal of this project is to help all the national government to have both animal and human priority genetic diseases data in the same platform using DHIS2 so in this presentation all data we are using is generated randomly using Senegal population and geography so let's see the methodology first we start by identifying and all the priority genetic diseases in the supported IDDS supported country project we develop we create and develop the data collection tool that we configure into the HIS2 that we generate dummy data for tests for diabetes cases and animal bites to export and we export this data in to SCAM for doing analysis to see the cluster events so here you have the data country form developed into DHIS2 the first one is the human data collection form and the second is the animal site in the animal site just what I want myself is we need to classify diseases by animal species so here the result of the SCAM analysis here we have the cluster the low cluster is represented by dark dots and the higher is represented by the red one so we can also cycle in the cluster zones okay let's see what will be the next step the next step will be to to configure this platform and integrate it with R to have real-time analysis and to automate some analysis into DHIS2 to create data visualization and after that it will to select IDDS project country to pilot this system and to see how to implement it in one of the countries supported by IDDS so the goal is to avoid any data sharing issues in the animal and the human side because when we talk about data infectious diseases or one health approach we need to have one platform that helps us to collect human and animal diseases in the same platform so for to host in this platform with government or the national government can choose to host system in the Ministry of Technology or in the other third parties depend on the country politics and to finalize I just want to thank all my colleagues Miquel and Lindsay and Roland who helps us to set up this abstract and to continue to work on this project and we will let you know what will be the results. Thank you and thank you all for me. Thank you so much for this presentation once again if you have any questions please go on the COP and you can write your questions you will make sure you receive an answer now let's hear from our next presenter we began from the Malawi MOH who is going to present surveillance for all to circumvent the barrier of network penetration and bring COVID surveillance to the masses case from Malawi oh let's stop this sharing yes Brigham please go ahead you can share your screen hello everyone I believe you can see my screen we can see thank you thank you hello again from the Minnesota of Malawi I work under the quality management directorates in the ministry and this is my presentation now as the title says this involves the work we did during the COVID-19 pandemic or as we are at now and we were trying to look for ways in which we could bring COVID-19 surveillance to everybody regardless of who they are you know in an efficient way as possible now as a background you know as a country Malawi we are working on setting up the one health surveillance platform by leveraging the DHS tracker implementation or tracker program and that was one track but then with the advent of COVID-19 that was fast tracks and focused towards COVID-19 the issues that were arising in terms of surveillance with the outbreak that's when we had diverged from the initial implementation to what we have now as one health surveillance now looking at the landscape of Malawi in terms of network penetration the trend that we had as of January 2020 population of about 18 million and then out of that 18 million we have 8.58 million people who have at least mobile phones that's just a mobile phone regardless of what factory it is and then from that 8.58 million we have about 2.8 million of people who have mobile phones that enable them to connect to the internet and then we needed a way or solution to fill the gap that is in between the 8.58 and the 2.81 million you know so that we can also provide surveillance for the people in that gap and speaking of the challenge to your right of the screen that's what our challenge was more or less like because such devices are what's most common within the population in the country and then we needed to leverage the capabilities of such devices for people using phones so that they can also be able to report or monitor what's happening during the pandemic now the solution that we came up with with you guys so that was we built an SMS and we built SMS and USB platforms on top of the capabilities so that we can register we can register the entities which are the people in this case we can have them submit their symptoms have them check the statistics of the pandemic both nationally and globally and then that was at least what we were able to achieve as the first iteration of the implementations and as an example that's I took the liberty of taking the screenshots for the two implementations in action and on the far left you see the SMS implementation which shows a few menus regarding what you can do with the app and then on the right you have the USB implementation which also has basically a menu on what the app is more or less capable of doing and as a summary we this is how it all looks like fit together when all is in action and maybe I was trying to look for the pointer but you should focus on the middle section where you have rapid flow unicef so that's basically what you would look into apart from that implementation there is the anchor dimension WhatsApp chat port but that assumes that people are connected to the internet which we were trying to avoid when we were coming up with this implementation so the good thing about SMS and DSST as you might be aware is that they work on any device regardless of what form factor it is so connected or not connected to the internet you should be able to use the key functionalities of those two platforms and then also write on top of the implementations that other people would be using using phones that have WhatsApp or directly using the DHS to track capture mobile application and then moving further to the challenges and lessons learned now in terms of the challenges we had a few things going on in that regard and much of it was to do with how to synchronize the data coming in from the different applications so we had data coming in from the WhatsApp chat port we had some coming in from the SMS application and one from the DSST implementation this is a case where somebody in the morning or today one would use WhatsApp chat port tomorrow they would use the DSST application and if you don't cut it for those writes you'd find that you have duplicate entries in the platform and that was maybe one of the challenges that we had and had to work around and the other one was collaborating with different partners that are sharing and that was brought about because those applications were not built by one player and they were built by several players helping out the ministry so we had a few things going on in terms of how to consolidate that information while not a few things and then as a way forward and as a way forward from the lessons learned we saw that it's very nice to adapt much of our implementations to comply with international standards as possible and that's what we will be working on you know from the project going forward so that's about it thank you very much for the time thank you so much now we're going to hear the next presenter Ibrahim Kamara is going to present Sierra Leone the rise and fall of the HMIS DHI2 data sets Ibrahim yes please thank you very much let me just try to share my screen yes can you see my screen please yes we can alright so my name is Ibrahim Kamara I'm the manager for the measure of health and sanitation in Sierra Leone and I'm co-presenting the rise and fall in size of the DHI2 database I want to start by acknowledging Nora and Khalif for their contribution to this work so Sierra Leone in case you don't know it's well known for being the demo site for DHI2 testing as well as leading on the EIDSR in the sub region we actually started using the DHI2 offline system in 2008 and in 2013 we moved to an online system whereby trying to integrate all program data into the DHI2 system and at this time we had more than 3000 data cells that are required to be completed and reported on a monthly basis and because of the decision to integrate all program data into the DHI2 we had to revise our forms around 2018 to 2019 to capture all program data as well as having separate data sets for HIV and TB program so in 2013 we had a lot of data collection tools that included age and gender desegregation this actually affected the quality of our DHI2 data system especially in terms of completeness and of timeliness so what happened most of our health facilities could not summarize and submit their form reforms on a monthly basis targeting the 15th of every month and we also realized that there is a lot of data inconsistency between what is reported in the register compared to what is in the summary form of the DHI2 so at around this time we realized that the data sets we have for the clinic registers was 2199 and the one we had for the hospital registers summary story was 1142 and that for the HIV was 923 so in 2018 we had to do some revision especially targeting the hospital data sets so we increased the data age desegregation as well as included sex to the summary forms so as a result we increased the data sets for the hospital forms from 1000 to 8000 in 2018 so in 2019 we did another form revision and this was because of the increase in data quality issues and we also realized that we have a lot of data elements that never analyzed or used to make any decision so on that note we used the WHA facility guide for Malaya EPI and RMNCH as one of our guidance documents to conduct the exercise we remove all we collapse all ages age desegregation as well as remove some gender in some of the data element that we think are not useful we ensured in all these process we ensured we involve the participation of the programs as well as the health partners in all the stages of the process and to presently we have 1000 1616 data sets in our monthly HIS2 system and we have conducted trainings for the newly revised suits and the health facility staff are very much excited about the changes we made so at this point I want to ask Nora Stoops to at least continue from this process from this point Nora are you there? Yes I am I think that it's a rise and fall but the rise again of Sierra Leone and the summary table shows us where we were what we went to and what we've come down as and I think that the process followed is crucial for countries that are looking to reduce their maximum data sets the support of the Ministry of Health in revising these in revisiting things and re-looking it was crucial it couldn't have been done without them so we now have a set of tools for all of these different services we have the registers and as I say the training has started and hopefully will be continued and full roll out and as Ibrahim said yes the clinic staff are very happy with the reduction in the number of data cells to be completed every month thank you. Thank you so much Ibrahim and Nora now let's hear to the final presentation of the session from Ryan Williams from the WHO is going to present D2 Docker making DHIS2 accessible to everyone Ryan you have the floor yes hi can you see my screen not yet not yet now we can see it yes perfect thank you great hi everyone my name is Ryan Williams I'm a Surveillance Officer at WHO working in the Global Malaria program and today I would like to talk about the D2 Docker making DHIS2 more accessible than non-IT specialists the D2 Docker was developed by ICT who is represented today by Nacho who is online and is available for any technical questions Lucia Fernandez who presented the metadata application was very involved in the development of this tool in case you want to contact her as well so the problem installing and running DHIS2 instant is not accessible to most people it requires a certain level of IT knowledge and IT infrastructure often lacking in among health specialists and professionals this limits the health professionals capacity to innovate and to use the tool and it hampers the full potential and the possibilities to improve on the health information system how many times during workshops or after presentations I have an approach about DHIS2 the participants say it's a great tool how can I get it some even have DHIS2 they are their country's national HMI system or other instances but they typically don't have enough rights to do much other than maybe enter data and create some reports this is not exploiting the full possibility or potential of DHIS2 and they would like to know how to get DHIS2 on their machine to use it more in depth and most often what they are asking is to get the same implementation that has been either presented in the presentation or used during the workshop so if I can give make use excel as an example they would like or they would not like to have an empty excel sheet they would like to have the pivot tables with the fancy charts and filters and slicers that they can use like they can plug in their own data and play around with a bit and experience different possibilities unfortunately DHIS2 is not readily installable and configurable by most of these health professionals so how do we provide them access well solution is D2 Docker which is a command line tool that allows you to manage a DHIS2 instance so you can start from scratch or you can start from an existing installation located in a registry in the cloud or Docker Hub so you can work locally on this instance on your desktop or laptop you can save your work you can do versioning you can publish your work and you can share it all this without being too complicated using a set of simplified commands so what is the Docker mentioned in D2 Docker well Docker is a technology that already exists it is like a virtual machine it is a tool that has been designed to create, deploy and run application using containers these containers are packaging of all the parts required for an application to run so what this means concerning DHIS2 is that within this image or container you have DHIS2 it is database, it is work server and so on so everything is encapsulated inside this image this image is uploaded to the cloud where it lives you can then download it and run DHIS2 on your machine of course this requires the Docker application but it is no big deal to install you can see here in the middle of this image there is a Docker Hub this is where the images or containers are published the different versions are located there and you can make them accessible to your clients the Docker Hub is an example of a registry there are many others which are a GitHub but for dockers one of the simplest pictures I have seen of Docker without getting into much detail I found on the web is this one which shows the Docker client the host which is normally sitting on your machine and interacting with the registry which is in the cloud where the different versions of DHIS2 that we place will be hosted or different applications or different implementation will be hosted and then go and perform certain operations like load and that is where they publish and that is where they will save their information too so how does D2 Docker work D2 Docker is a wrapper that wraps over the Docker and Docker compose and this manages the DHIS2 server instances from the command line Docker2 D2 Docker is implemented in Python 3 no knowledge of Docker containers are required D2 Docker uses 4 containers 2 custom images the main thing to know from here is that one is the core and one is the data now the core contains what the applications and the data is the data that you can load from a backup of an existing DHIS2 server or you can start from scratch it only takes a few minutes to load you can store images and track changes using any Docker registry as you saw in the previous slide Docker Hub was the registry that we use and with this command line which has been simplified you don't need to know how to build your own containers or how to configure this Docker you just pull from existing images and that will be your starting point so how do we use it we use it to create specialized packages in specific versions of DHIS2 for countries so this can include our AP modules, ENTO modules and in these specific modules that we create an example is we would create say install a version of DHIS2 say 2.30 for example we will install then the ENTO module and then we'll make it specifically the installation for Mozambique and this also as a second example could be version 3.4 AP model for Ghana that our naming convention has the ISO 2.0 code for each country now from the generic package we tailor it to a specific location we then modify this package save and upload it to the Docker Hub so this allows us to readily develop and modify the DHIS2 instances for specific situations, make revisions and updates and to readily share it with our clients which are countries, national military control programs and so on. So it's very useful for development, extremely useful for testing and training and an example that we use it recently for migrating our data from one version of DHIS2 to another and also we use it to change server and also very helpful for developing apps all running on your local machine. Here's an example typically what I just mentioned a specific version has been DHIS2 has been installed we load the Intel module into this virtual machine, we make our changes, we save, publish to the Docker Hub where we have our repository, a WTO repository and there we would assign access to a client in Mozambique and there they can pull the image from the Docker Hub and readily have this version that we designed up and running on their machine locally the simplified set of commands this is not a complete list but it's basically what you need. I mentioned earlier that you have two basic images.core which represents the actual DHIS2 includes the version the first command is to create that core you can create it from an existing war file on your local machine or it can be downloaded from the web directly and then the data itself the second command you create the data now to create data command can actually read in an SQL backup of an existing server that you already have this could be your HMIS, this could be something that we share or someone else shares with you can be loaded as the data side of that HH2 installation both of these combined will be your particular DHIS2 instance then you can start the image using the start command which pulls a new image from the server the command line indicates the name of the file you want to or the version you want to pull and if it's already on your machine then it would load up that version and if it's not there it will look into the hub and pull the latest version that's available then there's the stop command which stops the server to commit and push now commit will then save the changes you've made and still maintain it on your local machine push will then publish it to the server so back to the cloud available for others to load and to pull down to their machines and work off you can also export your work to a file you can import from a file and you can run SQL commands against the database running inside the application so key points no knowledge about these docker containers needed this has been preconfigured for you by us, by ICT by whoever has knowledge of this the end user has no need to know anything about these docker containers no need to know how about how this thing is built it's been done for you the simplified set of commands that allow you to load, download, modify, save and share your work and it makes it more accessible to health professionals and gives them the access to create, improve and innovate thank you thank you so much Ryan so we are now at the end of our session so thank you to each of you who have attended this very interesting session it's now the end of the day so what I would like to do because normally the use case is also the opportunity for each of us to meet this year it's not possible to do it but however I would like to invite each of you to switch on your camera so that we can say goodbye in image yes that's good so thank you to everyone and hopefully we will meet again we will see each other tomorrow at the end of the annual conference any questions do not hesitate to write them on the COP so have a nice evening or nice morning or day depending on where you are and we talk soon thank you thank you bye