Webinar on ICT in the Ebola Response



meeting the needs of communities and the needs on the ground we also had a very large digital media and communications component which I will talk about later in the presentation another thing about the e cap program that was unique during the Ebola response was that we did it operate in all 15 counties of Liberia which if you know Liberia is no small feat to the next slide if you can go to the next slide please alright so be a bit of a delay okay so to understand the different technology we used I think it's important that people can understand a way that reverberated because we had different pieces of technology targeting different people within the program so most akkor itself along with psi provided training to around 900 partner staff members these were the people working on the ground doing the work we provide training in terms of the messaging behavior change M&E system and also we provided a LG smartphone to each of these nine hundred individuals this phone was used for data collection and also for engaging with the other technological aspects of the program and these phones were donated by the pool Allen foundation through net hope now these partners staff members then went on to train fifteen thousand individuals in communities to do the messaging and it was the staff members that were supervising these community members in the messaging so it was a cascade system and we the different technology targeted different individuals within the cascade so if you go to the next slide so with India we use quite a bit different a number of different types of Technology with the idea being that by using all these different types together they would create quite a strong platform so first of all we had an EK Facebook page this may sound not very exciting but given that we were such a big network of 77 NGOs we needed to build some momentum around the network and the Facebook page particularly in Liberia where people are quite active on Facebook was a great way of getting individuals within our partners to engage in what they were doing in to share information additionally we had a website which shared information on the research we are done updates in the epidemic the data and other information that partners needed to know including the different things partners for doing this website was open to the public with the idea being it would improve coordination on the website we could partners than anyone could see exactly where we were working what communities what districts and the type of work we were doing within each of those communities we also distributed audio and video content on each of the mobile phones that were distributed across the country to the partner staff members these audio and video content was intended to be used to refresh the training materials of those managing the social mobilizes and also to engage difficult individuals within communities on challenging topics so for example for something like stigma if someone's working sort of in an area where there hasn't been any Ebola survivors you could use the video content to train people to really understand why we're talking about the topic and relate to it in a better way we also established an SMS flat for targeting those community level people at the bottom level of the castes who did not have smartphones this was done using the rapid pro system under UNICEF in addition to this we had whatsapp groups for all the individuals with smartphones and these groups work for peer-to-peer learning and sharing stories and experiences and finally we also had a large-scale digital data collection component where we were collecting 15,000 knowledge attitude practice surveys every month from across the country and sharing those results with anyone and everyone who is interested through the website now the bay name of sort of this discussion today is not just to talk about what we use but to talk about what actually worked and I would like to be realistic here and let you know that not all of these different technologies worked so we can go to the next slide I like to sort of work out like inform you on what we think worked well in and and won't really didn't work well and as was mentioned in the introduction obviously a big issue was connectivity for a lot of the internet based systems you had people in a southeast of Liberia couldn't really connect and it really didn't work our website was not accessible to them and they couldn't engage with our SMS system it failed there's no I can't say it in any other way we had a lot of issues with the network providers here in Liberia we couldn't get contracts we couldn't get the system up and running and so after one month we disbanded it with some of the phone based system so the audio and video content the data collection the whatsapp we had a lot of challenges with troubleshooting particularly with people in a southeast over time we developed ways to respond to this but at the beginning for the first few months we really struggled so when a phone went down it was down and there was no way of fixing it um the other big challenge for this was it was a major time and stark investment from the Mercy Corps level and I think that is something to think about in future epidemics whether this is actually worth the time and staff level of investment but despite talking about the negatives there were still many strengths so things like the website and the whatsapp group were fantastic was for creating PTP learning networks which really enhance the social mobilization that was occurring at the community level and I think for the Ebola epidemic here and I grew this was what we really needed with people young people learning from other young people on how to best inform their communities and that was something you couldn't have established without using technology it'll also really encourage staff um to to work a bit harder so um so for example when you have people that have to travel three or four and three or four days to get to a very remote community it's very a lot of them are not eager to do that but by knowing that they could share the photos of what they'd done talk about what they've done on the walk separate people were willing to go the extra mile to go to those far far remote communities because they knew that they had a big network around them that were supporting to bring them and encouraging them to do the work so fast that made a big difference additionally another major strength was our ability to share information from a mostly calm down and from partners up particularly it goes in the southeast where is very hard to communicate some toughness phone networks don't even work women the incident sister was working the website and the phones were a fantastic way to get information for them fast and to respond real-time this was particularly important with issues of community entry where someone could message on whatsapp and say we're having an issue and we could send a staff member there straight away or we could call which was more reliable than some of the text messages where you didn't know if they've gone through or not so um I think this is it gives you an idea of what had you know what worked well and where we had challenges um and I don't want to go over time so if you can go to the next slide I'll go through the main lessons learned oh I know this is emergencies so I talked about a lot of the stuff involves planning and time so it's not always possible but I'm going to save them anyway um the one thing I think the main thing we learn is that having the right team to set up their systems is really important um if you don't have the right people people that know I t that know how to troubleshoot that know how to you know make citizens more flexible from the back end there's no way it will work and so investing in the right team at the beginning is really important the other thing is investing time in training if you want people to engage in technology they need to know how to use it but something like whatsapp that was very easy to use once we had the system set up but as soon as there was any floor in the system there was no way arrêt so investing time in training people in how to fix things planning obviously is not easy to do in an emergency but there are certain things you can do um you know in a week's time you can plan a bit more than what we did we should have done a bit more research about making plan B is for SMS system that we didn't do um we could have gone back to a very low-tech SMS system but because we hadn't even thought of it we kept pushing for this complex system and it never really eventuated so yes all in all the big lessons I think we learn from our program is you know making sure you think through what you use and make sure the technology is appropriate for the country that you're working in um but if you do that I think it could have a fantastic impact on something like the Ebola epidemic I know that our use of technology really made our program much better much more enhanced and I think the impact was actually felt at the community level but again it requires a lot of time and investment I think that's the the main things um I've seen in the question but I'll someone says that I said that the SMS system was difficult um I think this is something that I'm not sure is an issue in all countries but in Liberia with the network providers here lone star in so calm we had a lot of trouble establishing basically um they made it they read to the system but then they didn't actually implement it we also said had some issues with the record pro system that UNICEF hasn't set up um but I think since now I think UNICEF has worked on fixing that um I know the other partners such as intra health I think are using method pro at the moment and it works fine but during the emergency period just it didn't work um I have it there's another question about the facebook page um the facebook page was like it's what I guess it goes back to what i was saying we're all this stuff on its own I don't think is that effective but as a package it worked quite well so a lot of the social mobilizes that we worked with were young people we were very active on Facebook so you use the Facebook page to share photos and videos and then to sort of talk about the successes of the program um and we use it as a way we had competitions like you know the best social mobilizer of the week and we sent out prizes and so it was a way to sort of motivate people and and build server momentum around the network and the work that was doing I don't think the facebook page on its own was would have been effective but I think in combination with the word separate where we could advertise it and share it if it worked well um the facebook page is actually still active because we have a second program called ecap to our with a lot of the same injuries and so we just continuing to use the same facebook page another question which is about combining data from the multiple sources to get a good overview so a lot of this basically requires a strong team at the back end so that means the people that are managing the what's up and the Facebook are sharing information with me who is running managing the monitoring and evaluation system who's also sharing all this information with the team that's working most closely with the partners um I think we didn't do this as well as we could have but I really think it comes down to sort of meetings and communication within the backend team and I know that for this the new program we're working on we're using some of the same systems and we've actually improved but it really comes down to sound really basic bit meetings team meetings sharing the information and then trying to make the website as a hub for all the information so a lot of the comments coming from the whatsapp was loaded up onto the website the data is loaded up to the website so trying to link them all together but i think it is it's a good point and it is a challenge but i think that's for me that's everything unless anyone has any other questions a girl since she is a question okay i'm seeing a dells question which is that the work required a major time and staff investment what i mean by this is so our program was quite an expert heavy um during the outbreak so we had a special guy running the digital digital system we had someone doing mass media and of the expert I was on the UM M&E side of things and I was around those dreamy obala I'll break you worked non-stop um not all programs have the opportunity to invest is heavily in experts and there's a lot of capacity at the national level to run some of these systems the SMS system could be run and someone to watch out groups could be random but the initial setup process require the right team and when we had challenges but we needed to adapt or troubleshoot you really need to get to have people to understood the technology so that they could attack them um uh so yeah that's what I mean by it takes you have to invest the right system the other thing is like for me from my end without large data collection component it was so big that we didn't really have enough time to probably use the data and engage with it as much as we did actually collecting it which is a big weakness of the system so um you just have to be willing to to realize that these systems are not easy things that people can run and then it will take a lot of time to make them run properly I guess my view is if you try and do something fifty percent you may as well not do it at all if you have to really try very hard to make a robust system because a lot of this tech stuff is it for it feels very easily so unless you're willing to constantly fix the challenges come up I'm it won't work well I'm seeing more comments um with this staff training I said that we did do training but it wasn't enough so we we did a lot of training with the NGO staff but again we were working with about 1000 South we were training across the whole country so for us to do a training with stuff it took a lot of time and so I think we did we did three different training rounds over the period and even that it wasn't enough so it's all that stuff trainings not worth it but it's quite time intensive and you need to invest a lot of time in it and Sophia just a minute or two left in this in this section and then we'll head on to the next presentation yeah I'm seen from ma that the plan B fellow connectivity was SMS um definitely that was the plan B but the plan B failed sadly um I think we've set up an SMS system now for the new program and we know better how to do it um I think our failure was maybe too engaged more with the current systems in Liberia see what people were using um we just went with the rapid process in which at the time wasn't functioning well but like I said I think right now it's functioning quite well just our it wasn't that positive um regards to what's up in the rural areas what I didn't work in the rural areas but some people went to the town so if everyone just submit data on the mail system they had to come to town to submit the data and so often they had internet then and they engaged in what's up then so even in the rural areas where there wasn't high levels of connectivity it was an engagement with whatsapp the big problem in the rural areas was that when what's up updated which what cept us because it's a application it doesn't update very well in library exit I don't have an agreement with Liberia so you had to manually update it and we couldn't accept when people were in very remote areas we didn't have enough staff to be able to go fix and troubleshoot all these phones um and I said with Elizabeth I um I guess you're right that it is an important investment of staff time and I do think like I said I think our system was worth it but my worry is that in future epidemics people will you know again try and use all these tech components that may be John allocate star for resources to them and so then the systems don't work so I think if you wanted to make it less stuff intensity of you have to use technology that locals stop or national staff can can work on end can adapt and that national like that liberians can troubleshoot so for example with whatsapp people couldn't work out how to troubleshoot when they needed to update so there's also other ways of working around it for example we made videos and PDFs of how to fix problems and sent them out onto all the tablets on the full of funds and that was a way of reducing time on stop the time burden but even then it's still quite a heavy burden thanks a lot I really yeah thanks thanks thanks very much I'm sure everyone is clapping I mute and it's great to hear both the positive effects as well as a good dose of realism about just how hard it is to make you know implement to maintain these systems because really yeah really resonate Scotland with us as well so i think we'll ship to the presentation yep so we're going to transition over to jean martine flower street station from World Food Programme so John Martine leads the mobile vulnerability analysis and mapping project his recent work focuses on using new technology to track food insecurity areas that are too dangerous for face-to-face data collection John Martinez worked on using text message robo calls and voice surveys to reach out to crisis affected communities in Africa and the Middle East is also worked on using cell phone metadata to estimate population movements at times of crises just one second while we switch over to that second power point and then we'll I'm going to you as well you everyone can hear me yeah we can get great great so decision my 10 i'm coming in from Rome good morning to you all and I'm going to have a few minutes to tell you about the mobile vulnerability analysis mapping project so the vulnerability analysis and mapping is about reaching out to people using other phones that they already have through SMS interactive voice response in order to inform on WPS food security responses if I could go to the next slide please and show you a little bit more about what part of wfb I work for um right so be I work for the wp food security analysis unit I the what you see here is the the way your website looks we have a network of 150 analysts worldwide whose job it is to track food insecurity we do things like baseline food security assessments with national governments we do monitoring systems we do market analysis we also have some work on nutrition we also have a commitment to providing public goods and commitment to open data you can find a lot of data sets and information on the website you see here so what I want to do today is give you an overview of the mobile van project and how we use it for boa so van was actually next slide please the VAM Department food security analysis Department of deputy was created in 1994 quite some time ago we're actually proud to work on new technologies when may when they come out really new things like a geospatial we were into the doing that in the 1980s and then when PDS came out of course they sounded really antiquated now that we were among the first to use them in work we're proud of that what I want to tell you a little bit more about today is how we're using on mobile phones for food security needs assessments so the environment vols contacting people through their mobile phones the future tunes that they have so no smartphones no data plans the basic forms and we which actually are the ones that people use in places like Liberia Guinea or Sierra Leone and we contact them in one of three ways so either interactive voice response we've used the instance of her voice free and open source software to run our interactive voice response codes we've had operators call people up and we've used a lot of SMS surveys and SMS ended up being remain tool for a free bowl on the idea here still reach out to a large sample of the population and ask food security questions through text messages aggregate that data in a dag regression back end I'll tell you a bit more about that that information is used in order to inform the humanitarian decision-making process so in terms of resource allocation or advocacy and the advantage that we have with flow within van is essentially threefold it's much cheaper than doing face-to-face surveys it's faster than other types of surveys in terms of survey turnaround time and it's also safer for staff and keeper we usually get an SMS a questionnaire done between three and five dollars per complete faster because an SMS round for 1000 to 2000 households will take us one to two weeks instead of four to six weeks when we send out the enumerators and it's safer because in a case like Ebola you will remember that it was very difficult to send people as for the field and we were able to access all the Sierra Leone Liberia of course some stuff one covers permitting but it was actually quite widespread and Isla guinea thanks to the other phone survey so those are the advantages of that other tool next slide please and what you'll see on the next slide is in factor or strategy in a Guinea Liberia Sierra Leone were a place where you first rolled out to after having a very limited pilot in the arts to in Somalia and the limited pilots allowed us to produce enough learning for us to be able to deploy very quickly when the emergency was declared on verbal look the next slide please the learnings we we drew from our experience with the bowl up with the automated SMS service you'll see them there so of course they're safer I told you they're safer there quicker and they're cheaper than the traditional face-to-face surveys but I'd like to focus on the issues that came up so of course there are their biases because of literacy and ownership cell phone coverage and access to cell phones is incomplete and Guinea Sierra Leone and Liberia I think we have about between 40 and 60 lines per 100 people so they are there they're resistive you towards Ruben areas towards better off households towards the young men so it's Tobias that that's there that we have to deal with one of the challenges was dealing with the volume and data we're getting thousands of surveys every month and that led us very quickly back to work on on automating the processing of the data we also realize that SMS was a not suited for for laundering more complex questions we found out that with short and simple questions were able to get good answers so strong ethical questions and require short and simple answers and that that's what ended up working for us next slide please the next slide shows you how in what we actually did it for three countries I think we've got 12 rounds of data how did that bowl affected countries we've began need a collection in a matter of weeks after the emergency was declared at least four wsp it went through SMS we had that some IV are there SMS works better than I PR by the way and the indicators we collected by SMS included food security coping so the strategies include did you eat last preference rooms or digest to borrow money or did you have to borrow food so this is a standard food security indicator this huge worldwide bed market prices and you also ask people what they thought of food security in their community that allowed us to produce the map you see here were able to track wage rates as you see on the graph on this side also able to do a word clouds and understand what people are talking about we're very and this was very new to us we realize that people are very happy to engage with us bye bye text messaging Licata between sixty and seventy percent of respondents would text something and they say things like oh the things it's getting better because the harvest is coming or the corn and a lot down in preventing us from selling or food that's that's how it went the questions are asked to the general population these sort of respondents that are randomly selected in each country and we were able to go to do these word clouds but also do a natural language processing and employed language natural language processing to assess the tone the statement that were being made and it was interesting to see that the tone of the conversation the rated plus 1 2-1 according to an algorithm was actually correlating quite well with the food security the questions we were able to triangulate information that way next slide please so I'm not moving to lessons learned and I mean we thought it worked pretty well to get reliable frequent data reliable in terms of the trends we we don't want to pretend that this was representative but we've got good trend data from from through thanks to SMS SMS Pro better than I BR you get cleaner data or SMS we think because people are able to see it applies that that they have before sending them in when there's ivr you don't have that much margin for error the design of the questions is important we had a lot of trouble with your questions initially but I think after a few rounds we were getting 95% clean data because we understood how to ask the questions engagement I told you about the open question being a successful way to engage with people and finally perhaps on creativity it was a restaurant to ask people what kind of toilet do you have and that goes back to the bias issues people are saying well you're only reaching the richer part of the population but in fact we asked people what kind of toilet do you use and that allowed us this woman on the poorest households that replied that they had either a dirt toilet or no toilet at all and were able to monitor trends through time and it was interesting to see much worse indicators and very different trends but that part of the population and were therefore able to produce information that spoke to people's needs so finally a next slide please this is not something that if you did on its own we had a lot of people backing us up on methodology we had the nielsen company helping us out Tulane University produced an evaluation of what we did it's posted online if you want check it out the temp that we use came from instead for her voice already talked about that the surveys were done by G opal the SMS actually worked quite well in librarians early on in our experience we got very quick because the volumes we needed in the data we needed in time thanks to GOP and we had contributions from the humanitarian ovation fund google.org cisco less ESO and you say to make all this happen final slide shows you links you might want to check out the final slide shows you that we actually wrote an article that was published in the overseas development institute humanitarian exchange on or experience with SMS and early warning for food security and Ebola there are few other articles you can check out there are generally interested in the project River blood that's available within van mourik enter twitter account at mobile jam thank you for listening over um so we have that question about a bullet patients to their families what we did was ask questions to the general population so these were randomly selected respondents and they would get the survey every month we had a monthly frequency and we did everything we could to maintain a panel in order to deliver trend data as I was explaining about the quality of the data that we had wouldn't really allow us to have quantitative estimates we tried to track the same people that get an idea of the trends and the trends were actually getting a lot better this time so about ivo and white didn't work out didn't work out so well and this was this is drafts counterintuitive because you think we'll ever see it's better to go with idea but when you ask people to estimate something to provide a number like the price of a bag of rice you'll get this you'll get mistakes data entry mistakes that people can't fix with ib are people who might feel rushed by the pace of the recording of people have to finish with the pound sign on their phone so that their little things with IVR that aren't very user-friendly we found that ivr probably rushed people into replying in the questionnaire SMS people can take it when they want with IV or by the way you get an IV or phone call you have to write or else you you can't stop Midway SMS you can take the question when you want you can stop replying you can take though you can go home and reply the privacy of your own home you can check your answers before sending them in and and it seems to us that the data quality was much superior by SMS than now denied er perhaps for that user user friendliness factor and this is Neela am I right that god of your questions were what are open-ended or open text like you didn't sort of favor questions even at via SMS was like respond one if you're this respond to if you're that were like the response is is always very structured was it was it more open-ended responses what we were able to use open-ended responses we always had so the questionnaire was very short we had about ten questions in the question yes ms questionnaire so it would come as a chain of messages back and forth the last one was an open question and in SMS it works great because it was text and we would read the text messages here in Rome on days after they were sent to us and processed them in the way I described I'd be all would have been more challenging because then you have to deal with transcribing the messages from from my VR to text and then not doing the processing I was mentioning so as soon as did have advantages from that point of view don't say those people at me oh okay so there's a question on feedback loops so the the data was shared on the wfp website it's available on an open-access basis and that was actually one of the successes for us is that under organizations user data and relied on our data for their reports of their analysis of the imola situation I can think of using that LSC and Sierra Leone has this program an economics program that used her data as well the European Union is well relied on on the data that we were collecting for their own analysis I think when we look at the website of course it's very much the UN agencies the government the NGOs some universities came and downloaded the data that we were collecting in this way so was posted online every month when we were done we're also posted the information on the humanitarian development on data exchange HDX I'm not sure everyone's familiar with that but it's up a platform run by ocha or sister agency and we entered into a partnership with pho job to make sure that the data we were collecting through bullet would be shared with the entire humanitarian community through this portal HDX humanitarian data exchange and with the people from whom the data was collected that's more complicated we're actually working on that in the project not in the imola countries would into your compliments avolio where the information we collect is put onto a server and people can call in to listen to the strategic information haven't been learning about that on the blog and I'd encourage you to check that out on the little over one minute left set there's one more connected maybe one more question so from Emily Nicholson wood platform do you use for developing and sending sms we were through mobile accord they have a a service called yopo we have tested in the Democratic Republic of Congo in the determine that it worked out well for the Kobe questions that I described so we thought let's roll this out in a bowl and it ended up working out pretty well so it's not just the platform is that the questions we introduced for below some of them are different from the pilot experience we had and getting those questions right actually took a lot of iteration the prices on the food prices were a challenge because we have to get the unit of measurement right we thought that everyone would know in a bag of rice costs and I mean I've worked in Liberia and Sierra Leone that resumption was wrong you have to ask for the cup of rice and when we started asking for the copies of the micro retail price we got very good data I asked him for a prices of a bag were of rice or palm oil or anything else was I was complicated so it was really getting the questions formulated in a way that was intelligible to the respondent that really took the most effort for us of course they open ended question that's something we hadn't tested before the ball was the opportunity for us to learn quite a lot so the lessons we picked up in a bow we're now applying in places like Yemen most recently Malawi so that's that's how our experience with the bowl is also helping us get better information in other emergency context as well over to you great thanks so much i think that's just the end of this thought and yeah it's really interesting to see how the different systems kind of had successes you know with SMS without SMS and that there are places where works well places where it doesn't it's on to the next presentation by piracy thank you so expectation is by Paul at IRC Paul Amendola is a technical advisor for health information systems at the International Rescue Committee and is based in New York he has designed and implemented data collection projects and epidemiology studies and developing sub-saharan African and Asian countries focusing on increasing data quality and analysis through data training reviewing and advising on information systems and database creation at Paul I'm much a meteorite now pull that up the key detail go okay is that okay yeah that's great okay great okay if we could go to the first slide please my name is Paul Amendola I'm the technical advisor for health information systems for IRC and slides about sort of the context what IRC was doing during the Ebola response we already took during the outbreak in the peon in early 2014 we were already operating programs in Sierra Leone Liberia Cote d'Ivoire and Molly during the outbreak the health response had a number of different activities our main activities though were focused on four of the hardest hit counties in Sierra Leone and Liberia we were providing a support to three Ebola holding units in Sierra Leone and support to one Hospital in Liberia in addition we were also equipping facilities for Ebola response and measuring the readiness and next slide please so would focus on three ICT interventions they might seem a little unrelated they happened in different areas but our lessons learned for them were were similar and I've been applied to other programs the first project was an electronic medical record system for a treatment unit and then a database and record system for a holding unit and then our facility readiness assessment and next slide please so IRA see was I was planning to open a bola treatment unit in Liberia our challenge there was to design a system to be able to move data between infected areas and non-infected areas paper wasn't working there were a couple different ideas of that other groups were using of I using whiteboards just shouting over areas and we're we had a we had about we had a couple of weeks we had about six weeks from before we had a treatment unit being deployed so we were trying to design and deploy a tablet based EMR in the project we were partnering with vecna using a clinic pack and the EMR was looking at patient-level data real-time data and implementing decision support sorry next slide please I'm sorry so they add the EMR sorry back to the the first day of our site we go back on I'm sorry so we never this was actually never deployed the training was done the software was developed the hardware was procured and set up but just sort of set up in time for the for the outbreak to be winding down so the the system was never actually used and next slide please okay in three Ebola isolation units in Sierra Leone we were using an Excel database to track patient triage to track patient outcomes and to to report to the Ministry of Health ideally the staff was spending a lot of time with their paper forms there was low connectivity so we couldn't have anything that was going to rely on internet connection but they they had access to to a laptop that could be used in the triage area it was a it was a line listing of patient records and had automatic completion of all the moh forms and a kept track of the number of patients in the ward all of their cumulative data all of the indicators they needed to track both for program monitoring and for donor reports the next slide please and then in Sierra Leone we were ahead of a consortium of 13 NGOs that were implementing facility readiness assessments that were done at every facility in Sierra Leone at least once a month we went through a couple different ideas but we were able to deploy a couple of hundred mobile phones with a with a short questionnaire at the time we were using odk that fed into a dhis instance for partner tracking each partner had access to the to the mobile data and to the dhis instance for reporting and throughout all three projects these are three major lessons learned so a user centered design for each project I can't sort of stress this one enough for the for the EMR in Liberia what was the first iterations we designed at headquarters with clinic staff when it was actually deployed to the field everything needed to be changed it costed a lot of time it cost a lot of money although the the basic idea of the patient flow through the facility was the same house clinicians saw the benefit of the system was different than the headquarter staff had seen for the excel database it was we ended up using Excel for the reason that we knew that the data was going to change a lot we needed a system that would be familiar enough with there with the end users in the in the field to be able to adapt as indicators changed as their treatments changed and didn't need a lot of back and technical support for for the sierra leone facility assessment a paper system wasn't feasible again the user centered design how the questionnaire was designed and then tested with the partners and then a introducing that earlier and testing with the end users earlier would have would have saved a lot of time it was designed the first stages were designed by IRC but having more of a consortium input would have been beneficial second lesson learned the standard ICT assessment we've seen the same assessment components that need to be checked before implementing each project sort of as sophie was saying staffing having the right staff profile in place for the project and having enough staff members for troubleshooting and for backstopping is essential that's sort of our first a checklist item before we deploy projects and that's that's definitely a lesson learned of from this experience other components for assessment we look for is if we're if we're entry seeing a new technology or if the technology already exists sort of with the other two presentations said the same thing if we're if if the existing technology is just being leveraged that there's a better chance of success making sure the data is going to be able to be used and accessed looking at what the mobile coverage is and the power supply situation is will the phones be able to be charged will the tablet to be able to be charged looking at logistics for what is the process to repair a mobile device what is the process to replace mobile device and what are your data protection and evaluation needs just just the just the fact just the need to make sure those assessments are done before any sort of ICT is introduced has been again one of our one of our biggest lessons learned but moving forward having in emergencies like Ebola having readily deployable systems that we know have been validated that we know have been tested in certain situations would be really beneficial even if the even if the content or the questionnaire has changed having reliable ICT systems we can deploy at a shorter notice it's something I think we're all trying to to move forward with great pony about three minutes left if there if there are some questions and I don't if you were ending but I had I had one on your last point which is I'm just like if those clinic pack systems are are they now like ready to boil you've done all the work like the next time could they just you know be pretty quick to deploy or build into user centered design the new country a new area where you kind of have to start from scratch and have the Saint you know with a big challenge of getting it out fast enough I think the the lessons learned from how we deploy would change but it they wouldn't be a they would be had they would have to be too much adaption that they were so evil a specific in terms of even the clinical decision support that was programmed but even even having a baseline set of modules even on Lake Winnipeg that could just be somewhat you know tweaked for four different context would be really helpful the next time do you by chance when I hazard a time lag i kept it was if you were new crisis was with starting today and you were just you know just starting to go to it how many weeks or months till you know it's a reasonable amount of time for for deployment I don't mean to put on the spot too much but you have a have a good i guess if if the staffing was there um we're we're fairly used to whoa the steps that would need to be done at maybe four or five weeks I think um for something at like a I treatment unit great great well it seems and that seems like a big big advanced compared to probably what it took this time that's a that's a significant improvement public you have a question of what system of data collection was used in the ETU the ETU it never actually was operable but it was it was using the vector software on the on the clinic pack but then did you use odk as well to collect data from the photos from not you to use that was from clinics right that was that was from clinics that was from the health facilities and I think our decision to use non open source software was timing we had about six weeks before we were ready to deploy and I normally that as the the open source should be a sticking point but there were so many there was there was such a time crunch that it seemed like our best option oh yeah we do we have a ICT assessment checklist that we can share but I think that puts us at about the end of the of the time slot thanks a lot Paul and great to add a lot to see a lot of common themes as well as some new perspectives it's like each each each presentation is doing that so we are on to the next how great thank you Paul so our next presenters immunol who is a research scientist at IBM Research Africa she studied the University of Cambridge and has PhDs from the University of Cambridge and chemical engineering and biotechnology she previously developed a bio ethanol combined lantern and still designed for use in rural India Nina it was just a moment and we will unmute you great can you hear me we can thank you excellent great so yes I work for IBM Research Africa we're fairly new the lab in two years ago we have a lab in Nairobi one in Joburg and I'll be one in West Africa coming up soon and so we started working actually with open government initiative following the Africa summit in fall of last year and open government initiative also it could you move to the next slide I'm edgy I asked us if we could help them because one of the things they wanted to know was to try and understand how well the Ebola response was going from the point of view of system so agi's mandate is to provide open channels for communication between policymakers and citizens so we have to set up the 644 hotline and which allowed individuals to call or text in the way it was set up was that the health messages were going out over radio and then there was a message about 644 afterwards and people could call or text in and one of the key things was that we didn't constrain the type of response so there was no survey based responsiveness this is just people calling in about anything they wanted to so the calls these were then transcribed by a GI call center volunteers and then the SMS is came in through an SMS gateway once we got those that we then worked on writing some code for message categorization which I'll talk about more in a second and then the results were reported to AGI and then they could go back to the so edgy I sits under the president's office in Syrians they would then go back through the president's office to government to report on what they were finding so if you could just move to the next slide example of the results that we got during the first September lockdown so we we categories based on what what was coming in and also based on the Regents these are just the cold data and we got people calling and texting in about all sorts of things those sorts of things that we were expecting things were directly related to a bowler dead body sick person and things like no soap which were obviously messages that were conveyed over public health messaging on radio but then we also got a lot of things that we weren't necessarily expecting stuff about lack of utilities compliments about how the response was going information about volunteers and and then this doesn't have the other category on it but then we also got a whole load of stuff that was just completely not related to bowler people just calling in about something completely unrelated and so this was interesting because actually over time we started to see these categories change so this is the first one we see a lot of no sensitization as time change you start to see the idea of non clients becoming more ubiquitous people understood what the response was supposed to be and how and they would report when somebody was not complying so things like my neighbor's child is sick and they haven't called 117 or the burial team came but then they were asking for a bribe so this sort of stuff started to it happen we started to get less of the know sensitization and I was interesting as you can see from the numbers of call numbers volume was very small the SMS volume was even smaller this is for a lot of reasons but maybe if we could move to the next slide so we suspect this is for a lot of reasons for the SMS so low literacy yay if you look at the color shadings on the chart for there was a lot of purple which was Western so Western urban and not set as much from the other parts of the country so we're Daphne game sampling bias from that and it was hard to know how much oh gee I was being accepted and one of the key problems first was that none of the responses could be verified and therefore it was hard to provide follow-up actions and you know if somebody says that there's a corrupt ferralti what do you do about that how do you know if other people are reporting it from the same area and so the system was useful at the time it was built I just saw somebody asked how quickly it was built it was built very very quickly within a few weeks we got cooperation from airtel who set up the 117 short code and we got some donation in kind from a Kenyan startup who provided the SMS gateway and so it was really put together very very quickly and and so following that we learnt quite a few things one is that this message categorization problem is not trivial and we didn't have high enough volumes to do statistical analysis for that's really what it needs to be able to categorize these messages on the fly and actually the creation of categories is hard unless you include a human in the loop and so the other thing is also the response is not being verified so one of the things we started to do was to work with humanitarian data exchange usual much I mentioned earlier to try and see what open data sets were available to be able to try and triangulate between different data sets and find how likely a report was to be true and what this sort of made us completely pivot and because what we realized that actually there was a lot of data out there that was available that was being made at Oakland people like World Food Program UNICEF were making a lot of data open it was just hard to use because it all existed in Excel files and there was a mishmash of data some geo tags I'm not different field names for the same things that things like date or admin unit would often be named differently as a field they would often be formatted differently and so a lot it was actually hard to do the data triangulation that we had thought would be easy to begin with so now maybe if you could move to the next slide so one of the things we were doing was mapping out how people are collecting data during the outbreak and trying to see how communication was happening in terms of information systems so obviously people are communicating with the Ministry of Health but how was systems being linked together there is area I know and are we missing on this for some of the data going into dhis too it was actually when we went to the minister it was very hard to retrieve that data off dhis to and use it reuse it again and and so if you could just move to the next slide so given that the ministry is now working much more closely on their internal hrs systems the one link that was still hard for the folks we spoke to in Sierra Leone which was Missy girl in unit up with the fact that a lot of this data is still going to get lost anything that's survey-based and that won't go into dhis but that you want to combine with the dhis type data facility assessments indicators so if you could just next slide so this is the system that we're now trying to build so we've kept some of the citizen engagement in right now we're trying to decide if we do that on our own or use existing systems or most likely going to use existing systems of the social mobilization action consortium snaps Mac or new report and we're just chatting to partners now about who's interested in what they want to do and using the community mobilizes this is also to get around the huge sampling bias that we ended up with by date trying to do direct contact and then looking at how to build out and our natural language processing that's what NLP stands for in this diagram and to be able to categorize these messages more effectively start to look at sort of more machine learning approaches to message categorization and really start to build those out which is one of one of IBM's big research strengths as language processing and so we think if we can get hold of the data for that in a large enough quantity we should be able to do that and then the other thing that we're looking at is incorporating it with you open data from humanitarian data exchange and so one of the key things here is to try and show what is the value of making your data open so for this we're working specifically on a health facility accessibility index they're trying to see if we have health facility assessments can we now start to come up with a composite score for them to measure their accessibility or their readiness for an epidemic and so we're doing this using a methodology initially similar to wealth index and then looking at using slightly more refined techniques such as decision trees and some more complex regression models with decision trees as well to be able to then come up with a score for facilities and their catchment areas on their accessibility and then the last part of this is M surveillance so this will most likely be working with goal who have a I think it's odk your kobo toolbox surveillance tool for measles right now and the idea of combining these three so the citizen collected data the surveillance state runs humanitarian data exchange is to try and provide a more holistic picture of what's going on and then this this visualization dashboard can be used by both partners on the Ministry of Health's to get a quick overview one of the things that we're really trying to do with this dashboard is to make it easy to use so lots of people are using a decade khobor toolbox conquer all of these tools which given the way many of them are set up still currently terminate in an excel file and or if they terminate in a server being able to pull that into a back-end system that makes it easy to visualize so there are a few parts to this visualization dashboard one is data cleaning so we're doing a few things around so we're doing it but in terms of data cleaning we're doing a few things around being able to put in a sort of coding list for geo-tagging so for facilities for example you have the ministry's geotagged master list now and for all the old date sets to be able to go back and Geo tag for the ones that haven't been yet also looking at things like deduplication of files so we found for example one of the Syrian schools files have set one school entered 15 times it's slightly different spelling slightly different GPS coordinates their building tools that will automatically do that cleaning and and then the other thing is making it really easy to upload both the tabular data so CSV in Excel data but also shape files so map files and one of the things we really wanted to do was build some analytics off openstreetmap surround travel times to health facilities and making it almost a drag-and-drop so that you can just put so we have this set up right now for tabular data where you upload your tabular data file get integrate the code at the back finds the geocoding column and will then pull the geodata out and you and map it for you automatically when you click on a point it will then pull up all of the individual fields for that point that we're collected in a survey or in that we're in that file so the idea about this is really not to now collect new data but just to work with the data that's their help to triangulate it and make it easier to use and the specific use case work on which is what this side is now on right now is a from the community data to really try and understand the demand side barriers to accessing care so what are the the softer factors which often aren't included in quantitative models for health care accessibility looking at things like you know why are people not you know what is trust in the health system like do people think that they will get good service if they go these sorts of questions and then looking at the supply side look by looking at the facility assessment seeing if this facility is an accessibility index will work if it's representative of and the reality and then finally the surveillance so looking at demand for care and if we have these three we wanna see if the end user can start to identify what are the high-risk areas so we know where an outbreak is we also know how roughly how prepared it is from the accessibility index and then we also know how open that community is from the community engagement for accessing care so between these three can we now start to see areas in which you may want to focus your community engagement or maybe change the way in which you're targeting the messaging and next slide sorry to interrupt you have about three minutes left okay sure so this this is the last night so this is just this is still a work in progress to build this platform out this is the front end that we have right now the health accessibility index are cheap as of today will be updated with the new wealth index type accessibility index and so we're just plugging the bits together the project will rentals july of next year and it's now funded by USA dat bola grand challenges so that sort of me done i can take any questions for the last two minutes great and do you see a couple in there in the chat or should we read them out oh right yeah and yeah so in terms of how long did it take to set up the original a GI system that was the setting up the system the biggest delay in that was actually getting the telco cooperation from Airtel to get the shortcode setup and to give us access to that for the SMS gateway the actual setting up of it on our side was was quite quick we have quite a few projects that are doing data collection through SMS portals so we had a lot of that code ready to go and our timeline for launching the dashboard so where we were in freetown last week talking to partners so now we're just finalizing who were going to work with and we really want to do this as a user centered design process so we are going to be working with the partners to really understand their needs and it won't be a hard launch it will phase it out so we have parts of this now working and ready so we want to just work with users tester out early get feedback and keep building and as I said the project ends in July so I then we should have hopefully a fully functional platform and there's might have been obvious but could you elaborate on what the no sensitization comment was was that people calling in for more information yeah so that that was actually people complaining that they'd been kept in their houses and the sensitization campaign hadn't come to them so this was during the first lockdown I gotcha no thanks on can you say a word about how that how the mobilizers will work well they just could you elaborate on that a tiny bit I'm serious yeah sure so this is with um we're planning on doing this with with gold who have the social mobilizers and they are using community leaders and they now have a form which so they run on kobo and on paper i think and they they have their questionnaires and there they have that system up and running it's been running in sierra leone for a few months now and i can't give you too many details i'm basing what we're trying to do is just pull that data and understand what they want to do from it so they have a lot of free text data that they currently don't use but would like to use so that we're going to help them with that and really try and understand okay what can we do some of the sentiment type analysis the World Food Program are doing as well as trying to do some of the categorisation super I think there's just one last question in the chat what happens a lot before ends they were probably move on yes 644 is carrying on a gia thus that's actually owned by open government initiative in Sierra Leone we want to do the same thing with this platform we don't want it to be IBM internal we want it to be owned taken on and owned by partners who will be able to build it use it do whatever they want with it afterwards super well thanks a lot and it's great to see how the different kind of you know directions of input and output complement each other and I think I recall right the gene martin from roll through creek am also mentioned working with HDX oh it's nice to see that emerging is kind of a place to pool open data yeah great thanks a lot i think we're up to the last presentation by interrupt yeah so our final presenter is amanda from intra health and she's the international technical advisor for HR h and knowledge management and serves as the m0 global coordinator working with local and international stakeholders to support your platform in West Africa mend also represents intra health in the K for health project bringing a health workforce lens to the global knowledge management project there's one moment try and pull this slide up on mute to great hi can everyone hear me we can that's right great good day everyone so are you guys gonna do the slides or do you want me to do it because we aren't able to go in presentation mode we were going to do the slides um will be a little creative when we get to one section but exactly yeah all right let's get going so thanks everyone for sitting through all of our great presenters and I appreciate dimake for setting up this webinars we can go to the next slide I'm going to talk to you today about em hero em hero is a two-way communication formidable call and send SMS to the health workers here are uses data from iris and so that the Ministry of Health can send those SMS text messages to health workers own mobile phones iris is a health worker information system that was developed by entra health and it allows you to select who you want to send messages to so you can select by cod roe or location and really be specific about contacting your frontline health workers Emmure works on a talk in text a phone as well as smartphones the system allows health workers to message back so this helps the Ministry of Health to better communicate with their health workforce we've worked with UNICEF to create em hero and use stuff negotiated with the mobile network operators in countries where m hero has been deployed so there is no cost to the health worker when they send a text message back to the Ministry of Health the Ministry of Health determines what messages or use cases they want to send and to whom as I said they can select I would they want to send it the Ministry of Health develops all of the content of the messages and the workflows test them out before they send them and then they take the data analyze it just to back up a bit in hero was established to build on existing systems including dhis to iris rapid pro and con care as you know someone mentioned before during an emergency response it's really easier to build on what's already there as opposed to introducing something new M hero harnesses the power of interoperability of of all of these systems strengthening health information systems where m hero has been deployed that's a very high level overview of how the system works and what it does and I'm really trying to paint the picture that the Ministry of Health is very much in the driver's seat of implementing in here oh we can go to the next slide though inter health and UNICEF established m hero during the Ebola response and we wanted to help improve communication between the Ministry of Health and frontline health workers so this slide illustrates the types of questions that we go you know back and forth from the system to the health worker to the information about what was going on in the imola response could be quickly captured at the Ministry of Health model and they could respond you go to the next slide this is where we're going to have a little bit of fun oh it's all there okay I guess if you're not in presentation mode the way this normally works is it kind of highlights step by step but we can tell the story by looking at this slide how the technology works so M hero takes information from iris as I said which is the bottom left and this is where we can get health worker information such as the health worker Kandra or their locations I say the Ministry of Health wants to send a message to all health workers at one facility or all of the nurses in one County iris would be able to disaggregate data information we also are interoperable with dhis to to ensure that the facility facility data is linked up so the information that we're sending to all of those health workers at one facility that we pull from iris that facility is is that information is also a dhis two and it is matching so am hero goes through the health worker registry which many of you know about through the opening re initiative and then through this fun interoperability layer it all comes together and then goes to rapid pro or con care whichever SMS platform you're using so that this is where the messages and the content can be developed we've been using rapid pro so far in our implementation of an hero and the way the messages are developed it's really quite simple and we've been using very simple workflows there's nothing too complex so the you know the information trees that are built in there are pretty simple we always test them before they send them but when we do they go straight out to the health workers you can see our health worker images there and then the data is flowing in two ways so we have data flowing out to the health workers and then there's sending us information so it comes back to the Ministry of Health and we're working to ensure that the information that comes back through rapid probe back through that interoperability layer back through the health worker registry can be updated into iris and dhis to as needed ok next slide so I want to talk a little bit about what we've been doing an implement diem hero we piloted in hero in liberia in november two thousand four we got some funding from unicef from us aid through the kafir health project and also a little bit of money from Johnson & Johnson to scale up and guinea but our initial pilot witham hero was to look at the health workers and validate are those health workers there there was a lot of movement of health workers with the Ebola outbreak going to able a treatment units leaving service and even many unfortunately who died so we did a validation use case to map out who was at what facility and what facilities were still open the Ministry of Health quickly adopted this and was very supportive of this and it's since then M hero has continued to grow at the Ministry of Health in Liberia they've done lots of different use cases all determined by them includes nutrition use cases mental health use cases measure evaluation even used every keystroke they were supporting the minister Cullinan an ICT we're also planning to scale am hero in Guinea and Sierra Leone in early 2016 we've been doing a lot of work in Guinea in particular and setting up iris which they did not have prior to the Ebola outbreak but really wanted so this has become a high priority for them because they're realizing the importance of knowing who their health workers are and where they are so ensure health is working with a lot of local stakeholders to scale up their iris and build that capacity at the Ministry of Health to use iris we're also working in Sierra Leone to scale the iris working with a wh 0 currently iris in Sierra Leone is only in the western region so we'll be scaling it up nationally and along with all of our plans to scale iris we're talking about em hero implementation many of the ministries of health or stakeholders in the ministries of health and guinea and sierra leone want to do that health worker validation exercise like we did in liberia with M hero and there's also some talk of doing some maternal and health maternal and child health use cases in getting we have a lot of things that we're planning next and I want to say that we have received funding us a grand challenge which is affording us they built the ability to scale in Guinea Sierra Leone and and continue to go in Liberia and we're also doing a lot of other things to improve the back end of the technology through the grant college so we're improving the analytics platform we're working with stakeholders to improve data you know that the information they back it comes back in a way that they are able to analyze and use we're working on to interoperability with other systems just come care what else is going to be launching a virtual learning force next year so this is something that we're designing with wahoo as well as other ministries of Health who are using am hero to bring users together and a way to learn online what works best for them what would they like to see and do a little bit of online learning together and sharing of information um I don't have a slide here but I think it is worth mentioning a couple of the challenges that we have faced implementing them hero and a lot of them have already been discussed in this webinar so I'll just echo the ICT infrastructure particularly in the rural areas has been a challenge a lot of the SMS that are sent out may not make it or there may not be a response because of the connectivity in rural areas the infrastructure at many of the ministries of Health has also been a big challenge server capacity network capacity computers particularly in Guinea has been a really big challenge as well as the capacity of the staff to manage setting up use cases and an even steven he's an iris to export the data they need that's been a challenge the availability of those staff to is a challenge they have many things to do and particularly with the Ebola response there have been many on staff at the Ministry of Health and though we're using systems that many of them already know it's still an stuff an extra layer work um I would also say that the governance of data has been a problem or a challenge many ministries of Health don't have strategies or policies on data and data security and data governance and so that's something I know has been discussed with many partners but it is something to be aware of as a challenge if next step so I just want to conclude by highlighting some of the benefits of them heroes and I can't say enough about how the the fact that M hero is really driven by the ministries of Health is it's a benefit we talked about users that are designed and I think this is quintessential user centered design everything that we're doing with implementing an hero with the analytics even by having the virtual course that were planning well us how we should be improving the platform for them and how the platform should be for them um i haven't mentioned much about the health workers but um hero does make them feel more involved and motivated we certainly found that and i think someone else mentioned that earlier in one of their presentations about how engagement with health workers and community motivates them and finally facilitating Health System strengthening we're looking at how all of this data comes together and connecting ministries with their health workers and wait they've never been connected before really does strengthen that overall system and not only for a bullet but for you no way beyond and below and the recovery and having a more robust system in the future i think in here no is a platform that is really beneficial for for facilitating that the last slide is a picture of a lot of logos and you can go there and this is just a highlight some of our many many partners that we've been working with through our implementation of em hero we're very transparent and collaborative and and working in the Ebola outbreak countries and even beyond we are going to be working and molly soon to do I GSR for using him hero for ideas are so um that's all for now I don't know if anyone has any questions I haven't seen any in the chat box yet oops oh thanks thanks manda and I think it was a great last presentation to to end and a great last side to end on to show kind of the power of collaboration and interoperability I think you brought a lot of the themes together I'm curious if you have a sense if you have data from the response rates of the you know from health workers and I'm curious how you seem like it'd be an obvious problem of the validation exercise that you ask are you there no response and how do you distinguish they're there but didn't respond from there not there yeah I'm curious how you could seems like a real problem when I'm peers but must have here's what your how you all double dash that's a really good question so our pilot in Liberia we targeted for facilities in four different counties um and i would say that our response rate was okay for a new platform and we've been finding that at with a lot of our different use cases that we've been implementing a lot of what we're doing is doing awareness-raising with different units and the Ministry of Health so they know what I'm hero is and how they can use it and doing a lot of training and you know some of that's caught on i mentioned the mental health unit had no information at all like a mugging is captured and a chance to about the mental health services provided to their health workforce and so they used them here to just find out what you know sort of cases are they seeing and what's important i need and that was a new way for them to get information that they weren't able to get before as far as validating health workers that didn't respond um I want to get back to you on that I'm not sure how that's mean used that or dealt with that challenge but it's a very good question sure I shared eyes is like a toughy and there is a question in that chat about whether you brought together all the health workers in the system but you know where their health workers who aren't registered in iris um um and how was that that's a really good question so for right now we're just um using iris and so public sector paid health workers are there with an iris in Liberia there have been a lot of talk about getting community health workers you know packing them including the mentiras in different countries but that's not where we are now or for the private sector um in question about what kind of responses do we get from health workers the questions that we ask are meant to be quite simple because asking complex questions as someone said earlier is is very hard to do through SMS so a lot of them are yes or no responses and then a couple of them are open-ended responses that we may say are you um you know did you do you have enough supplies that was in the example yes or no do you need more yes or no and then the Ministry of Health can follow up you know more concretely with the health facility manager about what might be needed so I hopefully that's a good enough answer about the kind of responses as we do more use cases and more close with him hero will be able and people get more used to it they'll be able to do a little bit more in-depth responses and we'd also like to eventually add ibr into it this is why i was asking a couple of questions earlier about ivr more as a way to share information probably then get a lot back because it is hard to capture data and data collection through ivr but it is a really good way to send out message blasts or information to a large cohort of people especially if we do community health workers where literacy might be a problem on SMS super thanks a lot and perhaps as a segue into the closing section I have a when you touch upon data governance and interoperability and I wonder you know it's very easy to talk about us all sharing our systems and you'll have a lot of experience and I wonder if you might you know share some of the kind of key next steps for us to like really really share our data effectively and achieve the goals of interoperability you know in the coming there's a lot of efforts underway like us a baa and stuff a litter what do you see as some of the priorities for making that real well that's a huge question and I'm really happy to see the transparency in which so many partners have been working under the ability and thanks to us a particularly the global development lab for setting up the calls every fridays that we have a chance to share challenges and talk about what we're doing and I think we just need to keep doing that and not be scared to share information even if it's something that doesn't work um it's really important for us to say this didn't work and this is why and this webinar has been an excellent platform for many of us to talk about that and you know we've got to keep doing that and find ways to you know coagulate all of those lessons learned and document them and be sure that other partners that are working in this domain that may not be on this webinar may not be on a us a call or aware of this so we're not repeating things over and over that aren't working well I think the things that we do need to focus on are in addition to transparency are this system interoperability because that's where the data lives and in order to really have a holistic picture of what's going on in a country making these systems work together and it is going to take a lot of time and it's going to take a lot of discussions and meetings and and and whatnot and we need to be sure to include our Ministry of Health stakeholders and in-country stakeholders in those discussions because at the end of the day all of this is there data in their systems and we just need to do what we need to do on our end to make sure that it is working for them well that was a great response to a big question and nearly and on some great elements so I think I think we're now going to officially finish this session and go to the to the ending session which which might just be a few minutes or longer depending on how much people have to say we've changed the rules of zoom so now you can unmute yourself and so it's a it's a free-for-all and I'll just invite people to see if they have any kind of you know closing comments or reflections or open questions that haven't got answered yet and feel feel feel free or to type things into that into the chat window if that's easier you if they kick it off guys I think it's like a you know a general take away at a high level from all this as it seems like you know lots of lots of progress made lots left to do and you know I'm certainly encourage it feels like we're you know we've made a progress that will enable us to do better on future emergencies and in a way I think the long the fact that this crisis was kind of a long one allowed for more of that kind of iteration than some of the other crisis that we've been more on the sidelines from so I could seems like we're in in better shape though it seems like a lot of the same challenges connectivity new systems you know having to you know will still be constant when we're trying to you know address feature emergency so both uh you know somewhat optimistic some what daunted from this but it's been it's been really great to see these different reflections and just how much our different systems complement each other that's it for me and then we had one question from earlier in the chat that we we ran out of time and you know I wonder if that's still a question they'd be interested in posing to the group you green which question that was um let me see and and just to reiterate because we we change the settings previously you could not unmute yourself if you'd like to ask a question you can just click on the microphone in the bar at the bottom to unmute yourself if anybody has any questions asked I will scroll back up in the chat I think we had one more Navy during Paul's presentation on that sword to the end of time hi this is we know I'm you know can you hear me hi so the question I had was about how I think somebody was talking about how it was a very international team from the tech development side so it was a broader question on how to ensure sustainability by including the local tech environment and this is something that we're thinking about as well how do you make sure you have local developers local tech folks who would be able to carry on and make sure that the system it doesn't there's not a small breakage and then it stops being used and I think that the question were asking was in reference to maybe some of the most Accords tech yells yeah and to be honest and this is something that I know most of our discussed quite a bit with John Hopkins ht3 program here in Liberia but a lot of the information we were collecting and engaging with during the course of the outbreak as sort of lost its relevance not much now with the new cases it's hard to keep it going I think a lot of the datum we were collecting was the rig specific and as the outbreak becomes I guess less of an issue then some of the data isn't as useful anymore um the one thing that I would say is it what we have healthy is that you know I feel like we're five steps are headed for the government and so it's very hard to get the government on board because they're not really at the level that we are at I know this is something xe3 here in the grid I'm working on quite a bit maybe Elizabeth might want it about that um thanks a lot and I'll give another second if people want to chime in on that important point this is Adele can you all hear me yes yes hi everyone so I had a related question which is that we hear a lot across all these different projects and others that one of the big challenges is that there's not the national capacity sometimes this is around technical skills sometimes it's around being empowered to make decisions rapidly based on information coming in and I'm curious if any of you are aware or if any of your projects have engaged in any kind of effort to build that capacity and overall thoughts on where to go from here because it seems like what's happening is that people are looking at sustainability of technologies and projects related to their particular areas of work but I'm just curious if there's a broader effort to address this underlying gap that's a great question and I think salt you know safer de Maggie's case we're certainly interested you know on the particular thing a building capacity to use our tools you know well we're training everybody and everybody we can to be able to configure and deploy their own mobile apps with our Comcare platform and that's obviously just want a little piece of the puzzle but but you know definitely agree and it's you know it's very hard to build that capacity during a crisis and it's a it's a better time to do it when you're less of a crisis mode um maybe also to answer that question adele is some sort of our program we engaged all these local ngos and at the start obviously doing most of the emergency period they were just using latest sort of work collecting data but they couldn't actually code the surveys and things that we use through owner and odk and as the program went when end we handed over all the electronic devices to all the local ngos and a lot of them have been working with me and my team not honor any official basis but on an unofficial basis and we've been training them to do the coding and giving advice and guidance on how to do that but that's just not with the national government level but that's with local NGOs and local staff here with the idea being that as local NGOs got to play a bigger role yeah they can set up their own monitoring evaluation systems and data collection systems using technology hiya doll this is Amanda with em Carol and Liberia we've trained several staff that were sort of volunteers at the Ministry of Health in a sense and very low level but we've taught them how to do all of the development of the workflows and use cases in rapid pro and they are part of the imperial team there we play them a low stipend and the goal and the intent is for the Ministry of Health to continue paying them after the end of our award but I think bigger term I we can work with um you know sort of task shifting of ICT in a way and and work with local NGOs I mean I don't think we need to look at how we're training people and work on mentoring and twinning for for ICT and informatics with a health training institutions across sub-saharan Africa because we really need to have more of a pool of people that we can work with and trained so that we as this field continues to grow which undoubtedly it will we'll have more people that we can work with super I think another encouraging thing is a lot of stuff is I'd say trending in the right direction you know connectivity is not getting good as fast as we'd better assess as we'd like but getting getting better our tools are all getting easier to use and you know general capacity is generally you know an interest is growing that's also you know a promising sign supernet sounds if anybody else has a has a last closing thought now be a good time to say it otherwise I think we could just end with with gratitude to all the presenters who put in the time to be here and and shared their thoughts really appreciated the candid and open reflection both on what worked and didn't and I'll say thanks to all the albums monkey team who put this together it's really uh it's been you know these kind of events I think are really just you know crucial for us to to stay in touch and be able to work effectively in the future thanks everybody thank you very much I thanks everyone will also be sending out a recording for this if we'd like to filter that and there's also some follow-up information possibly that you want to share on that assessment so I will help facilitate that and talk soon thanks so much thank you for organizing this thanks a lot wyd thank everyone

Be First to Comment

Leave a Reply

Your email address will not be published. Required fields are marked *