 I would like to thank the previous presenters for discussing the role of technology in saving time. And this is really aligned with my presentation right now. So my name is Ziad, and I'll be talking about a project conducted in Central African Republic. So I'll be using the word CAR to stands for Central African Republic. Also, the project name in French, Projet d'Alerte Precoce, which means Project of Alerts and Prevention. So it's known by the word PAAP. The Ministry of Health in CAR has a requirement to collect data on 20 diseases every week. It's known by the word weekly reports. And due to the context of CAR, it's a challenge to deliver these reports, all of them, and also on time at the end of every week. So MSF has done an intervention to test a software or a phone application called Argus, which was developed by the World Health Organization in Lyon. From your left-hand side to your right-hand side, I will explain the structure. So it consists of a phone app that is installed on an Android phone, where the representative of every health care center would enter the data on 20 diseases. So a number of patients and number of deaths as well, based on the clinical symptoms. On the back end of the phone, every disease will be sent in form of an SMS, which means if the report has 20 diseases, there will be 20 SMS sent from every phone. The server consists of a laptop that is connected to a USB drive that has a SIM card inside. And it will receive all the messages. And then it would be flipped into the laptop using an open source software called Frontline SMS. And this software is connected to a dashboard plus a GIS software developed by MSF to visualize the data. So the mission and also the Ministry of Health can see the reports every week, plus all the stats for the diseases and the mortality. And also it can be exported into various formats of reports and database. It was done in a province in the southwest of CAR. It's called the Mamberi Kadeyi. It has 21 health care centers. The size of this province is equivalent to the size of Belgium. It's almost exactly the size of Belgium. Here you have the location of these health care centers. Also I would like to mention here that the value added that NSF is not operating in all of these centers. So this is a value added as well to know what's happening in the different health care centers where MSF doesn't have presence. The colors of these circles represent the cell networks used. In the northwestern part of the province, there is no cell coverage from CAR. So we have used cell network from Cameroon. And the rest is CAR, three cell networks. The headquarter of the Ministry of Health of the province is here and also the server is located here. This is a summary of the portfolio of the diseases. So we have received, during 15 epidemiological weeks, over 3,000 cases. I would like to remind you here that these reports are based on clinical symptoms only and plus on the practice in these health care centers. So there is a caveat to the story how the classification is made. Plus that we have also received alerts. But due to the timeline today, I'm not presenting the data on the alerts. This graph, it shows the completeness of the reports. So we expect to receive 21 reports every week. So out of these reports, how many we have received? So what we have done here, we have compared PAP, the intervention, to the reports made by papers for the previous year, 2015, for the province of Mamberi Kediye. Plus we took data from an adjacent province, also comparable to the size and criteria of Mamberi Kediye. It's called the NANA Mamberi for 2016. On your left-hand side, the vertical line shows the percentage of completeness or proportion or the percentage of reports have been submitted every week. And on the horizontal line is the number of epidemiological weeks. So you can see clearly that the completeness percentage is quite high and the median was 81%. WHO has a threshold for a minimum 80% to have a completeness level. So in case there is an outbreak that with such completeness, you can ensure how you can react on the ground. So you can see here that almost over two-fold the difference between the Mamberi Kediye and PAP and also almost less than double with NANA Mamberi. There was also a statistical difference between them as well. When it comes to time, did they submit the reports on time every week? So the same comparison was made. Here at the beginning, we noticed that the timeliness was low, so we have done an investigation to understand why. So there was various factors sometimes if the cell network is busy by sending SMSs. Sometimes there were some bugs in the software. Sometimes the staff, they have made some errors. So here we have done a workshop training. That's why you don't see data in the middle on this week. We have done a workshop and also we worked on the bugs with the IT consultants for the software. And then we saw an increase. Plus we have done a further analysis, Kaplan-Meier survival analysis, to look at the number of days it takes to complete the report, the reception of the report. And we saw also an improvement over time. We are not presenting the data here due to the also short time of the presentation. One of the questions, OK, we show that this is working. We have done an utility assessment by the staff that they found this is really excellent. And then the question was about money. That what if MSF, as we always say, we are not staying. We would want to hand over how much is going to cost the MOH and WHO to pay for it. So we did the costing analysis based on operational. So it's less than 1,000 US dollars per health care center. The yearly costs. So the operational cost over 15 weeks was a little bit over $40,000 for the operations for the SMSs. In terms of an improvement in the reporting system, it was really significant improvement. We could see it also with the staff, with the MOH. There's satisfaction as well with this. There were some anecdotes about it that they said, oh, there were some sites that they had an epidemic of silence. And now they are sending us the data. And some they have called the system as the invisible Pigeon messenger of MSF by really transmitting the data from remote areas to MOH and also with MSF. And also starting of this month, there has been an integration of PAP into the MSF mission activities. And also there is a possibility to do a handover to WHO and MOH. I would like to acknowledge everyone has participated in this from the mission and WHO colleagues in Lyon and CAR and also the IT consultancy and also our colleague Prince, or Jean Coulis, who is the supervisor of PAP, who sadly couldn't make it with us today due to visa issues. So with this end, I would just thank you also. Thank you, Ziad. We have a time for one or two quick questions. I might have missed it, but is it zero reporting that they must report every disease every week, even if there's no cases? Indeed. It's called the zero reporting policy that even if there is no case, it has to be reported as zero to make sure that it's a zero and not was missed by accident. Yeah, thank you for this. Is there a question behind you? Bernie MSF, can you just explain what the $900 is paying for? I'm happy to do it, but maybe it will take some time. So I'm looking at the chair right now. I have two or three minutes. OK. So we did the costing with a mindset that how much it will cost to run it as operational. So we have excluded logistics, for example, in terms of taking MSF vehicle, taking also the expats costs. What if this would be run locally? How much it costs? We looked at our local operational costs for the technology part. Are there any other questions? We have one at the back there. Is today MSF, I have a couple of questions. One is, it's surprising not to find malaria amongst the list of reported diseases, since it's the most prevalent disease, most incident disease in the area. So that would be one. But then what kind of supervision is accompanying this to ensure that the quality of the data is actually good quality data, and they're not sending anything, whatever? So malaria is among the alerts, and that's why the data is not presenting for the alerts. What we have used here is that it was the MOH requirement, so malaria was not part of it. For the quality, can you please repeat the question? How do you ensure the quality of the data? How often are the health care workers getting supervised to ensure that the data that they're sending is actually good quality data and relates to what they're seeing at the health center? So the quality here, you can think of it in the process at different levels, either how the diagnostic is made, if this is actually a true case or not, or if it's in terms of data entry as well. When the data is received, there is an MOH staff that he is the supervisor of the surveillance system that he will be calling and confirming the cases. So he will be the one making a call, whether the cases have been reported are they true or not. So he was taking lead on this, because it was a collaboration between MSF and the MOH. When it comes to the diagnostics, it depends on certain cases. MSF, when there was a possibility, MSF was doing confirmed diagnostics in the lab. But the policy of MOH usually afterward, even if there is an error, that they don't fix this. So because mind you, this was a surveillance system. So this is here, MSF was doing this for the first time. Oops, I made a mistake here. So back to your question, data quality in terms of diagnostics and in terms of data entry, there was monitoring for it for both of them. Okay, so thank you, Ziad. Thank you. We have more time for questions, everyone. Thank you.