 Thank you. It's OK like that. Thank you for this introduction. And so I feel sorry that this is me standing in front of you here today to present the work that we have done in Central African Republic with this MSF E-Care tool that we presented last year. Frank is a national nurse there. He is currently supervising the activity, and it would have meant much better to bring the field's perspective and experience. Just also a note to tell you that I apologize. We have switched the presentation because then I will be doing two presentations of the different tools that we designed with the same technology and support systems. But so I will first speak about the main project that was designed in this algorithm to improve the management of childhood illnesses in primary health care and then give you the talk that was initially planned first on the derivate tools we did for a vaccine campaign. So where is it? Here. Everyone was doing very well. OK, so you know that in our MSF context, when we talk about primary care, we talk about context where we lack qualified health workers. These health workers have access to a very limited number of diagnostic tools and that they work in an environment where the childhood mortality due to infectious disease is very high. And all these results often in insufficient quality of care delivered to our patient, but also a huge over-prescription of antimicrobials. And so in order to improve the quality of care and rational prescriptions of antibiotics for the children aged 2 to 59 months in primary care, we developed MSFE care with a group of expert pediatrician. So first, this MSFE care is a clinical algorithm. So it's clinical pathways that will state very clearly the step-by-step procedures that the clinician should follow in order to appropriately identify the disease and prescribe the treatment required. This is based on a simple syndromic approach, meaning that we only use sign-in symptoms that the lay-out health worker we work with are able to recognize. And we only base the decision on available resources in primary health care. And this material was transformed into an Android application that helped the clinician navigate through the complex protocols, helped them to increase the adherence to the recommendation we make and limit the interpretation that often occurs when we give the usual guidelines that we had developed before. So last year we presented the feasibility, the result of a feasibility study that was done just for a few days in Congo. But what we are happy to share with you is that thanks to Frank's work, we have implemented it now in three peripheral health centers in Central African Republic and that the 24 health worker that we are conducting a consultation for under five are now using it. So this staff, where our nurses and community health worker, that we are appointed by the MOH or the committee gestural to handle this consultation. And to implement the tool, we have sent the attrition for four months to prepare the deployment and help in the training and implementation. But we also used the actual supervision team that was already there, which was mainly one expatriate midwife that is supervising all the activities in this peripheral health center supported by MFF and Frank, so the national nurse, more in charge of the supervision of the pediatric consultations. The training happened in four days in each of the health center one day of actual training and then followed by three days of face to face supervision where the supervisor helped the clinician in the first days of use during the consultations. At the end of the training, two tablets were left by a health center to be used and one solar panel had been installed so that they have power access. The three health center have no internet access so we have developed an online peer-to-peer data transfer system so that at each supervision visit the supervisor can get the data from the tablet and bring them back to Berberati where they are synchronized with our central server. Throughout this intervention, we tried to assess whether the tool we had developed was appropriate with regard to the clinical situation that the clinician encounters and whether the clinician were ready to follow the recommendation given by the tablet in term of diagnostic and treatment. We also monitored the level of use looking at the number of consultation weekly, monthly and also referring it to the routine health information system looking at what proportion of the total consultation are done actually with the tool. Then one of our main objectives was to improve the rational prescription of antibiotics and to look at that, we have done some consultation observation before the introduction of the tool. Now we can also look at what data are automatically collected in the system because the clinician tells us what he finally prescribed at the end of the consultation. And because we knew that this was not collected the same way, we also tried to collect data from the routine consultation register where the clinician report their prescription and we compare a sample of 200 consultation before in January 2016 to 200 in January 2017. And so here since the end of November we have registered in the system 5,300 consultations and if we look at the consultation that we're doing at that time we saw that if the uptick was different from one health facility to the other, from February on we have more than 80% of the consultation done with the tool. The median consultation duration that is recorded within the application was six minutes in this consultation with an interquartial range of four to 10 minutes. So when we look at whether our tool was appropriate to cover the clinical situation encounter we saw that almost all reported symptoms in the system were addressed by MSF ECR. And in more than 98% of the consultation a diagnosis and treatment was proposed by MSF ECR. So it was offering a solution or it was covering the situations. Now when we look at the reported adherence in 96% of the consultation clinicians say that they would follow the recommendation of the tablet and this was even higher with regard to antibiotic prescription decision. We have just sent now an anthropologist to assess more in-depth what is the perception of the tool not only from the user side but also from the community and the caretakers. But already at the end of February we administered a questionnaire to the users and what was interesting is to see that the majority of them reported that the tool was easy to use also they had to get used to it in the first days that it had improved the consultation and especially with regard to helping them in calculating the dose of the drug they wanted to prescribe and their perception was that it was very well accepted by the community. And a lot of them asked us to extend this so bringing them some more tools for other aid groups but also for other medical activities. So one of our objective was to decrease the antimicrobial prescription and what we saw before the implementation of the intervention both through the register and through the observation of consultation was that half of the patient coming for an acute illness were receiving an antibiotic. This is really the same result that we find everywhere when we look at a health facility in sub-Saharan Africa where the proportion of fever due to malaria is high still half of the patient receive antibiotics and this we know is too much when we compare it to the etiology or fever study that we have even if this is scarce. And after the introduction of the tool we managed to decrease this proportion to about 20% when we look at the register and the full dataset that we have. Now another interesting things in this project is that this decision super tool is also a data collection tool and that helps us to have some insight into the... I mean we can look at what's happening in the consultation process more closely and what is interesting is that when just looking at the proportion of the diagnosis we have we can identify some problem. Here what we saw is that in Gambula proportion of lower respiratory tract infection was quite important and higher than in the order and higher than we expect and this is something that we can take and then go back to the clinics and discuss with the clinician how they reached this high level of lower respiratory tract infection diagnosis. So to conclude on that what we saw is that this tool had that value. We had positive impact on the quality of care and rational prescription of antibiotics. Discussing with the users and the supervisor they had the feeling that this tool had allowed to bring some knowledge that the clinician had received in previous training but this bringing the tool helped them to actually apply what they had learned in their consultation process and that this was driving them through a systematic procedure that they could continue doing if we remove the tablet because they learn by doing in front of the patient in front of the specific clinical presentation. And then it can help us better target the supervision we do. We had a lot of question of whether this would be feasible and replicable in our MSF context. What we saw here is that it's as required a very limited technical support. Most of the technical support was taken by Frank there. Also he mentioned that it would facilitate his work if the tablet, if the center were connected to internet so that he would not have to do this process of collecting the data and synchronizing. Still it was feasible within his work load and all the supervision clinical support was integrated into the setup that was already there before the intervention. So and the coordination team, supervision team and users all ask us to continue and to extend. So that's also something positive for us. So now we are in a phase that we face the challenge on how do we transfer this from the pilot to scale and we identify some key aspect that we still need to work on so that we can make it available for the users. We realize that we need to strengthen the final version of the algorithm but also better allow the customization of it so that it can be adapted to different contexts. And we need to integrate this tool within the log, medical and IT departments so that they help us to maintain and update the hardware, software but also make sure that we can regularly update our clinical content. And that's what we are trying to do now to create this support to, I mean, all this environment that to support the deployment in our MSF, Switzerland missions but also in the other oppression centers that have expressed their interest. So just stop here. Thanks a lot for your attention. Thanks for all the people who are still and have been involved in the project. And especially thanks for Frank who is doing all this work in the field and Sergio who was implementing it. So, singi la mingi. Thank you.