 Thank you. Everyone, good morning again. I will present on implementation of routine viral load and outside cascade analysis. UN did set targets. UN did set targets for reaching 1990 by the year 2020. And the first 90 was to make sure that by 2020, 90% of people living with HIV and AIDS near the HIV status. They also said of those, 90% have to be retained on unretro-viral therapy. And finally, they said 90% of the population of people who are on ARVs have a suppressed viral load. But four years before these targets are met, we still have a global coverage of viral load of less than 30%. And we also, in particularly Malawi and Zimbabwe, some of the countries I will present on, we have a viral load coverage of 17% and 5% respectively. And this, we say, we want to achieve 90% of suppressed viral load by 2020. So this is some food for thought for everyone to see whether this is achievable or not. So my presentation is that from 2012, we did start implementing routine viral load in MSF projects in five countries, and there were six projects that we're managing, in Lisoto, Malawi, Mozambique, and Zimbabwe. All the settings we did, our projects were raw settings. Now, so all the sites were implementing a scaled up viral load using dried blood sports samples. These are quite stable and you can transport them after a longer period of time. They don't need any kind of media to store them. They also were using a centralized viral load platform, which made sure that we have a high throughput for all the tests that were being done. So all the countries that I've mentioned were performing viral load, and except for Malawi, which had a two-year period for all the patients when it arrived. Now, what we did is that we're following up a specific algorithm. And the specific algorithm was to also assess that there has to be viral load coverage for all the patients when it arrived. So that's the first thing we had to look at. Then also, we looked at that the step two was to make sure that when the viral load is done, then there is action on the results. And the first action depends on whether the result is below or above the threshold. In this case, which was 1000 copies per meal of the viral load. So if the viral load was below the threshold, there was need for us to offer differentiated care for the patients. For example, offering them clubs and commit ART groups or first tracking is a mode of art delivery. And I think the next presenter will talk about this in detail. Then also, if the viral load was above the threshold, there was need for counseling to be done. And subsequently, after the counseling, there was need to make sure that if a patient again is the second viral load below the threshold, then they will remain on first line. And if the patient is a viral load above the threshold, then they again will have to be switched to second line. So basically, this was the algorithm that we're trying to follow over the past period. Okay. So what we did is that we analyzed, analysis was done between January and November 2015. We reviewed clinical and laboratory records to make sure that to determine whether the steps of the viral cascade were being implemented and again within a defined period of time according to the guidelines of that particular country. So the results we found were presented to the program staff and the barriers for implementation we identified. Then what we found out that from the 24,200 skis-to-three patients that we analyzed, the overall viral load coverage was skis-to-five percent and it ranged between 32 and 91 percent. Then also, we also looked at viral load results that were above the threshold and again in this scenario, we realized majority of the size viral load proportion of patients, the viral load above the threshold between 10 and 20 percent except for Mozambique, which had a quite high proportion of patients that did the viral load which was around 20, 40 percent. And again, reasons for why it was like this Mozambique were not very clear. Now, to explain why we had either bad or good coverage, the first success story was that in some of the settings, there was task shifting or sample preparation to lay workers. That means people who are not nurses or doctors were taught on how to prepare samples and they would do it. Then also in some instances, we are using electronic medical records so that this flag up patients who are due for viral load and then tells the nurse or the clinician or whoever was responsible to make sure they bleed the patient for viral load. And the third success story was that there was demand creation with the patients and this was done either through the civil society and patient education to make sure that the patient when they come to the clinic they emphasize the clinician that I need my viral load, it has to be taken. And the civil society was pushing the service providers to make sure that they provide the viral load to the patient. The main challenges were that at times at some instances in the clinics, there was poor patient triage and patient flow was not very well done. Therefore, there were missed opportunities and some patient would go with the viral load done even if it was due. Then there was also a prolonged turnaround time from the laboratory in terms of delivery of the results. So at the end of the day nurses and the clinicians will end up getting demotivated in repeating viral loads and they won't be getting their results. Then again in terms of action on the viral load that was above the threshold, we were using some tools to record the enhanced adherence counseling and the recording was between 37 and 70 percent. And also repeating of the viral load after I faced viral load was above the threshold, the coverage was between 23 and 16 percent. And the switch to second line was quite poor with a switch of between 10 to 30 percent. And again we realized that there were some reasons as to why there was poor action or good action on high viral load or a viral load above the threshold. The first success is that if there was a dedicated focal passing to identify and follow up patients, this was quite well done. And also if there was flagging up of results to make sure that if you see a result which is high, it was easy to identify and make sure you take action. There was again a use of the enhanced counseling register and high viral load forms which are shown on the right. These were tools that we're using as well to make sure that recording is done for enhanced adherence counseling. And also from the laboratory, there were monthly lists that were sent to the health facilities and the program managers so that they follow closely with the patients and action taken upon them if they had a viral load above the threshold. Some of the barriers included that there was a poor patient triage as I said earlier on and again they won't be dedicated staff member to perform an adherence counseling. And when we mean dedicated staff member, somebody who is employed on a salary and making sure they are following up counseling needs of the patients. There was again in some instances lack of supervision and follow up of the clinics and follow up of the sites. And also a major barrier was lack of task shifting and decentralization of second and art. I want to take you back to 10 years ago when we started treating HIV and AIDS. We said nurses are not capable of initiating patients on ARVs and as a result there were so many delays, many patients died and this resulted in us not being able to quickly curb the HIV and AIDS epidemic. Therefore we are saying this can be repeated if we still have got those bottlenecks where task shifting of second line, ARRT initiation is not yet done and this is common in most of the facilities and in most of the high bed and countries. So again we also looked at the cost of no action. If we do viral load there's a cost implication involved and again if you look in Shanghara in Mozambique we spend 8,000 on performing a viral load test but those tests were not used and therefore that man was lost. And we're saying this has to be addressed as well. Again in terms of the patient if I remain sick I'm going to see the second line. There are costs to the patient and again in the public health arena if you don't treat a patient and they remain with the high viral load, a viral load above the threshold, there are chances of spreading the HIV are high and they keep on infecting new people and again this brings a burden on managing those new infections and making sure that our own treatment. So in conclusion I would want to say that scaling up of viral load is feasible in resource poor settings and again it's very important that we analyze the viral load cascade at site level as it is essential to ensure that tests are taken and results are utilized. There's also need to make sure that when you invest in viral load there is also investment in programmatic implementation of viral load as in establishing a viral load testing capacity. It's not just about buying machines, it's not just about bringing a lot of laboratory agents but it's a world to make sure that programs are able to handle what is involved in terms of viral load. There's need to manage patients with the high viral load, there's need to make sure that everything that we put in as a cost is put to use and again there's a very urgent need to make sure that task shifting is done and decentralization of second line heart initiation and follow-up is done. Like what I said earlier on in most settings this is what is causing such a poor switch rate and this making patients finding it very difficult to again be doing well on air of ease. So if this is not addressed very early we are likely to run the risk of going back 10 years ago when people were dying and nothing was being done. So with this having been said I want to thank you very much for your listening and now to acknowledge the ministries of health in the different countries and also the field teams MSL field teams in the different countries and the patients particularly in Lisoto, Malawi, Mozambique and Zimbabwe were part of this. Thank you very much for your listening.