 Our third panelist, Philippa Boul, Philippa is the non-communicable disease advisor and leader of the chronic disease group in MSF Switzerland. Her presentation will be on chronic non-communicable care in the untypical refugee setting in northern Lebanon. Thank you. I'm interested, very interested. Hello. Hi, so I move us of course to a very different setting now that of the Middle East context and of a refugee setting and how we look to address the management of a large cohort of non-communicable disease patients in this setting. So Lebanon is a country of almost... hosting almost 6 million people at the moment. Of these, almost one... there is approximately 1.2 million Syrian refugees registered in the country. And since the beginning of 2012, MSF launched the medical response to the big influx of refugees from Syria. And we're currently based in three locations, the cities of Saida and Tripoli and the broad region of the Beka Valley. And in two of these projects in the Beka Valley and Tripoli, we integrate non-communicable disease care into five of the primary health care clinics that we run. Now, just a note on Lebanon. The medical system there is highly privatised. It's quite specialised. So, for example, patients with cardiovascular disease and diabetes are often used to seeing endocrinologists and cardiologists. And there is quite a strong pharmaceutical lobby, so use of generic medicines is not that common and not that well-accepted. There is a very high prevalence of NCDs in the region, as Faud has already pointed out. For example, in Syria before the war, attributable mortality to NCDs was about 77%. So, MSF so far really has relatively limited experience in the management of NCDs and certainly not in this open refugee setting where refugees were spread throughout the country and not sort of confined in camps. And in the middle income contest where the medical, the health system is quite different to the systems in many places in which we work. So, therefore, for us it was very important to understand the feasibility of the model which we implemented, which was different to sort of the existing model of care there. And to understand a couple of particular aspects of our program, one being a home glucose monitoring program that was implemented last year, the other to understand the high rate of defaults that we found so that we could adjust the program appropriately. We therefore undertook a descriptive study using retrospective review of program data, data which had been collected in aggregate tables designed by the field team. And so, in order to respond to this high level of NCDs, amongst this Syrian refugees, there are a couple of things that we needed to do given that we didn't have existing tools or an existing experience on which to base the response. So, one of the first things we did was to work on developing guidelines to support our teams. So, these were based on existing international guidelines but very much adapted to be simplified algorithms that could be easily followed by our medical staff and rationalizing the drug list and the sort of broad range of drugs that's normally used in the country to a very essential drug list and then also having a simplified list of investigations for routine follow-up of the patients. We developed a model of care which was based on a general practitioner system. So, and again, the general practitioners in the country were not so used to managing these types of patients, particularly the more complicated patients. So, in a way, this was a type of task shifting we were already doing in this setting, but we task shifted further to use nurses to do the routine follow-up of patients and community health workers as a key part of the patient education team. Then we integrated this care into the primary health care where acute patients were being seen and we were also providing sexual and reproductive health care. But we had an appointment system with regular follow-up for these patients and we had specified patient files that we developed. We had to support our teams with this and we held a workshop in the country with some international experts to discuss with our field doctors and nurses on these protocols of management. And we have an international physician who's based in the field to support with the daily management. This is just an example of some of the guidelines that we developed and the type of patient algorithms that we have inside the guidelines. And then we have very much step-by-step use of drugs and we have tables to help the doctors shift patients from often the quite complicated medication regimes they are on to the more simplified regimens according to our essential drug list. Then in the middle of last year we implemented a home glucose monitoring program and this was to really assist our doctors who one of the things they struggled with the most was the insulin management of patients and this is in a context where many of the patients did not have access to glucometers or were unable to afford the quite expensive strips for glucometer use. And so our doctors were struggling to really adjust the insulin appropriately based on the measurements we were getting in the clinic. And so we designed a program where for select patients, namely those on insulin and those who are unstable on the insulin but who were adherent to the program so they'd been coming to us regularly for a period of at least three months were given glucometers with a certain number of strips and instructions on when and how to use those strips. And really the objective was to help the physicians with adjustment of the insulin. We followed the patients with HBA1Cs before provision of glucometers and at three months afterwards and had specific questionnaires designed for the patients and the practitioners at that three month mark and our infield expert did a file review of the files at this time. So in terms of the results we've had so far I'll go very quickly through numbers but we've seen over eight and a half thousand patients pass through our care between 2012 and 2014 of which almost 3,600 at the end of last year were in routine follow up and had been seen at least in the last three months. And this represents about 20% of the consultations that are seen in our clinics. I won't go, as I said in too much detail through this but just to show that we still do have a significant number of defaulters the 33% at the end of last year which is why we undertook the defulter analysis. In terms of the morbidities that we see the most common diagnosis is hypertension followed by type 2 diabetes and cardiovascular disease. Now what this graph doesn't show is in fact the co-morbidities and if you look at the number of diagnoses we have it's more than twice the number of patients so a large cohort of our patients are really being managed with multi-morbidities. The defulter tracing sincere apologies there are some errors on this but I'm going to upload the right line after this in terms of the numbers. At the end of last year though so look at me, not the numbers. At the end of last year because of the high rates of defaulters the team actually did a complete file review and tried to trace every patient that had not been seen within the last three months. So they managed to find in both Becker and Tripoli about 30% of the patients and of these they found that quite a number of patients actually moved locations so some of them back to Syria and others within Lebanon. Then of the other reasons for defaulting in fact they differed a bit between the project locations and the team found it was quite interesting to analyse these and see how to adjust the project and found a couple of things. For example some patients had decided to stop taking their medication some patients had difficulty with access to the program and in Tripoli in fact as of last week opened a new clinic in the north of the area partly in response to realising that patients were having difficulty coming back to us for follow-up. The team in Tripoli disaggregated the defaulters by year and certainly found the defaulter rate had decreased from 2012 to 2014 as had a number of the reasons for defaulting. So for example one of the reasons for service in the clinics and this decreased by 2014 and in fact in that time the team had made some quite significant emphasis on trying to implement an appointment system and decrease the waiting times and change the patient flow. The home glucose monitoring system by the end of 2014 we had 85 patients enrolled in the program the majority of these patients with type 1 diabetes with a mean HBA1C of 10.28 on admission to the program now unfortunately at the time of analysis we didn't have very many follow-up HBA1C of those we did have all of them had decreased. We found that 61% of patients were self adjusting their insulin and in fact this wasn't the primary or the initial objective of the program it was initially to firstly help the clinicians with the monitoring and certainly 88% of them on file review were found to be appropriately managing to adjust the insulin according to the results so it was providing proving helpful for them but certainly the team now will look at how to support the patients better as they are using this themselves and on the quality of questionnaires patients were actually feeling a lot more comfortable having the glucometer at home with the management of their disease. So in conclusion in terms of the default tracing I haven't gone through a lot of detail with that but a lot of what the team found were things that they felt able to try and address the default is in a more timely manner particularly and so now they implement a routine system of trying to trace patients who've not been seen in three months to see what the reasons are for them not attending the clinic. For example there were a small number who were unable to attend for apparently being bed bound and we're going to have a system of social workers being able to do some home visits for these patients and so we will do this regularly from now on and then further review to see whether we need to continue. The home glucose monitoring results I mean the initial results from both the qualitative but the very limited quantitative showed some improvements in the patient control but certainly we need to be monitoring that a bit more closely with our doctors to get better results out of it. In terms of the lessons learned from the cohort analysis using these simplified protocols having the expert to support our doctors has enabled this task shifting. Certainly at the beginning of the program it was very difficult for our doctors to manage some of these patients and particularly those on insulin and then for the nurses to do the routine follow up but so far it's working successfully with this thing and we've managed to rationalise the costs in the sense that we previously had a very broad list of medications all of our medication is bought in Lebanon and there was a high use of investigations and now we have these very simplified list of investigations and medications that are used. The appointment system and having the dedicated patient files and a dedicated patient service circuit has allowed for improved follow up of the patients. We were quite limited in this analysis though by some difficulties with the data tools which were inconsistent between the projects and at the moment we're in the process of implementing a better data system that will be used throughout the mission. This will be an ongoing work the other significant component of the ongoing work is putting a lot more work in emphasis into the patient education system and we're looking at further areas of operational research one of which will be looking at simplifying medication administration such as use of the poly pill for secondary prevention of cardiovascular disease. We've now been able to utilise components of the project in other settings so in the emergency context in Iraq we're able to utilise some of the things such as the patient files and the algorithms as well as Syria and very different contexts such as South Sudan where we adapt it to the context. Big thanks to the field teams for their work in this program. I heard that the teams in Lebanon may be watching today so I hope they are because I present on their behalf and also big thanks to the innovation unit in OCG who are the ones who helped us to develop the guidelines that we're using. Thank you very much. Wonderful group all on time so that's great. I'm happy for that. So starting with technical questions let's start from behind please the lady in yellow. Doris from Vienna Evaluation Unit thank you very much Philippa I have just one question if I have understood right you said you have changed a drug regimen to a less to a simpler one so I would like to know if you know accepted by the patients to have less drugs and first change to another drug regimen. Sure no in fact and when I say simpler it's really simpler compared to the existing context not certainly compared to perhaps the way MSF would normally do it it has been a very difficult thing and it's required a lot of efforts in terms of patient education and certainly in the follow-up of default to tracing one of the reasons for defaulting was dissatisfaction with the drugs or was patients deciding to get their drugs from elsewhere and in fact that had slightly increased over time as we had actually rationalised the list because at the very beginning we did have a much broader medication list in the emergency context before we rationalised it. So it is an issue and it's something that as I said the patient support component of the program is something we're really putting a lot of work and effort into doing at the moment and we've just implemented we have now a new dedicated patient support nurse as the normal nurse doctor and community health worker we have someone to follow up the more complicated patients and provide more ongoing education in the end though I guess we are still seeing hopefully those who are the more vulnerable and unable to get medication elsewhere and we still do have very large numbers in our cohort so I mean the acceptance is it's not completely unaccepted and we will continue to work on proper explanations to the patients to explain why these regimens are needed. For the clinicians of course it was also different to what they were doing and this is one of the reasons we had this workshop with international specialists to explain the evidence base and we really made sure it was very evidence based guidelines and we could support the clinicians to understand that and how to utilise these drugs. Question please here. Hello, Alvaro from Action Against Cancer in Spain we have some programs in Lebanon in the Bacavali they're actually CAST transfer programs and we are using mobile phone technologies to communicate with beneficiaries I wonder if this is something you explore and what would be your take on using massive SMS communication in order to decrease default rates to understand what's the situation of the cohorts. Thank you for the question I'm going to come and talk to you because it's something I'm very interested in I'd be interested in your results we have discussed it within the team and of course we used this for the default of tracing but we found significant difficulties and a number of patients who had changed telephone numbers regularly or the telephone number was not working anymore we now have a chance to facilitate that but the initial discussions we weren't so sure how successful it would be in terms of the proportion of patients would be able to follow up but it's certainly something we're thinking about and I'd love to hear your experience. If I can have the privilege of being here also I will not raise but just to add the complexity of the situation when working in a context of let's say Lebanon in this case which is very private system of health facilities where there's like a sort of non-camp refugees there's no camps in official camps in Lebanon so I can see how work would be difficult to present sometimes a different approach for treating with a problem like there it's a third question last question now so please. Hi there thank you very much for that really interesting presentation I was just wondering if you had a method because you were talking about catchment areas you were talking about access to these clinics did you have a method for identifying clusters of high rates of NCDs and particularly I guess diabetes? Not in terms of identifying clusters as such certainly we chose in the Becker Valley we chose our clinic locations strategically to try and cover not only different parts of the Becker Valley but of course places where there was less access to care in general and also different types of population groups and then in Tripoli as I said we were initially based in one location but now we're spreading to another because we realised we actually we do routinely collect information about where the patients come from and we realised we needed to spread elsewhere but in terms of the NCDs we're not looking for clustering as such we have done some access to healthcare studies we've done a series of those studies some of them quite recently at the end of last year to look at different health needs and different aspects of access to healthcare so I can discuss those with you later if you like Thank you very much Felipa