 Hwyl i arcofwiel, mae'n gwybod hwn o'i gwybod yma, 5.30 o'ch meddwl ystod, mae'n fwy gweithio ein bod chi oedd yn cael ei bod yn gweithio ddigonol. Felly mae'n gweithio efallai yma yna 5 mwy, mae'n gweithio bod ffwrdd. Felly, gallu. Mae'n gweithio'r gweithio, rwy'n gwybod. Mae'n gweithio'r gweithio'r gweithio'r gweithio. Mae'n gweithio'r gweithio'r gweithio. I can't start, I think. Do you want me to do it? OK, I've been asked to make a start, otherwise there'll be no drinks. You will be desperate for our drinks at the end. Just like to welcome everybody to our very first seminar of the academic year for the International Centre for Evidence in Disability. So we're very excited to have the first one. I would like to just explain a little bit. We'll have the seminar and then we'll have an opportunity for some questions. And then at seven or before, if there are no burning questions, there'll be drinks and nibbles available and we'll direct you in the right direction for those of you not familiar with the London School of Hygiene and Tropical Medicine. I'm just going to hand over to Alan Foster, Professor Alan Foster, who is a co-director of the Centre for Evidence and Disability. He's also the president and CEO of CBM. For those of you who don't know, he's one of the key international organisations working on disability and development and is also key funder for this study. I'll let him introduce the study in more detail. Thank you, Maria, and good evening everyone. I'm just going to say a few words and a few slides to really introduce the speakers and also the purpose of the study. Maybe a bit of background to the centre. The centre was started at the school two years ago. It came out of about 10, 12 years ago a group of us moved here who were very much kind of ophthalmologists and involved in prevention of blindness. And we had a centre with the WHO called the International Centre for iHealth. And that work gradually developed at the school and as it developed then of course we started working with epidemiologists and social scientists and health economists, et cetera, et cetera, in the school who then actually kind of said, you know, we're very interested in eyes, but it's a bit narrow. And so gradually what's happened is we've created a research group that's addressing different impairments starting with visual impairment and blindness but then moving on to hearing impairment, physical impairment and as you'll see looking at disability in a wider context. So that's kind of the background to what happened. So the centre is new, it's, as I say, a couple of years old and the centre is across the whole school and if you're interested to know more about the centre or to collaborate with the centre I think there's going to be some leaflets about the centre and it's got an email on it and you can contact any of us. And this is the mission statement of the centre, working in partnership to improve health and wellbeing among persons with disabilities through excellence in research, teaching and knowledge. So it's actually taken the school mission statement and then adapting it for persons with disabilities. And we developed values by which we want to work with and those values are that we want to be inclusive and I think that will come out in some of the talks that you'll hear a rights-based approach to disability that we work in partnership with many organisations around the world and many supporting organisations and that we aim for excellence in everything that we do. Many of you will know disability and be very familiar with it but some won't. I, by training a medic, I'm a doctor and I have been on a very steep learning curve about disability over the last five or seven years. This is the definition of disability according to the international classification of functioning and health from the World Health Organisation and it's very much addressing the person with disability in his or her world. So from a medical point of view we're often interested in the health conditions, the diseases. Those health conditions may lead to an impairment in the body. That impairment may be sensory or motor, intellectual and that's the impairment. The impairment then can lead to reduced activities. So for example, somebody who cannot see will have a problem flying a plane. The activity is reduced because of the impairment and then that reduction in activities means there is reduction in participation in society and there are other factors besides the individual that aggravate that situation of reduced activities and reduced participation. Those factors are environmental but they're also often attitudinal and the way people think about persons with disability and act towards them. So that whole interchange of factors is what actually produces the disability. Therefore we need to look at the person with a health condition that's causing an impairment in the context of the whole environment, the person in his or her world. You'll see this picture of this lady later on in the talk but I want to use her to help that understanding of what we mean by disability. This lady I do not know. I guess Ayla may know her or others. She will have a health condition which might have been TB of the spine or a spina bifida or a degenerative condition. That's the health condition, the medical aspect of it. The result has been that she's got physical impairment. The activity she cannot do is walk and because she cannot walk and have that activity she may have limited participation. As a child she may not be able to go to school. She may not have employment. She may not be able to get married. All different aspects were external factors and attitudes begin to disable her and exclude her from her society. That's the context of understanding disability. The person in his or her world. So why did we do this study? We had three main purposes to do the study. The first was a previous study indicated that volunteers at the village level called key informants could identify blind children in a village and refer them for services. So this was a very cost effective way to identify blind children in a large rural area and refer them for medical and surgical treatment. The key informants were able to do that and were volunteers. So the question was raised if key informants could do it for blind children could they do it for children who were physically impaired children who were hearing impaired children who had epilepsy or other problems like cleft lip or clubfoot conditions like that could the key informants be used to find those children? So that was the first question and you'll see the results of that. The second thing was having identified those children were we able to network with existing services to provide the services that they need holistically and they may be health services, education services other aspects but could we get the existing providers to be ready to provide those services? So again not looking to suddenly bring in lots of extra capacity but rather to use the capacity that was there. So the networking of services and having done that and we won't go on to this in this particular presentation what were the results of children uptake in those services what was the impact on their lives of uptake in those services? So that was the second reason. The third was if the key informants could identify children with impairments could we get cause specific prevalence data for impairments in children that would then enable us to plan future services? So for example if you have an NGO that is doing wheelchairs how many wheelchairs per million population are needed? What's the requirement? Or how many hearing aids? Or how many physiotherapies are required for children with cerebral palsy? We need to have an estimate of the magnitude of cause specific impairments in order to advise governments and other agencies regarding planning of services in the future. So they were the three things. Could key informants identify children with different impairments? Could we network services? And having done that what would be the impact would children take up those services? What were the barriers to uptake and then what would be the impact on their lives? And we can't address all that in this presentation but that's the second purpose. And then thirdly can we get figures for planning so that we can advise agencies on what services are needed for children with disability? So my final job is to introduce our two speakers and first I'd like to introduce Ayla McTaggart. Ayla joined our group a couple of years ago has been involved in running this project in Bangladesh and also in Pakistan. The project is now actually coming to an end. I think it's finishing now. And Ayla is staying on and actually starting to do a PhD on disability in Africa which I think starts and India. So Ayla will be one of our speakers. You can see already she talks a lot. Evening to up to me. Can you imagine that? I've told her to speak slowly, right? So if she speaks quietly, just go like that from the back. Not quietly fast. Just go like that. And second the principal investigator for the study was Professor GV Murty. GV and I have known each other I think 20 odd years now. GV comes from a public health medicine background trained in India, worked for many years for the Indian government. He then worked at the World Health Organization on blindness in children. He is now the vice president of the Public Health Foundation in India and he is the director of the India Institute for Public Health in Hyderabad which has the Disability Centre. GV works half in India and half with us here. So he has a kind of joint position between the London School and PHFI in India. So with that I'm going to hand over to Ayla and there'll be a chance for questions at the end. Is that right, Maria? Okay, hi everybody and thank you, Alan, for the introduction. I do speak really quite fast so please take him seriously on getting me to slow down. I'm going to talk through the sort of background and rationale of the study in a bit more detail and the methodology itself and then I'll hand over to Professor Murty to talk you through the results and implications of some of the findings. As mentioned, the study was funded by CBM and implemented by both the International Centre for Eye Health and the International Centre for Evidence and Disability both here at the London School. We partnered with the Child Sight Foundation in Bangladesh and the Comprehensive Health and Education Forum in Pakistan. So as I mentioned briefly, that's just the outline there, a bit of background on the method and how we used it in this study, results, implications and feasibility and then we'll have time for questions at the end. So sort of as a background starting point, the need for data, and Alan's already mentioned this briefly in his introduction, but we know anecdotally that we estimate there to be quite a large proportion of children with disabilities in the world and that the majority of these is estimated up to 80% are in low and middle-income countries. Having been said, there are very few robust estimates that are comparable and are available internationally for use for the various functions that Alan mentioned in terms of needing to plan for services and general statistical information. As an example here, the UNICEF 10-item screening tool estimates child disability prevalence in Bangladesh and Pakistan at 8.2% and 14.7% between the two countries, but other estimates that we have compiled looking at other literature in the two countries have ranges between 0.64% and 15%. So it's a pretty wide range to be working with. Aside from the needs for estimates, there's also, of course, the UN Convention on the Rights of Persons with Disabilities which very strongly advocates for the need for evidence and the importance of collecting comparable data so that we can, as mentioned, formulate and implement these appropriate policies. Again, it's touched upon by Alan. It's really important that we have this complete information in terms of prevalence of specific conditions and impairments, causality, their impacts on child's activity, limitations or restrictions in participation. And with that information, how we can then use it to facilitate early detection, plan appropriate services and to eradicate those barriers to participation. So prevailing data collection methods for identifying and enumerating children from the disabilities in a lot of low and middle-income countries to date have focused on population-based surveys and or national censuses. These have their caveats. For example, the large sample size is needed for population-based surveys, the fact that they can be costly and time-consuming, mis-children in specific settings and may face problems of non-reporting that you often find in situations where stigma is related to disability, which is also found in census situations, the same situations of stigma and questionnaire design being different and non-comparable across different methods. Given that, again, Alan has already briefly introduced what the key informant methodology is, but it's an alternative to a census or a population-based survey approach which focuses on the training of community volunteers named key informants. So, in this regard, the key informants then go out into their villages, their local areas and case-find children with targeted impairments based on a training process that they've been through. So they will spread messages through their daily activities within the community, list children that they identify with a targeted impairment and then refer those onwards to the appropriate services. Consequently, it's community-based, it's participatory, it involves local networks and it can be suitable for some of the children that may be harder to reach in other methodologies using population-based surveys or via censuses. We'll go into it in more detail, but just for anybody unfamiliar with the methodology, I don't know if you can read that at the back, but a quick overview, a flowchart of how the method works. So, the key initial beginning point is the mapping of a community's social networks and sensitisation of the community in whole, as well as mapping the available referral services and engaging those service providers. Once that's been done, the next step is to identify and recruit key informants from within those villages at roughly a minimum of 1 per 1,000 population, but it's actually based village-wise, so it's 1 per village depending on the size of the village, which obviously varies. The next key step, obviously, is the training of those key informants on identification of children with the impairment as is targeted by the study, and I'll show you some of the previous studies that have different target criteria, and then allowing supporting them in the time that it takes for them to go back into their communities case-fined based on the criteria and communicate the messages as per their training. The next step is to then invite the children who have been identified to a screening camp where they can be screened by a medical team, again based on your criteria that may be specialists in one particular area, whereas you'll see in our study it was a multidisciplinary team where the children then examined, diagnosed, and if they're found to have an unmet healthcare need referred on for treatment, obviously, beyond that, there is the follow-up and identification of services that aren't available trying to identify referral routes in those situations, and then at the end, obviously, in terms of the study design, then you've got the dissemination of results, which is sort of what we're doing now, and we developed a piloted KI retraining module, which is not part of the method as a whole, but we'll talk about that a bit later. As mentioned, it's previously been validated for child blindness in Bangladesh, Ghana, China, Malawi, Malawi and Iran, and for epilepsy in India. I know it's very small. The references are there. If anybody is interested, please just contact us, and I can give you those reference points. Given that validation, again, Alan briefly mentioned what the key objectives of this study was, but what we were looking at is whether we could expand that, knowing what it had been successful in terms of identification of children with particular targeted impairments or health conditions, whether we could expand upon that, so whether or not key informants would be able to identify children with physical impairments, sensory impairments of visual or hearing or epilepsy within this setting, and then, if so, whether we could use that information to assess prevalence and to plan for appropriate services. I just want to mention at this point that we did not include children with intellectual impairments in this particular study, and that was after a lot of deliberation by the project team based on the two key components, that, one, we did not have the tools at our disposal to actually train the key informants in identification of intellectual impairment in a field setting, and, two, that the referral services were not available at the time, so we had no onward referral route for any children that may have been identified. So, it's a four-year study. I started in 2008 and finishing, as Anna mentioned, hopefully in the next couple of weeks. We worked in three districts in Bangladesh and one in Pakistan, and the project developed in a phase-wise manner, so we did the piloting of the field activities in Bangladesh initially, and then they were adapted for use in Pakistan. As a result of that, the focus here on the results that we show you will be from the Bangladesh analysis, but we'll mention some of the main findings from Pakistan as well. There's a touch more information on where we were in each country for anybody who might be familiar with them. We were in Rajshahi division of Bangladesh. We conducted the pilot in five sub-districts of Selajganj, and the main study in Nator and Bogra, covering a total of one million population. In Pakistan, we were in the district of Sialkot in Punjab, and we were working in three sub-districts, covering a total population of 400,000. Very quickly, the key definitions that we use within the study in terms of identifying whether a child did or did not have the targeted criteria. These were quite strict criteria so that we could look at the validity of the study. A child, obviously very straightforward, you had definition. In terms of a physical impairment within the study, we defined that as a substantial impairment of six months duration, or from birth if the child was younger than six months, affecting the child's functions, as per not being able to easily do one or more of the key domains of the Washington group questions, which I'll come on to next. For visual impairment, hearing impairment, these were bilateral, so for visual impairment, it was a preventing vision of less than 660 in the better eye, and for hearing the hearing of 30 decibels hearing loss in both ears, at least, or failure of autoacoustic emission testing in both ears. Now, we've also got the caveat there that also of a strong clinical suspicion by an ears and throat specialist, because we did have the situation at some of the screening camps where the children presented with discharging ears or had communication problems that made it difficult for us to actually screen as per our field testing diagnostic tools. So in those situations, we allow the NT specialist to make a statement, a judgment statement that they presume the child to have a severe bilateral hearing impairment that the child was then sent on for further testing, and for epilepsy, it was simply a history of two or more tonic lung seizures within the past three months. So the Washington group questions, most people probably will be familiar with them, but for those who aren't, they were developed by the World Bank. They are a set of questions that focus on an individual's self-report, a basic function of domains, a basic action, sorry, or functional domains. There is an extended set, we actually use the short set which focuses on a person's ability to see, hear, walk, climb steps, talk, or conduct self-care, for example, washing or dressing. The responses are on a threshold if a person is asked whether they have no problem in carrying out the activity, some problem, lots of problem, or whether they are unable to do the activity. And there is a severity threshold that you can use to then calculate whether or not that person is at greater risk of disability. It is a question set that is used quite frequently, quite widely, in population-based surveys and disability surveys. So we included these to compare the self-reported disability against the clinically evaluated criteria that we had, the objective criteria of impairment. So moving on to the training of the key informants, this was led by the project field staff named community mobilisers who would spend between four and six weeks in each sub-district in the study, networking, recruiting key informants, providing the training and the ongoing support, organising the examination days and facilitating and assisting with the on-ward referrals. As I mentioned, in the flowchart we recruited one key informant per village based on the social networks and capital within that. If you remember, the key first point was mapping the community social networks and taking the key informants from within those based on that mapping exercise. So a range of professions, council members, teachers, religious leaders, NGO workers. We trained a total of 1,510 key informants in Bangladesh and 589 in Pakistan, which works out roughly about 650 to 670 total population per K.I. If you look at it population-wise. The training itself consisted of the development of a flip chart which was used to promote the concepts of the different targeted impairments and also generalised discussion on child disability. There was also a lot of dialogue, a lot of time for discussion, and I have a couple of print-out copies of the flipbook here. If anybody's interested, please come and ask me at the end, I can show you those. They were adapted in both Bangladesh and Pakistan so that they would be contextually relevant. Cs, as I mentioned, they focus on specific messages about specific health conditions and the target impairment criteria and the generalised discussion. The picture of one of our community mobilises conducting a training session. The little insert picture on the left is from Bangladesh and on the right is from Pakistan. In terms of the involvement of the methodology, we conducted it phase-wise. Given that it had previously been validated from one impairment at a time, either for child blindness or for epilepsy in children, there was a concern at the outset of the project that perhaps by trying to train key informants to identify multiple types of impairments at the same time that this may have meant that children were being missed. So in the pilot, the five sub-districts had different, slightly different processes in each. In the first fall, key informants were trained on one impairment type only, on the identification of children with either visual impairment in one sub-district, hearing impairment physically on the other and epilepsy on the other. And then in the fifth, looking at training them on all the impairment types simultaneously. Additionally to the children that the key informants then went out and listed a proportion of children that hadn't been listed by the key informants from within that community was also asked if they could attend the camp so that we could look at the difference there between the children being listed and not being listed. The results from that exercise from that pilot that actually by combining all the impairment types together for providing the training on identification of different types of impairments simultaneously, we weren't missing children, we weren't losing anything by that. It was also found this isn't elucidated by those numbers but also found that the community uptake was much higher in the combined approach. Obviously also we realised that there was quite a high incidence of multiple impairment. So for all these reasons combined it was decided to follow on with the main study with a combined camp approach. So these are just some of the numbers from Bangladesh from that combined approach from the main study. In total we had 57 camps organised and just shy of 4,000 children listed by the key informants. Of those 95% actually attended the camps and were examined for the specific impairments that were targeted. So screen just under 5,500 children who hadn't been listed by the key informants for comparison. As I said it was a multidisciplinary team made up as you can see of the pediatricians and specialists in various domains. So each child invited to a clinical camp was screened by the entire team regardless of what the preliminary diagnosis had been by the key informant and all children were provided with referrals as needed. And it's important I think to note that that was all children brought to the camp with any unmet healthcare need was provided with that referral service not just those who screened positive as per the study. The referrals themselves that as we mentioned was preceded by a referral network mapping in East District to understand what services would be available be they local services, video or subsidised private services and we attempted wherever possible to make links with those services in advance and to refer the child on to the nearest possible centre if not with support provided by the field staff, the community mobilisers that I mentioned and as I said yes referral was offered to all children presenting with unmet healthcare needs. And then finally alongside the key informant method that we were using we were trying to validate the methodology and to understand whether or not it was working in comparison to a population based survey we also conducted quite a large household survey in 15 clusters in the same sub districts. So this was a cluster randomised sample using probability proportion to size and all children within the selected clusters between the ages of 0 and 18 were examined using the same clinical examination form that we used within the CIM screening camps. A few assumptions there from the in the survey design, the disability prevalence at 16 per thousand design effective to 95% confidence interval and an 85% response rate. So within that sample we enumerated 8,470 children and examined 96% of those. It's just a few pictures of our diagnostic tools. There's the tumbling-E and the OAE machine. And one year post-intervention we followed up with 287 of the children to ascertain the impact of both the referral that they've been provided through the study and the intervention itself and we'll briefly mention the results of that as well. And I think that rounds up I hope how we designed and implemented the study in that. I can ask Professor Murty to come and explain what we found out as a result. Thank you Ayla. I think this is the most appropriate room to talk about disability. All of us have exercised for us a vital spine. I have to look up, you have to look down. So it's a good place to have therapy while we listen to the presentation. Ayla has just mentioned how our MI5 agents were spread across Bangladesh. The key informants were spies on the community trying to gather information on the different conditions. But unlike MI5 agents, they were not armed. They didn't have any weapons with them. And also their cover was blown because everybody knew that they were agents. And luckily there was no Anna Chapman in these MI5 agents. So we were on the safer side. I'll try and look at some of the results based on the three major objectives which Alan had mentioned earlier. Looking first at who were these children who were picked up by the key informants and you look at their age distribution you find that in Bangladesh about a third were children aged 6 to 10. Just under a third were children aged 1 to 5 in about 15% were aged 15 to 18 years. Look at the distribution of gender. 56% were male amongst those identified by the key informants and 44% were female. I'm going to come back to the Pakistan distribution after taking you through the next slide. We actually wanted to look at the proportions amongst children who were listed by the key informants and examined at the screening camps were the ratios between males and females were they similar or were they divergent from what the household survey had shown. If you look at the household survey you look at the numbers examined and look at the proportions male and female you find that the proportion was almost the same. So in terms of identifying children to be examined at the household survey there was no gender difference. But when you look at the screen positive for any of the disabilities or the impairments which Ayla has just mentioned you find that even in the household there was a difference in the ratios 55% amongst males compared to the females. So this distribution that we had at the camps seemed to be a true gender difference in Bangladesh. I would not be able to talk about the exact reasons because we did not go into that but that seems to be the pattern. If you look at the Pakistan cohort and I'd like to mention that in relation to the data from Pakistan it came in very late because it was the last stage of the study and we are still in the process of doing the complete analysis with some of the basic demographic information which we have which I am just presenting. If you look at the Kim identification of children by the age and gender you find that between Pakistan and Bangladesh there were a lot of similarities there was hardly any difference and the proportions that were brought up by the key informants in both the countries. The other thing we looked at was the method equally valid in both the situations. Because it was developed in Bangladesh we also needed to try and look at how it behaved in Pakistan and we find and we look at the sensitivity consistently the sensitivity has been high where is specificity in both Bangladesh and Pakistan has been low. And the low specificity that we looked at when we did a sub-analysis what we found was that a large number of children in relation to suffering as diagnosed by the key informants they had some problem. Overall less than 5% of the children who came to the Kimcams with those children who did not have any health problem. But the majority had either mild or moderate grades of impairment unilateral involvement which did not fit into the criteria that we had actually selected for the study. We also had children who had congestive heart failure rheumatic heart disease learning disabilities which were not targeted impairments as part of this particular study. We also had children who came with respiratory infections and other infections because the availability of services in Bangladesh was minimal and here was an opportunity to access healthcare. The further results that I am going to present are only in relation to Bangladesh because we have not had the complete data analysis done for Pakistan as yet. One of the first objectives was whether key informants could identify children with disabilities. The specified targeted impairments that we are talking about and the high sensitivity meant that they were picking up that the case detection was high. The second thing that we wanted to look at was could we use this high detection rate by the key informants to try and estimate the prevalence in those communities. For this particular purpose because the key informants were able to identify almost everybody amongst those who had the severe grades of impairment that we were looking at the population of the villages from where the key informants did the case identification we took that as the denominator which is shown here as 258,000 and then looked at the numbers to work out the prevalence rates. Look at the second column which looks at the figures that we have from the household survey where 8,120 were examined and you compare the prevalence Let's look at physical first. You have 6.2% per thousand in the Kim with the 95% confidence intervals indicated there and 8% in the population based household survey using the same diagnostic criteria to label them as physically impaired. In relation to bilateral vision impairment you have 0.7% per thousand and the Kim is 0.5% per thousand children in the household survey. If you look at the next row that was a problem area for us we looked at confirmed bilateral hearing impairment and you find that it is as though as 0.3% in the key informant method using the key informant method as against 6.4% in the household survey. Now what we realized was that a large number of children couldn't get the objective test done because they had a discharging ear. So if you look at presumed hearing impairment based on the clinical diagnosis of the ENT surgeons it is much higher but for the comparison we wanted to use the confirmed hearing impaired which is looking at either the caustic emissions the child failing on that or pure tone audiometry showing that the child had hearing impairment. But that was one area of concern which we were able to flag up in relation to using the Kim method. If you look at specific health conditions and their prevalence the prevalence in relation to cerebral palsy again the overlapping conference intervals 3.7 we service 2.6 look at epilepsy 1.5 we service 2.2 again an overlap on the conference intervals if you look at a combination of one or more of the above the physical, bilateral vision or hearing then you find that it is 9 per thousand children in Kim we service 14.7 per thousand in the household survey. If I exclude hearing impairment because of the problems that I just mentioned with hearing impairment and then look at one or more of the above excluding hearing impairment you find that there is concordance between the two estimates from the Kim as well as the household survey. We looked at the Washington group criteria also to look at the prevalence of disability and we dropped the question on communication here we used 5 questions and then when we look at the comparison we found a higher prevalence in the Kim compared to the household survey. If we try and extrapolate this to a million population which is useful as a planning tool to plan for a million population you are looking at about if I go right down to one or more excluding hearing impairment you are looking at about 3,000. If we had better tools for hearing impaired it would be much higher than the 3,700 which has been shown in the estimates here. But that is giving us information on how we could provide for wheelchairs for catropediatric catrag surgery or hearing aids if required from using a method like the Kim to help us in planning appropriate interventions. We looked at the Washington group criteria in greater detail and if you look at the key informant method and try and look at the Washington group criteria there amongst the entire population which was examined at the key informant camps nearly 10% had a problem with the domain of vision a quarter 25% looking at either some difficulty a lot of difficulty or unable to do it about a quarter. If you look at mobility, communication and self-care nearly 50% of the children examined of the key informant camps they had problems with mobility communication and self-care. Trying to look at the impairment versus the Washington group criteria if you look at vision and you look at the column of visual impairment here 71.7% that's nearly 3 out of every 4 children who had a problem on the domain of vision on Washington group criteria were detected to have visual impairment. When you look at hearing impairment it's even higher the domain of complaining about hearing and the complaint about hearing is the diagnosis of hearing impairment the concordance was 81% looking at physical about 70%. If you look down the hearing impairment column and you look at communication from the Washington group criteria you find that nearly 3 out of 4 children who had a hearing impairment did have a communication problem. So there is a good concordance between the Washington group criteria and clinical diagnosis in the results that we were able to put together. The gender difference between the different conditions was again very consistent it was nearly 2 is to 3 that is 60% were males and 40% were females and this was consistent across all the conditions though the proportions are very little bit. If you look at the history of onset and look at one parameter that is history of onset at birth in relation to the different conditions physical impairments nearly 2 thirds it was a history of onset at birth. In relation to vision impairment about half at birth impairment about a third and if you look at several policy the last bar that you see there 3 out of 4 children the parents said that it was present from birth so a significant proportion of the different impairments were present at birth but if you look at a comparison between them you look at hearing impairment 6% it means that the opportunity to have preventive measures in a condition like hearing impairment is much higher because most of it is because of otitis media which is later on in life and can actually be acted upon and we know that for vision 50% 45% to 50% and children we also try to look at the access to school this could be either mainstream schooling or it could pertain to special schools there were only 8 children in the entire cohort who were actually going to a special school and we also put together the proportion of children in a normative database who were going to school so that you could compare between the children who did not have any impairment and the proportion of them accessed schooling in Bangladesh if you look at all these conditions the worst situation is in relation to epilepsy or cerebral palsy where the access is poorer than for the other conditions but even for the children the normative database 16% did not when we looked at reasons why these children did not go to school amongst those 993 who did not go to school the biggest chunk was the provider's issue so something which is completely avoidable something which can be changed with 45% nearly half the children not being able to go to school because the school teachers or school admin did not allow them to go to school the second biggest group was parent refusal again another group which can actually be changed with changing attitudes and perceptions that's something which can be different which means that out of every 10 children who were disabled 8 did not go to school because the society or the community or the family did not permit them to go to school so this is something which we can work on to improve the participation of these children in Bangladesh we also looked at whether these children had prior access to any services those who were diagnosed to have an impairment you again find that about 40% that is 4 out of every 10 children did access a rehabilitative services in the past but a large proportion nearly half of them were advised about something but were not provided anything specifically and that's why the sort of access to service is further compounded by this lack of provision of services and you find that in relation to all these conditions the access to services was the least with the hearing impaired in relation to the key informant method we need to keep in mind that a large proportion of them as I had mentioned also suffered from unilateral conditions or discharging here at the time of the camp so that is why they may not have access to services in the past looking at the reasons why they did not access services again something where action can be taken where NGO institutions can play a big role is that there was a lack of awareness in more than a third 38% and in 58% it was a financial issue so making services affordable increasing awareness of services which again can if there is more awareness and services themselves become more affordable because of the volumes so they could be a business model if not an organization supporting the monetary costs of these services issues like transport and other problems were very minimal we also looked at the barriers through two different mechanisms one was a qualitative study done independently and one was a quantitative study and these three barriers came out to be significant and these were the socio-economic status of the families maternal literacy and the distance to a referral to a referral unit like Ayla had mentioned the study in Bangladesh was done in Sirajganj which is the nearest to Dhaka and the service availability was in relation to many conditions only at Dhaka and Dhaka is about 180 to 200 km away from the nearest point and that is from Sirajganj and so distance to referral for specialist services in Bangladesh was a problem even otherwise when we did the resources mapping the follow up impact were 237 children 287 children were followed up and when we looked at some of the responses from them all of them felt that attending the camp benefited them and 80% were happy in the sort of referral network or system that was set up to help them in these camps 5% did not take up the services offered they didn't have to pay for these services but they had out of pocket expenses like transportation etc for repeat visits for one visit it was covered but for repeat visits it was not covered and a third of the children did not go for the repeat follow up visits that were required they went once but did not follow up as required later on and the problems there were the availability of a person to transport them resistance from the family because they did not see a difference if it had cerebral palsy or issues related to transport where transporting a disabled child was a problem so based on that if you were to look at what the implications are from the study that was done here we were able to do a baseline mapping of the available referral programs in Bangladesh and data which added a lot of value on the prevalence the fact that we could try and do a crude prevalence from the key informant to help us in planning for services for these targeted impairments in the Bangladesh situation we were able to gather data on the onset and causality the types of refills needed and the quantum of refills needed look at problems related to access in education to rehabilitation services and the barriers which hindered the referral update if you look at how we could maximise the output using a method like GEM foremost you have to have a strong engagement of the local stakeholders and these are important if you are looking at long time sustainability and it requires commitment from the regional and district level government facilities the educational and rehabilitative service providers community leadership NGOs and the disabled persons organisation so all of them need to be involved in and a similar effort was made in the study in Bangladesh if you do that of a method like GEM you also could work for advocacy at the community level the involvement of the key informants was very crucial for the success of this sort of a method but you need to have the key informants interested and motivated in the long term they could help you in spreading messages even if they are not working for you on a full time basis part time basis the very fact that they have been their awareness levels have been increased means they will put that into practice they spread messages they let people know where they could go for referral even if you are not there later on they help in motivating populations for follow up and could also be of great use in trying to target the barriers to the uptake of services so the important take home messages firstly we feel that a key informant method is an effective way of identifying children with specific impairments the information which the key informants were able to provide can be used to plan appropriate services and this is not just in relation to the medical services but because of data on access to education rehabilitative support one can actually use this sort of data to tailor it to specific needs for education as well as for rehab support the key informants we found could be used to spread messages in the community and follow them up because a large number of key informants came back and said that can you give us more skill to actually try and do something for our communities because we are the local leadership we therefore feel that the camp is a sustainable and participatory community based approach and that the key informants can be excellent for advocacy for children with disabilities and help and that in turn helps snowballs their participation levels there were limitations and I mentioned limitation relation to hearing impairment that was a concern and we still need to work on methods and how we could improve detection by key informants or try and look at a two stage screening process as some people have done where you add a potential screener like the acoustic admissions test along with the key informant list that are provided the opportunity exists now using this model to develop a more holistic tool which would look at other childhood disabilities we did not look at intellectual impairments in the context of Bangladesh because after a lot of debate we found that there was a lack of services and if you identify somebody in the community what is it that we could do was a big question mark for us and that's why we were not able to do that but there are methods there could be countries, there could be communities where such services are available and we could try and have a more holistic tool we have to be prepared for the fact that in countries like Bangladesh or in areas where there is poorer access to health care you would have a large number of children with mild rate unilateral conditions as well as other health conditions who may come up and your specificity if you are using a method like the came with strict definitions as we did may appear to be low but here are children who need services which have not been provided with them and that's why they are here to seek attention of the camps the opportunity in terms of sustainability would lie in trying to have a replacement or a local adaptation for the community mobilizer so that you don't need somebody from an organization or an employee but if you could look at building a capacity using the existing health system in the country to try and build that capacity which is equivalent to community mobilizers then it may be something which one could look at in the long run at the end of the day there was a lot of material that we collected and when we sat down and did an analysis we saw that there was a great opportunity to develop a toolkit or a package on key informant methods and advocacy which could be used in different countries so there are these opportunities and we need to look at these opportunities so the limitations themselves are opportunities for further improving the key informant methodology thank you for your kind attention and the report the final report is due to be released in December 2012 and results from Pakistan we hope we'll be able to get it before we have the report out thank you about 25 minutes for any questions and I'll let GV or Ayla decide who's going to answer depending on the question so any questions everyone's thinking about their drinks yes my name is James Thornberry and I work for Cents International I've been very keen to learn how this data once has been published how it can be used for additional advocacy at international level as I'm sure you're aware and probably people are more aware that I am in this room of the debates around the what's going to happen post 2015 and the fact that none of the NDGs even mentioned disability and evidently there is a there's a very strong argument for disabilities to be explicitly mentioned post 2015 in whatever post 2015 claim on post 2015 one can think of very many ways that this could be cut but how do you see this data being used by yourself as a lobbying tool to assist us in getting disability that's specifically mentioned post 2015 I completely agree that it's incredibly important that we at the moment stand at a good tipping point as it were in the run up to the post 2015 agenda this is a very valuable time for this information to be in the public domain and that already there's a lot of discussion that post 2015 should be about the most vulnerable and the most marginalised but often that doesn't actually specifically mention disability so it's important we have an obligation to make sure that this information does does get out there into the public we will have the main report launching at the end of the month that's going to be quite a chunky report we're also putting together smaller summary reports and policy briefs and a few messages that we hope that you've been able to take away from the talk today but then are transferable and can be used and circulated and of course as we're funded by CBM which has quite a key role in this sector through CBM's fantastic network that it has that goes down to sort of regional, country, local levels that is a very key dissemination strategy for us to make sure that it doesn't just stay within a scientific academic realm but actually becomes quite usable for information that can be used sort of at the local level and obviously of course importantly for advocacy I mean only to add that we are relatively recently part of the bond disability networking group and we are also feeding into that group as well so that's an extra opportunity within the UK Any other question? The main concern again when you're looking at the undertow would relate to hearing and communication that would be the major chunk and like I said there were concerns about hearing impaired because we were not able to examine them using an objective criteria but at the household survey level if I could use that as an example in terms of testing for the OAA there was no problem even for children in less than two years in terms of mother's parental perception there was no problem but using objective test at the key informant camp there was a problem with these two areas but the others like physical where mobility was a problem obviously unless the child starts walking mother may not have realised it so there would be a problem like that we have not done a segregated analysis by different age groups but that's a good idea and we'll have a look at that in relation to education and civilitative support we had very limited information collected as part of this because that was not the major focus but one can tailor a method like the camp to actually collect that information and we were able to see that that sort of information would flow very well with the key informants who are local leaders have taken the communities but you need to tailor it we just looked at were these children going to school if not what were the reasons and these only were for children above the age of five so it's not early schooling or pre-schooling was not actually targeted in this particular study maybe it's one small addition to that one thing that was of note in that education Professor Murphy mentioned there's quite a small section that we have on education but one thing that we found is perhaps as somebody who has more background in education is something you'd expect we found that a lot of the older obviously there's a drop-off as the children get older there's less and less in school but of those who still are in school predominantly they were actually in school grades below the school grade for their age so that's one piece of information that we did have from it this was parental perception so obviously there is a caveat there and the pie chart that was there we had a multi multiple choice answer for that with the opportunity for parents to offer different things so you've got to bear in mind that there may be misunderstandings there and it is the parent's perception and whilst we did try very hard to actually get the child's perception to have a child filling out the form as much as possible in reality we found very the vast majority even for the older children those who didn't have community problems community care communication problems sorry a very complete blank but still the parents or the caregiver would insist on filling in the form so you've got to bear that in mind that's quite important too but at the same time if you the parents may feel ashamed to mention that they didn't allow them to go to school but if you put those two categories together children parents not allowing them and schools not accepting that's 80% so whether they pass the buck on to each other ultimately in terms of action it's at 80% we need to target Claire, that's an easy question the chart which looked at different impairments that actually shows that it is very similar across all the age groups all the impairments but we don't have age stratified data to say if you look at this particular slide in relation to all the impairments this gender difference were seen but the age stratification was the higher gender difference at 0 to 5 or 6 to 10 I don't think we've looked at that right now but that's something which can be done yeah that was my first worry increased mortality either natural or artificial because when you look at the household survey the proportions available were similar so that actually set at rest some of the concerns because when we went into the villages a number of males and females were similar but in terms of disability if there is female infanticide in the first six months of life then that can reflect in this but we've not gone into those causes could be one of the causes that there is higher mortality both naturally as well as an induced mortality in some of these situations yes at the back okay that's a very interesting question I completely agree that's a really important question we didn't hold particular focus group discussion type analysis post intervention but of course we did have the follow up questionnaire for that cohort of 287 which looked it didn't look specifically at whether or not there was a feeling of stigma but that the overall experience felt by the families who we went to follow up was positive I think that's yet another reason why we found that it was that we had to exclude intellectual impairment because obviously if you are labelling a child as having something that makes them different and not providing any support to allow that child to increase their participation to have an equal opportunity then actually I think most of us would agree that's a negative intervention to be having I would also say that the children in the study with unmet health cennies were all provided with that follow up support the referral intervention medical rehabilitation etc so the I guess I suppose it is an assumption there the assumption is that the whole cohort not just those 287 that we followed up that it would have been a positive experience in destigmatising rather than stigmatising because it was increasing the child's opportunity to participate and one other thing to mention I put on the slide there in the flowchart that we had this retraining module this second module that we piloted we didn't actually go into it very much in the presentation we piloted its second module for the key informants that went beyond identification and equipped them with a lot more key messages and understanding about disability in the community so we actually left that as an ongoing method for key informants to actually try to break down stigmas wherever possible to actually try to unpick ideas and to carry that forward and that was a really positive outcome of the initial training that focus on identification is talking to those key informants, those members of the community who had been picked based on their social network that they themselves felt that they had very much had their eyes open to disability they all spoke very strongly that they felt that they had learnt a lot that they had thought wrongly about disability that they had these misconceptions and they wanted to be a part of destigmatising the concept amongst their community so I'd say that I think we hope that it hasn't provided any negative stigma attached to the child by the labelling but it's not something that we looked at specifically. I might just add and it's taking us on to another study but we have done a parent training support group for children with cerebral palsy with this group because there were no services and we're just finishing that now bringing parents together and providing training to them and one of the key feedbacks has been the kind of relief of meeting other parents who've got children with cerebral palsy and that sense of isolation even within the same village they didn't know that those other mothers had children with cerebral palsy and that's been one of the outcomes of the parent support groups that's been a follow-on from this so that might go some way to answering your question. Right, I've got two more questions and then I might ask Alan if he wants to make any concluding remark so I can't remember your name but from SENSE International I think. Just to possibly have a comment but we were having a discussion earlier on today in my organisation about the relationship between having a child with disabilities and the formation of the family and whether there was a relationship, whether causal or not between having a child with disabilities and having a single-headed family member as a household and evidently the consequences of that in your socio-economic position and whether there is a causal range or not. Do you have any data on that? Unfortunately no, the socio-economic data that we collected was quite was based on literacy income and rural urban demographic we didn't actually look further but I completely agree that that's something that we look at in our centre as being really important not consequence because it's not cause unnecessarily but a relationship that is very important to ascertain very much. One last one from Lisa at the top. I'm sure you already know that the Washington group are currently in the process of developing an extended set on functioning purely for children. The set that we used, the short set was the set developed for adults which we had to use as a proxy at the time given that there was no set for children and one of the major differences I think that there will be once the set comes out for children which will be an extended set at the moment there is no discussion for a short set for children that will come further down the line but it will be directed specifically at a caregiver in relation to their child whereas of course we had to adapt the short set for adults which is about self-reports so I think that's a really important difference that we'll see when the Washington group have validated or piloted the new questions I think that will make a difference but we did see there was a concordance that was as you saw in terms of what the clinically diagnosed impairment and the functioning domain were there was concordance there but I think that we'll have to see how the questions finalise, I think they should be launching them within the coming weeks or months OK hand over to Alan for some concluding comment Thank you very much Thank you to GVN, thank you to Arla for excellent presentations but more importantly all the work you've done I think for me I listened to it and I hadn't heard this before and the written report is not out yet listened to it I think there was four messages for me one is that by enlarge the key informants work, they probably don't work for hearing loss at the moment but it's probably better than nothing still even for hearing loss and they are volunteers and so this is and there's lots of the added value they're coming from the community they can do advocacy afterwards and so on so that would be a very kind of I would say positive message of the key informants it works by enlarge the second is if we are looking at severe disability in children remember these definitions with severe disability then a take home number seems to be approximating 4000 children per million population the figure here was 3700 but it will underestimate hearing impairment there will be more of those so we're probably looking at 4000 children per million total population with severe visual hearing or physical impairment and that's an important take home message it is for Bangladesh doesn't mean it's for everywhere but it is for Bangladesh and most of those children half had never accessed services before and of those that had access services a lot of it was just advice and then you had the thing of limited access to education as well so that's telling us about that situation that's perhaps the second message the third message which to be aware of is this is specific to that situation so we can't turn round and say key informants will work in Africa and there's 4000 severely disabled children per million population in Africa we need more information to know that because key informants may work in Bangladesh we think they work in Pakistan but we're not actually quite sure about that it's a different group of people so we need to know that the last thing and probably this is the most important and actually the most worrying I think for me personally is so what for those kids we heard the follow up where you followed up 200 and something but there was nearly 4000 and that follow up was one year and whatever we do with key informants whatever data we get whatever that is if it doesn't make a difference to those children's lives and to the parents to their families then it's the so what question and so I think there's a lot of responsibility on CBM and I speak for CBM but also the other partners there to not just stop and say well this was a nice study and we published a few papers and that's it but we really need to know what needs to keep being done in order for those children and their families to get better quality of life more participation for the society to be more inclusive and that's a lot more work kind of going down into the future years so thank you very much for attending the seminar thank you for your questions thank you to our presenters oh now you get a drink I wonder why you're all stay that's why there's a drink isn't there where is that out this way or the quickest way I think is just going back down to the front entrance again I think straight down to the front entrance and then just go through the double black doors on your right downstairs on the left but there should be a little sign to show you and you'll see all the drinks laid out in the lower ground thank you thank you very much for it was it all that it was good was it still too far oh