 Pen i'n gwneud am yma Ben Cooper, roeddwn i ni ddiolch chi i fynd i ddisgwylion gwaith ar beth oedd yn ymgyrchol i fynd i am hynny'n gwybod rhywun yn ymddangos am gwybod arall. A dyna'r unrhyw deilig ymgyrch, rydyn ni'n gweithio i'w gweithio'r hynny o'r 20 miliwn cyfrifiadau sylwgr arweinydd, oherwydd 300 miliwn cyfrifiadau sy'n gweithio a oherwydd 70,000 Cael eu ddweud o'r unigwyd, o'r ddau cymdeithasol, oherwydd byddwch dros ymateb sydd gennym cynhyrchu cynhyrchu cynhyrchu cymdeithasol, ac mae'n gwestiynau ei ddiweddy o chytffynu yn ychynig o'r ddweud o'r ddweud o'r ddweud, ac mae'n gwestiynau yn gweithio'r ddweud o'r ddweud, a'r gwestiynau yn wefbitigio'n ddechrau, yma, a'r ddweud i'r ddweud i'r ddweud. Chad. Transmission is fecal oral, and although a hepatitis E virus has been isolated from surface water bodies in Darfur and in South Sudan, there's an increasing body of evidence that's pointing to person to person transmission within households, being the most likely form of transmission to explain the dynamics that we see. In terms of control, as you've just heard, the impact of water and sanitation responses is uncertain, ac mae'n gweld yn cael ei bwysig o'r wneud hyfforddiant yw'r cymdeithasol iawn. Ond mae'n bwysig o'r herfodol i'r vaccine i'r hefyd, ac mae'n bwysig o'r hyfforddiant i'r hefyd. Mae'n gweinio hv239 o'r hechelin. Mae'n gweinio yw'r cyfrifydd pan hwnnw. memor yw'r cyfrifydd yn wneud i'r Chynau, oherwydd mae'r 16-to-65-yreol sy'n gyda'r newydd yn ei bwysig. A rhefnoddau effektifol yn unig gynhyrchu'r dynnu tyfu, a'r trol rhan amdano rhaglion yn y Lansed in 2010. Mae'r ddweud i'r dynnu'n ddweud yma'r ddweud i'r Ynrydd. Ond mae'r llyfr yn sgwrdd i'r ddweud i'r dweud i'r ddweud i'r ddweud i'r ysgrin. Felly, rydyn ni'n gwybod i'r cyfnod o'r moddyl math. First of all, by estimating key epidemiological parameters that are involved in transmission, so things like the basic reproduction number, which indicates how many secondary cases would result from a single infected case, and then using these parameters to build or construct a model and evaluating the effect of the vaccine in an outbreak situation. And then evaluating further the benefit of extending the vaccine to pregnant women and younger and older age groups, which the vaccine is not currently licensed for. So, as we said earlier, MSF responded to a large outbreak in Uganda between 2007 and 2009. It was in refugee camps in the Kitcom district and resulted in about 10,000 cases and 113 deaths. About 50% of those cases were in just three camps, Agora, Madio Pay, where the outbreak started, and Palaga. So, we fitted a deterministic compartmental transmission dynamic model to these three outbreaks. And that basically just means that we can divide the population into compartments and depending on their infection status, and these people can move between compartments depending on the transmission dynamics. And I'll explain that a little bit better in the next slide. For model one, you can see here that we have an SEIR model. So, individuals can be distributed within any of these four compartments. S is for susceptible, that means they're not immune and not yet infected. E is for exposed, they're latently infected, but not yet infectious. I is for infected and infectious, and R is for recovered and immune. And the solid line or solid arrow shows the potential movement of individuals between compartments and the dotted line shows influence from one compartment to the other. So, the infectious people can influence the rate at which susceptible people move to the infected state. I'm going to present the results of model one, but as you can see, we explored a whole range of models, six models in fact, that represent the uncertainty that we have around transmission dynamics. But the results were broadly similar. So, this is where we fitted the model to the three outbreaks in Agorio, Mariope and Palaga. The circles indicate the real data, the observed cases, that's the number of cases per week, that we saw in each of these outbreaks, and the grey shaded area represents the uncertainty around that. And the solid coloured lines indicate the model fits, and the coloured shading, the uncertainty around that. So, you can see that the models fitted pretty well to the three outbreaks. I'm going to skip over this slide a little bit in the interest of time. This shows how we used the models to ascertain the model parameters a little bit better. But I just want to bring your attention to this lower row of figures where we estimated the basic reproduction number, or R0, for each of the outbreaks. And you can see that there is some variability between the three outbreaks, but that the values are relatively high. Not as high as, say, measles, which is around 15 and highly infectious, but higher than Ebola, which is around 2, and kind of equivalent to polio, which is around 8 or 9. So, using the Bayesian analysis, we created vaccine effectiveness distributions from an analysis of that data from the large randomized control trial that was published in the Lancet. We assumed 80% coverage and dose intervals of one month between the first and second dose, and five months between the second and the third dose. There was no evidence in that original trial for any effect from a single dose, so that was excluded from the analysis. So, we plugged this effectiveness distribution into the model, and this is what we see. So, this figure shows the percentage reduction in the number of cases which is given in grey, the number of total deaths in orange, and the number of deaths of pregnant women given in blue. This is a reactive vaccination response to doses after the first 100 cases. You can see that we can expect quite reasonable reductions in both cases and deaths up to about 30% using this kind of approach. This is the same scenario, but just reacting a little bit earlier after the first 50 cases rather than the first 100 cases. You can see that we will see a much larger effect up to about 45% reduction in cases and deaths if we react a little bit earlier. But something to bear in mind is that the vaccine is only licensed for use in 16 to 65-year-olds that are not pregnant. So, the effectiveness in all of these other groups is assuming that the vaccine is safe and effective in those groups. But you can see that if we don't vaccinate pregnant women, that actually we will see very, very little effect in reducing deaths in pregnant women. And that's because we're not acting fast enough and achieving herd immunity when it's needed, which is basically before an outbreak starts. So, we decided to explore that, and this is pre-emptive vaccination before an outbreak starts. And you can see this is for two doses before the first case that we are already seeing much, much larger effects than in a reactive vaccination response. And this is again assuming that the vaccine can be shown to be safe and effective in pregnant women and older and younger age groups. And something to note here is that we're even seeing a larger effect even if we don't vaccinate pregnant women. If we vaccinate everyone else, except pregnant women, we will still see a large effect, and that's because we're starting to see an indirect effect as a vaccination approach. And if we look here, this is the group, the 16 to 65 year olds that are not pregnant. If we simply add the pregnant women, although we don't see a very large difference in the number of cases, we see quite substantial benefits from in terms of total deaths and in terms of total deaths in pregnant women. And then if we manage to vaccinate using three doses before the first case, so in a pre-emptive approach, and vaccinate everyone, we can see almost 100% of cases being prevented and herd immunity being reached. So, the conclusions. Reactive vaccination can lead to important reductions in mortality, particularly if it's implemented early in an outbreak. The potential for this potential for much greater impact if the vaccination approach is preemptive, particularly if the vaccine can be shown to be safe and effective in pregnant women and in children and older age groups. The results were robust to extensive sensitivity analysis with different model assumptions, but it's unclear to what extent these results were generalized to an urban setting, such as the Chad outbreak, which in the previous talk, and endemic settings, and this is an important area for future research. And I'd like to acknowledge the Uganda Ministry of Health and the WHO and the MSF Uganda team at that time that helped to generate the results and respond to the outbreak. But I'd particularly like to mention Geoff Mercer, who supported us for some of the earlier hepatitis E work. He died suddenly in 2014 and will be remembered sadly. Thank you.