Next, I want to talk to you about Rotavirus Surveillance in Bangladesh. This is a surveillance system that I helped to establish and we're going to use this as an example of surveillance system to think through CDC's nine attributes. First, let me tell you briefly what rotavirus is. So it's a viral illness that causes acute gastroenteritis. It causes fever, watery diarrhea, vomiting, and abdominal pain. Now there are different genotypes of rotavirus and you can become infected multiple times but it's really the first infection that's most important because that's the one that's most likely to be severe. Children are at highest risk for severe disease because that's often when you have your first infection. The main clinical outcome from severe rotavirus infection is dehydration and that can be dangerous particularly for very young children. So the most important treatment for rotavirus is rehydration, making sure that you're replacing fluids that are being lost through diarrhea and vomiting. So that's why it's important that people are able to seek care and get treatment for rotavirus infection particularly when they're severe. The best way to prevent rotavirus is through use of vaccines, and we have very good vaccines that are used in many countries. However, in some of the most vulnerable places in the world, the vaccines aren't being used yet and part of the reason they're not being used is because of a lack of data that could be used to advocate for introduction of vaccines in some countries. That's why The Global Rotavirus Surveillance Network was formed. In 2008, a number of countries got together along with the WHO and they decided to join efforts and concentrate efforts to collect data globally using standardized case definitions and laboratory tests to start to build the evidence that rotavirus vaccine was an important public health tool and could have significant public health benefits in many countries. At that time, many countries still had no surveillance for rotavirus, and Bangladesh was one of those countries with no surveillance despite the fact that many of the effective vaccines that were being used globally had been tested in Bangladesh. So we knew enough to know rotavirus was a problem in Bangladesh, but again we weren't using the vaccine. In 2012, we set up a surveillance platform that was based in hospitals, seven different hospitals around the country in fact. This initiative was led by the ICDDRB which is a Health and Population Research Institute located in Dhaka. They've operated in Bangladesh for more than 60 years and often work very closely with the government of Bangladesh to put together surveillance programs such as this one. The objectives of the surveillance were to create baseline data on burden of disease so that we could make a case if you are aware that the burden of disease was considerable so that policymakers could consider whether or not introducing the vaccine would be a good use of public health resources. We wanted to specifically document and quantify the contribution of rotavirus to hospitalizations. So we weren't trying to find all cases of rotavirus in the country, but we were trying to say something about the proportion of children hospitalized who had rotavirus. The case definitions we used were those that were standard worldwide. So there was a suspect case definition which was any child less than five years of age who was hospitalized and reported having at least three episodes of watery diarrhea per day as part of their illness and those children had rectal swab collected, and if we found rotavirus antigens in that rectal swab, we considered them a confirmed case of rotavirus. In the hospitals, our surveillance physicians created a line list of all children who met the suspect case definition, and every fourth child was selected for laboratory testing. That decision was made based on the resources available to us. We couldn't test every child because we didn't have enough money but we thought we probably wouldn't need to meet the objectives of our surveillance. Demographic and clinical information were also collected from these children using handheld tablet devices and they were uploaded to a server in near real time. So we very quickly could see how many children were being enrolled at any given time. The rectal swabs were collected and stored frozen in the field and were shipped back to Dhaka where they were tested in the lab once per month. Once the samples arrived, the rotavirus antigen tests were performed and the results reported back to the surveillance system within a few weeks. The results from the surveillance were regularly analyzed and reported. So, this slide shows you an example from a quarterly rotavirus surveillance report. This report was made available through our government partners website. So what this table shows you is the proportion of children aged less than five years hospitalized with acute gastroenteritis. That's the suspect case definition that I told you about and those that we sampled that had evidence of rotavirus infection. It also lists a number of children who had intussusception. I didn't tell you about that part of the surveillance because it's not important to our story, but I'll just briefly mention here that intussusception is a condition where the bowel folds in on itself and causes acute pain in the abdomen and is life-threatening. Intussusception is sometimes caused by rotavirus infection and has sometimes been linked with the rotavirus vaccine. So it was an important component to include in our surveillance here but you don't need to know more about that right now. So if you look back at the table, the first column shows us the hospital name. So you'll see all the seven different hospitals listed there. The second column says A-G-E, which is the abbreviation for acute gastroenteritis. So that's the number of acute gastroenteritis hospitalizations that each hospital reported during the surveillance period. The next column shows you the children among those who were sampled. Remember that I told you that only approximately one in four children were sampled for the surveillance because of resource limitations. There's one exception here, LAMB Hospital, which is the fourth row down, and that's because that was a small hospital and so we decided to enroll and sample every patient there. The next column over shows the proportion of tests that we had run that were positive for rotavirus from those children's sampled. You can see, in this report, 81 percent of the kids that we tested had rotavirus antigen in their rectal swab suggesting that January in 2014, there was a lot of rotavirus causing hospitalizations. In fact, the vast majority of hospitalization of kids for gastroenteritis was due to rotavirus. This is another figure from that report just to show at another way that we visualize the data for people who are interested in the surveillance. So this shows the proportion of hospitalized acute gastroenteritis cases with evidence of rotavirus by month. Along the x-axis, you'll see the months and years of the surveillance. The y-axis shows the proportion that were rotavirus-positive among the kids we sampled. You can see here a strong seasonal pattern where throughout the winter months such as January, that I just showed you, approximately 80 percent of all kids who are hospitalized with that syndrome will have rotavirus. That proportion decreases quite a bit in the summer, but we were able to pick up very strong seasonal patterns. Finally, I just wanted to show you the reporting structure. It's quite simple. So the hospitals enrolled in surveillance report the data back to ICDDRB and they send the samples to ICDDRB where they're tested and the data are correlated and analyzed. The analyzed data then go to the Institute for Epidemiology Disease Control and Research, which is the organization at the Ministry of Health and Family Welfare, that's responsible for disease surveillance and reporting those results back out to other government partners and to the public. So that institute then makes the data publicly available and helps to gather stakeholders together on a regular basis to disseminate the findings and discuss what the best public health response should be. So, as I mentioned, there are yearly seminars with national stakeholders to talk about these results and how they might be used for public health. There's quarterly reports. We've published our findings in international peer-reviewed journals, and you should have read one of those for this lecture. The current goal is that rotavirus vaccine could be introduced in Bangladesh in 2019 due in large part to the evidence base that's been generated from the data from the surveillance system.
