Welcome back, everybody. We'll continue with our discussion of some examples of programs that have tackled this enormous challenge of improving the health of mothers and children in high mortality settings. I'd like to take a few minutes now to talk about a more specific and targeted approach to health improvement at the community level, and this is an approach called Care Groups. I'll first be talking about several projects that were implemented by World Relief, an NGO that has its US based here in Baltimore. They were the first organization to implement this. and one of their own staff who works in Mozambique, Dr. Peter Ernst, who you see in the photograph in the bottom left. This was his idea and it proved to be so successful that is, it has continued to grow and expand in many different ways as I'll mention. A care group is a group of ten volunteers or so, sometimes there are 12, but somewhere in that range. It's a group of ten or 12 women who meet on a periodic basis, sometimes it's every two weeks, sometimes it's every month. But, they meet with a paid supervisor for a couple of hours to learn an important health message. Then, the care group volunteer visits 10 of the households there are, that are around her and she shares that message with her neighbors. And so, the system is set up at the beginning so that every house is included in this process, and every care group volunteer is responsible for roughly 10 or 12 households. In this case, you see, in this diagram, the care group volunteers are called WHE's which in Cambodia means Women Health Educators. So, the actual names of the volunteers may vary from place to place. So, the care group is the, group of volunteers that each take responsibility for visiting their neighbors. So, over time as these care groups continue to meet among themselves and learn more, they become quite the powerful groups. They become very capable of conveying messages and educating their neighbors with basic health education. This is just one small example of many, of some of the rapid uptake of these behaviors that are being promoted by care group volunteers. this is from Cambodia, and this is in two different parts of the same project area. And this is the uptake of hand-washing practices and you can see how quickly this behavior changes as a result of promotion of this by care group volunteers. In Cambodia, also you can see that there were very low levels of immunization. But with promotion of immunization and teaching mothers about why these are important and helping to link mothers and their children with immunization services when the government provided programs came to the village, the project had a very dramatic improvement in immunization coverage. These are just two of many child survival interventions that were promoted by this project and many other care group projects in other locations. But, they uniformly have been able to achieve very similar rapid uptake of coverage and achieving very high levels of coverage that have sustained themselves over time as long as these projects functioned. And you see, in this next graph here, the estimate of the decline in mortality in the Cambodia Care Group program. This data was collected by the care group volunteers themselves. and so, one of the interesting things about care groups is that when they meet each month, they report on the births and deaths among their 10 households that took place the previous month. and that information can be tabulated very easily by the project and mortality rates can be generated from that. And, in this case, an analysis of these data showed the mortality rates that were being measured by the volunteers in the blue compared with the overall national mortality rates as determined by the UNICEF surveys that were done there and the projected decline into the future. But this data is still not published yet. But, I think it's a good example of the power of this approach. There is one published article on the effectiveness of care groups in terms of lowering under fiv e mortality. And this is from another world relief project in southern Mozambique that was carried out in the 1990s. And, in this case, you see on the graph, the decline in under five mortality that was recorded by the care group volunteers. This is the line that says vital registration. These are the vital events that the care group volunteers themselves recorded. But, what made this study important is that it was possible to independently measure decline in under five mortality through a pregnancy history survey that was done by the independent interviewers. And this also showed a fairly dramatic decline in under five mortalities, as well in the project area. If I had the opportunity to go back to this same area, and do an assessment of an expanded impact of this care group program in a larger set of districts compared to the original project. And you see some of the findings from this based on the vital events data that were collected by the care group volunteers. You see here in the large black line the national decline under 5 mortality as recorded by UNICEF statistics. And you also see the red dotted line the DHS, the Demographic and Health Survey data for declining under 5 mortality in the Gaza Province where the project was working. For the Vurhonga II project, which is the project that was reported in the article I just mentioned you see their decline compared to the national rates and the projected provincial rates But the expanded impact project you see in the far right in the red line there, again these are data that were collected by the care group volunteers themselves. And so they weren't independently collected but I think they do give a sense that the care group process can be very effective in reducing mortality. And we know it's very effective in expanding the coverage of key child survival interventions. I had the privilege recently of leading an evaluation of another care group project in the Sofala province of Mozambique in the central portion of the country. And, in this case, the program reached a larger population, 1.2 million people. Whereas the world relief project, the expanded program reached I think about 250,000 people. So, this a larger scale operation here. And, the food for the hungry divided their program area into 2 parts just as they did in Cambodia, if you recall. But, in this case, the first several years were devoted to implementing the care group project in area A and then in the last year and a half or so of the project they implemented the same program in area B. And the project was focused primarily on nutrition. But to a lesser degree, on control of diarrheal diseases by using the care group model. And you see here, based on household surveys, the project was able to demonstrate a fairly dramatic improvement in nutrition. The percentage of undernourished children declined substantially in both area A and area B. Comparing the baseline levels to the end of project levels five years later. And there was extensive evidence to demonstrate how this might have occurred. There were many activities that the project had that were geared towards improving nutrition from promotion of exclusive breastfeeding, improving complementary feeding. Practices increasing the intake of specific nutritious foods in the area and so forth as you see on this slide. So there was a lot of confirming evidence that this change in nutrition that was observed to really had a basis for it as a result of the project activities. It is also interesting that there were dramatic improvements in coverage in both area A and B And that these changes were able to be achieved within 12-18 months. In area A, the program went on for 3 and 1/2 years before beginning in area B. And in area B, the activities only lasted about 18 months, but some are dramatic improvements in coverage achieved through this approach. And again, in this case, the care group volunteers were reporting births and deaths at the time of their meetings, and this information was reported back into the projec t information system. And it was possible to compute the mortality rates based on this information. And in the graph on the next slide here, you see what this information showed. It's a little bit complicated, and I will try to explain it. But, the solid green line is the under 5 mortality rate that was taking place for the whole province, the Sofala Provence, based on DHS data. And calculating and estimating the decline on a yearly basis from that data. And that is our best estimate of what would actually be occurring for most of the Sofala province. The solid blue line that you see were the vital events that were reported in area A. But unfortunately, the project did not begin reporting these vital events until more than a year after project activities began functioning. So presumably, there was some significant mortality decline that took place during that initial period when vital events weren't being registered. The solid red line are the vital events that were reported in area B. And in this case, they were able to start registering vital events at the very beginning of activities in area B. So, I think it's reasonable to assume that the levels of mortality in areas A and B were more or less similar before the project began. And so, we've drawn these dotted lines in there to estimate what we think might have been the case, although we certainly don't have the numbers to confirm this. But, I think you can make the case with the limited data we have here and comparing this with the effectiveness of other care group projects as well that there has been a fairly dramatic decline in mortalities result of this methodology. Using just the change in coverage data and using a tool called LiST, LiST, that's Lives Saved Tool. It's a new procedure that's being developed here at the Department of International Health with other colleagues around the world. It's a way to estimate the numbers of lives saved as a result of changes in coverage in key child survival interventions. And so, using this inte rnet-based tool, we have an indirect estimate of the number of lives saved by this program of somewhere between 4,500 and almost 7,000 child deaths. You see here the term corrected and uncorrected. So, for the corrected version, we have subtracted out what we think the underlying secular decline would have been without the project. And with the uncorrected deaths, this change has not been subtracted out. But the cost of this care group project in Mozambique as implemented by Food for the Hungry is very small. It's only $2.87 per beneficiary per year, and by beneficiary, I mean mother and child. we've estimated a cost per life saved of $441 dollars, and a cost per DALY averted of $14.72. Which when compared with other approaches, make it very highly cost effective and effective intervention. There have been a number of other care group projects that have been reported by other organizations, and they are shown on this next slide And they're all very effective terms of reducing under five mortality as estimated by the LiST, too. And very low cost is based on the calculated cost per beneficiary per year. When one looks at the cost per DALY averted of this particular care group project, you see in the next figure here that it is a particularly cost effective program compared to the other USAID child survival projects that have been put through this analysis. And in comparison with other more comprehensive programs. But even taking this into account, we see that all of these approaches are highly cost effective based on criteria established by the World Bank and the World Health Organization Commission on Macroeconomics and Health. Let us take another break here before we move on to our final example. And, we'll see you in a minute.
