 Hello everyone and welcome to this GAVI webinar series hosted by the HIST Center here at the University of Oslo. I see we're just a few minutes past the hour, so I think we'll get right into it. So this webinar is focused on the use of DHS-2 to support immunization supply chains. Just to then state from the start that DHS-2 is not an ELMIS and is not trying to replace that specific software for supply chain management, but rather a last mile solution digitizing service delivery points, health facilities and integrating that into the digital environment. And that's really the use case and the approach that we're taking, leveraging the use of DHS-2 in many different contexts. So the outcome level objective that we're looking to achieve is improvement of supply chain management, the availability of medicines, the ability to make decisions with that data, and then improving the overall health outcome. So really that's the approach and methodology we use. For this webinar specifically, we'll be then having three different presentations. First, a country mapping of vaccine stock data in DHS-2 instances that will be done by Augustin Dushim and his brother group. That'll be followed by an overview of Thrive 360 National Control Tower for immunization supply chain. So that is a project jointly with UNICEF, which was implemented together with HIST Nigeria, and we'll have Barnabas and Kumba presenting an overview of those dashboards. And then third and last will be George McGuire from the HIST Center here at the University of Oslo. Going over different data use potential and capabilities with the data that will be presented and possibilities for improving supply chain management and how this can actually be operationalized. Again, aiming at those improvements in the supply chain, stock availability and contributing to overall improved health outcomes. So welcome once again, and I'll hand over then to my colleague in HIST Rwanda Augustin Dushim to get us going. Over to you Augustin. Thank you very much everyone. As Brino said, I'm Dushim Augustin from HIST Rwanda. I'm LMA's implementer specialist. So I'm talking about the country mapping of DHS-2 for logistics regarding the vaccines. The objective of this assessment was to find out which data for LMAs is collected into DHS-2 and how this can be used to improve stock management and show other countries how this can be done. So there is a list of 13 countries that have been assessed. There is Burundi, Komoros, Congo, Braza, Djibouti, Central African Republic, Rwanda, Sudan, Chad, Mozambique, Guinea, Bissau, South Tomeprincip, Cape Verde and Angola. So there is a list of criteria or information that have been assessed to be collected by each country in this program. First of all, there is a data survey that has been collected. The LMAs information that is already available in DHS-2, the answer was yes or no. The frequency of information that is collected, either by weekly, monthly or quarterly, the level of data collection, either by community level, facility level, district or region, the level in which this information is entered into DHS-2, the information about expiration, the dashboard to display this information, the stock on hand at the beginning of the period, the quantity should receive the lost or expired stock on hand at the end of the period replenishment and the type of data entry form that is used. So regarding the methodology, the assessment started on 4th to 14th February 2024. We have reviewed the available information regarding the stock in 2DHS-2. We have worked with history teams, our colleagues for additional information and the other clarification. Then we have completed the data collection tool, it was an Excel spreadsheet based on the information that have been downloaded in the reporting forms from DHS-2. Then we have analyzed the data collected in 2DHS-2. Here we have to know that only the availability of the reporting forms was assessed, the quality of information collected was not part of this assessment. The findings, this is a summary of vaccine information that is already available in 2DHS-2 by each country and by each information. Yes, AY stands for yes means that this information is collected and N stands for no means that this information is not collected for this particular information and a particular country. We have to know that the key data on opening balance, quantity should, quantity received, quantity lost and closing balance is available in 2DHS-2 for most of the countries. And most of these countries do not collect the product information, replenishment and there is no dashboards. The custom forms is the only commonly used reporting form used by different countries that have been assessed. Regarding the number of information that is collected by each country, we can see that 5 countries, this is Burundi, Central Africa, Chad, Mozambique and Guinea-Bissau. Out of the 13 countries that have been assessed, 7 criteria or 7 information have been collected by these countries while only 4 countries did not have any information regarding the stock data in DHS-2 for vaccines. Regarding the findings by type information that is collected in 2DHS-2, we can see that 9 out of 13 countries that have been assessed, they are collecting stock data on opening balance, quantity should, quantity received, closing balance while only one country has information on expiration date and replenishment recorded in 2DHS-2. This is the level where the stock data is collected and where the data is entered in 2DHS-2. Out of the 9 countries that have been assessed and collecting data on vaccines, 5 countries are collecting data at facility level while the 4 are collecting at district level. Regarding the data entering 2DHS-2, only 2 facilities to our countries are entering data at facility level while the remaining 7 are sending data from facility to be entered in 2DHS-2 at district level. This has the consequence that there is a delay on when data is available compared to when data has been collected and this delay can go up to 1 month. Here we have to mention that all the assessed countries, they are collecting LMS vaccines data as aggregate and on a monthly basis. What type of data entry forms that is used to capture stock data in 2DHS-2? Out of the 9 countries that are collecting vaccines stock data in 2DHS-2, 8 of them are using the custom forms while only one country is using the default form. This means that the custom data forms are mostly used and these ones do not render on mobile devices as designed. For example, at the tables, you'll find that they are displayed as a long list and then the default data entry forms needs to be used. This is a list of indicators regarding the information that is already collected in 2DHS-2 in different countries. This is a list of indicators that can be generated from this information already available. At facility level, we have this coefficient of variation stock availability, stock percentage, stock outcount, stock outlength, stock coverage time and so on. This only requires to collect the information in 2DHS-2 regarding the quantity issued, stock on hand and mark complete on the forms during the data entry process. On the district level, we need some indicators that can be already generated because the information required is already available in 2DHS-2. This is a percentage of district reporting stock availability, percentage of district having electronic vaccines and supply stock management system, stock out event. At national level, there is stock out events, stock out days at national level and this requires only the stock on hand to be collected. At the end of the period, the quantity issued and the stock out days. This is already collected in 2DHS-2 means that all these indicators can be generated because the required data elements are already collected in 2DHS-2 for most of the countries. In summary, the findings show that to the vaccine stock are collected as aggregating 2DHS-2 and on marked basis. The key data on opening balance, quantity issued, received, quantity lost and closing balance is already collected in 2DHS-2 for most of the countries. This means that the indicators that are needed can be calculated for these countries based on the information that is already available in 2DHS-2. The dashboards can also be developed based on this information available in 2DHS-2. We have to mention that most of the countries do not collect information in 2DHS-2 regarding the product expiration, replenishment and they don't have these dashboards to show the collected information. Most of the countries are also using the custom forms to capture the vaccine stock data in 2DHS-2. We mentioned that the consequence of this is that the data collected cannot be used at mobile devices as intended. It means that default data entry forms need to be configured. And also most of the assessed countries collect data at facility level and they have to send data to be entered at district level which may cause a delay to be available for this stock collected. For challenges, there is limited access for supply chain teams. Generally, they don't have access in 2DHS-2 for the stock data collected and there is issue of decision making. There is a different configuration in each country for the metadata data entry forms, indicators and different visualizations. We understand that because different countries have to respond to their current needs. However, the harmonization of this configuration will be necessary to have integration, reporting and analysis. The indicators and the visualizations are not configured and then the data is not used for assessment. We have not been able to obtain information about the use and quality of the information that is available in 2DHS-2. The real potential immediate improvement that can be met are the default forms need to be configured in 2DHS-2 to facilitate the use of mobile devices. There is also need to configure the dashboards for data use and also configure the indicators to be monitored on a regular basis for stock management. There is need to measure the timeliness and completeness of the data for the data quality improvement. And also there is need to assess and improve the access to LMI data in 2DHS-2 by expanding it to a larger number of staff. This is called breaking down silos. For potential improvement that can be done for long term, this intends to address structural and underlying issues. To stock availability by digitizing facilities and integration with central LMI or ERP. This needs to train the program staff on the use of their availability LMI data in 2DHS-2 and also expand the data entry to be done at facility level. This means digitizing facilities instead of making data entry at the district level. This will improve the timeliness and also improve the decision making especially about the resupply. We also need to integrate 2DHS-2 with the central LMI for end-to-end visibility and logistic workflow integration. Thank you very much. I'll hand over to my colleague Barnabas for next presentation. Thank you. Thanks Augustine for the detailed presentation. Good afternoon, morning depending on where you are. Can someone confirm if you can hear me? I need a confirmation. Slightly far, but we hear you in front of us. Okay. All right, so my name is Barnabas and I'll be taking this session. I work for HIST Nigeria and I'm going to be taking the presentation on TRIC360 Nigeria in Nigeria. So as a background, we tried TRIC360 project that was initiated by UNICEF to help bring together all sources of logistics data in order to help the logistic working of countries to be able to have an idea of what's happening and to aid them in taking the form decisions. So in Nigeria, we have made use of various sources of data. You would agree with me that data exists in different forms and sources. So here. Sorry about that. Someone was limited, so I had to mute. So we have made use of about five sources of data in our set of the routine data that exists on the NLMS. My interest due to Nigeria has a very robust national information system that is set up using DHS2 and collects data from about 40,000 health facilities in the country. We also made use of country estimates that the country has for some of the sources of our data. The country also started using the NLMS as the upstream logistic system. Sometime last year and the data from that system was also used for this setup. And of course the last source of data is the predictions based on the various data that will have the system. So we're able to set up several dashboards about nine of them. You see the data from all those sources. The first one is with forecasts and allocations. We're trying to look at data estimates that the country makes every year for their needs and then the supplies that we can get and is provided to those various places. And we have a dashboard for stock adequacy. We have a dashboard for stock consumption, stock status and coverage, stock trends, stock out events, those children, organization coverage trends are there. So this first dashboard here that talks about forecasts and allocations here, if you look at the left hand side here we're trying to compare two years. We're looking at the key interest antigens, how much was forecasted as what the country would need and how much was received within that period. Based on that data we're able to now compute the sort of shortages, shortfalls by each of the antigens here. And we get very good output and how the decision makers to take in necessary decisions that they need to take. This might be a form of trying to make additional purchases or the people in logistics. This is the shortfalls put in the table format by states and you can see for the key disadvantages. Each stage you will see the shortfalls and then you should be able to source that data. We also set up a dashboard for stock adequacy. Here we're trying to look at the various parameters for stock availability. So here we're looking at stock out, we're looking at adequacy stock, we're looking at under stock and we're looking at over stock. So on the left hand side we're able to have like a trend by month for each of the trace antigens. And on the right hand side there we can see BCG, the indicator based on the stock status for that particular period. Here we also broke this down into the various antigens so that you could easily tell the performance or the availability of each of the antigens. And you would agree with me that this would be a very useful indicator for those responsible for managing logistics and stocks at the various levels. The good thing about the CHS2 is that you can drill down to know where there are gaps and then you know what action you need to take. So people at the state level when they log in they're able to see the situation in their state. If you're an LGA person you're able to see what is happening within your LGA. The stock status and current as brought here we're trying to calculate from the routine data the stock available and how long it can take. This will stop coverage in coverage nine weeks. So the first one here you could see BCG, the quantity of BCG available at this moment can take for 7.1 weeks. This is at the national level. And you drill down by state you'll be able to see the situation state by state by LGA you should be able to also see by LGA are able to also get the spider map to show us the situation as well. This is a continuation of the same dashboard where we're able to put this on maps. The data is to come with the same gravitational lightation tools. You can analyze your data on maps, you can analyze your charts by charts. Now make this available on maps so that you can know where the performance is bad and you can easily intervene. Down here you have the analysis by the various antigens. What we have on the screen now is the initial coverage trends. What we have here is the initial coverage trends. And what we've done here is to try to bring about show the coverage by months. You can see the charts are looking at January and February. So we're able to compare this with the mix and mix and unique estimates. You could see the performance as compared to those over estimates for the country. The stock trends is data from the routine NHMIS data. Data that has been reported you're looking at the stock available per month over the last 12 months for each of the antigens. You can see looking at this and you'll be able to tell if there are data quality issues or not. You'll see that the flow is a kind of smooth, no spikes. That shows that there is no cost for where you see a spike. You might want to also, should I say, investigate. We were able to also put up the zero dose trend. These are estimates that the country makes. You might want to know that the country has about 775 LGs local government areas. That's what we call them in NHM. So this is just giving us the country's focus on 150 out of the 775 LGs. So you would see that that's why the chart here is just pointing at some states. There are some states that don't have those LGs. And the maps here also you can see where the reds are. Those are the LGs of focus for the zero dose children. This data was imported into the system. These are estimates that are done outside the system by being able to bring them into the system. And then put them side by side with the data that is available in the system. The 2030 scorecard dash for this another attempt to try to maybe give scores based on some key indicators. The indicators we used to be able to get this done were the 12 months stuck out, zero dose coverage and the ifm. So here, based on the performance in each of those indicators, a score was assigned to each location. And that is what brought about this. This is something that as the country uses it, a lot of tweaking will be done to make it a form to what the country needs. But at the moment this is like walking progress, but the results seem to be very useful. This is another prediction based on the average consumption indication. Here where you see doing predictions based on the standard deviations and trying to see locations where stuck out, there's a possibility of having stuck out as time moves on. So the usefulness of this is the decision makers should be able to know that if particular attention is not paid to these particular locations, then in the future, in the nearest future, there will be stuck out in those locations. And we all know the implication of having stuck out. These are stuck out events. So when we talk about stuck out events here, we're trying to look at a count of facilities where there was at least one stuck out event within the period. So if you're looking at the four key tracer antigens, if amongst the four, there was one that had stuck out, that facility gets counted. So that's why we have this. And then by the right hand side here, we'll be able to get a table that gives us a detail by antigen. And of course you can read down to each facility to know where there was a stuck out for a particular antigen. I mentioned earlier that the country started the use to open the numbers as a mainstream logistic management information system. So we're able to get the data that is available. The opening balances are the LGA code stores into the system. Remember we said VHS2 is used at the facility level. But we want to be able to also have an idea of what's available at the code stores. That will also help the NLWG and other key decision makers to know what they need to do. If they have enough at the LGA code stores, they will know that they need to push them to where they are needed. So this is also a useful information that we're able to bring into the system and put it side by side with other sources of data. So I will stop my presentation here and now I'm about to join my wife from the HIP Center. Thank you. Okay, thanks a lot, Barnabas. That's an impressive demonstration of the analysis that can be done. I'm going to start my screen sharing now. Apologies if the screen doesn't have the perfect shape. It's because I have a large screen. Brenu, can you confirm that you can see my screen correctly? That is perfect. Yeah. Okay. Thank you. So I'm going to continue on what Augustine and Barnabas have presented. And my topic is data use and data quality analysis that can be done in DHS2. So that will focus on the Thrive360 and then make a short comment on other functionality. So this just has an overview on the data quality. It's an extensive topic. I'm presenting briefly what is available in DHS2 and what is relevant for LMIS, for Logistics Management Information Systems in general, as well as for Thrive360. So the first point is that the data that is collected needs to be relevant. That means that we should only collect data that is actually going to be used by somebody. So we see that there's a tendency to collect a lot of data and sometimes people are not entirely sure who is going to use that data for what purpose, but keep in mind that it's a big burden on the health worker usually or on the storekeepers to collect this data, whether it is used or not. Then accessibility. That seems like very obvious that the data that is collected needs to be accessible to a wide range of stakeholders. One thing that we stress in DHS2 is that data should also be available at the facility level, maybe not all data, but rather than having paper records. Our dream is that one day the data that is collected digitally at the facility level and also available digitally at the facility level, because keep in mind that if you are a health worker or a physician at the hospital or clinic, you might also want to know which items are out of stock and that would be great if you could go on your and check that on your mobile phone rather than having to go to the pharmacy and ask or call them. Data completeness, so this is also a really important issue that all the required data fields that are configured in the data entry form, as they call in DHS2, in the reporting form, that they actually need to be filled. So you should avoid having what we tend to see is there's many, many columns, but some of them are systematically not filled. Then it's maybe you should reconsider maybe streamlining the data entry form if for some reason the data is not collected or if it is important to collect all the data, as I mentioned, if it's relevant, then you need to take measures to make sure that all the data fields are actually eventually filled in by all the facilities. And the data needs to be correct. That's obvious, but also very important. So that means that the data that is entered into DHS2 needs to correspond to, for example, the stock on hand or the stock distributed issued consumption, whatever it is called, actually corresponds to the numbers that of the respective transaction. So those have to match because you want to report what is actually happening in the real world. Something that is maybe not as obvious, but it is really important is to have the correct data ranges. We do see quite a bit of negative numbers for the stock on hand. So by definition, stock on hand cannot be negative and also transactions are usually recorded as positive values. And normally they are recorded as integers. So this is something that is important and that can be easily checked in DHS2. The timeliness of reports. So the reports need to be submitted, usually collected at the end of months, but submitted as quickly as possible. So this is one of the advantages. If you are collecting DHS2 logistics data on a mobile device in the facility, as soon as you have finished your data entry and if you have a network and you synchronize your mobile device, then that data is available on the central server and available to anybody who has access to DHS2 instantly. So you avoid any delay by sharing paper reports. So that's what we are aiming for in the long-term. And regularity. So the reports need to be submitted each and every month, usually, or maybe you have a daily reporting, but use it months and that needs to be done consistently. Then the data use. So data logistics data is mainly used for stock management. There's a lot of reporting requirements and dashboard requirements, but ultimately keep in mind that logistics data still needs to be used primarily for managing your day-to-day stock. So you need to know what is in stock. The stock issues the stock receipt to keep track of that accurately. So the stock on hand and the transaction quantities, those are the most important data that we are using in DHS2 for managing our stocks. And then stock replenishment, this is really important. Sometimes not given the importance that it should have, because ultimately everything that logisticians do professionally is to ensure that healthcare goods are available to patients. So you need to make sure that the stock data is collected accurately and timely so that the calculations, whether it is done at the facility, district level, or other level, all the data, the accurate data is available as quickly as possible and the data is updated so that the stock replenishment can be done correctly and stocks are replenished accordingly. Then analytics, I think you're familiar for various data analysis. Logistic service performance management, as we call it, so for measuring and managing quality of logistics services, for example, the percentage of stock items that are in stock. Data triangulation means relating HMS and LMIS data and using that for analysis. And then the reporting, as I mentioned, I think you're all familiar with that, for the routine reporting and analysis according to the policies and national needs. But keep in mind the ultimate purpose of all of the data quality analysis and the data use is ensuring that no child is left without a vaccine because it is out of stock in the facility. That's something that should always guide us. That's also why we stress the importance of stock replenishment because analysis, data collection, analysis reporting, that's all necessary and good, but ultimately it needs to result in having the vaccine in the refrigerator when it is needed to vaccinate a child. Okay, right. So, very quickly, the data quality tools that are available in THRs too. So the value type, so you can block certain values. For example, if somebody were to attempt to enter a negative value for stock on hand, then you can set the system to prevent this so that rather than having to check data and correct it, the wrong data entry is corrected on the spot. Then you have several tools after entering the data. So there's a tool called the validation rules that I will demonstrate briefly that can pick up negative values, can pick up missing values, or blank data fields, inconsistent values. So you can do quite sophisticated calculations and calculate whether the stock balances with the stock on hand beginning of the month and receive the stock issues where it all adds up or not. And you can do outlier detection for distributed quantities. So if you have very large quantities that are likely to be a typing error, because you type 10,000 instead of 100, you can detect them. And then finally, you have the analytics. You can again calculate stock discrepancies. You can have the missing data values count and put them on the dashboard and reporting completeness and the timeliness are integrated native functionality in the DHR store, DHR store core. So very briefly in our performance management concept, there's much to say, but we have these basically these three categories, symptoms, diagnosis and therapy to relate to the work of health professionals that are often dealing with these issues. So the symptoms is called the manifestation indicator that just indicates there's a problem. So for example, if you have a stock out, you obviously have a problem, but it's just a symptom like a headache. It doesn't indicate the reason for your stock out. Then you have, we have the diagnosis that would be analyzing the root cause and determining the underlying problem. So usually this will take a professional and national ELMIS system. There's not a lot of root cause analysis that can be done in DHS too, but it can give some hints. And then the most importantly, you need to take corrective action to correct the problem. So we don't want to just report and analyze stock outs. We want to actually fix the problem and reduce the number of stock outs. And that takes some kind of corrective action. And in most cases that will also be measures that the supply managers need to take in the national ELMIS system. So the diagnosis without the therapy doesn't lead to any improvements. So there's no purpose in just every month reporting that we have stock outs and that they're increasing or decreasing. Somebody needs to take some action. But in order to design, to develop a suitable therapy, you still need a diagnosis. And that does require determining symptoms. So we're not saying that collecting symptoms is useless. On the contrary, it is necessary. We're just saying, don't leave it just with a symptom detection. Somebody still needs to take corrective action. And our last point is that I will present, there's also some symptomatic treatment. Symptomatic treatment like giving, if somebody has broken their leg and you give them an energetic, that doesn't solve the underlying medical condition. But symptomatic treatment is still something that is necessary and can be part of the therapy. So keep in mind that we need to fix the underlying problem, but you also might have to temporarily, hopefully when they treat the symptoms. So I'm going to be very practical after this kind of theoretical and introduction in presenting four measures, four actions that anybody can take with DHIs too already. So this is the overview. Somebody done your conference, ask a very good question from all the things, the presentations that you have seen in the last two days. What is going to change Monday morning when you go to your office? What are you going to do differently than to what you have been doing until now? And these are four very practical things that you can do with Thrive 360 or DHIs to stock data analysis. Very practical and fairly simple. So I'm going to share this, I'm going to talk about the stock coverage time for the shortages and almost expiry. And the last row is basically the curative treatment. For that, you will need a national LMS system to complement analysis in Thrive 360 or DHIs too. Okay, so first of all, missing values, very simple. You have a screenshot of what it actually looks like in DHIs too. So this is taken from an actual DHIs too data entry form. On the left side, there are six data fields and the seventh data field is yes, no option. And of those five fields are filled and two are missing. And you can see a screenshot below inserted on the validation results and you can see left side missing values is two. So that means that out of the seven, two values are missing. And good news, we just had a demo on the V41 that is coming out in May. There you will even have a list that will indicate itemized list. It will not only say two data values are missing, but it will give you exact list. PCG vaccine, those as opened is missing. And so this is very easy to check. You just need to run off configure your validation rules and then press the button run validation that can be done at any time with any, for any facility, any month. And it will show you a list of all the missing data values and that hopefully will prompt then to either the facility can check themselves or it can be done with some advice. You can see on the right side now all the data fields are filled in and the validation result is a green tick as it should be. That means all the data that should be collected is actually available. So that's the basis because of course if you are missing data and you calculate indicators, especially when you aggregate the indicators across items across facilities, but you're missing data fields, then those indicators are not going to give you correct results. So this is just as I mentioned to increase the data completeness and the hope is that by just visualizing and giving feedback or users obtaining the feedback themselves from this analytics that will raise awareness and hopefully everyone will be motivated to have 100% data report completeness and eventually that data will increase, improve so it will take some time. Eventually that's a good way to monitor and to have accurate figures to know whether is it only one item that was missing data or is it all items, is it many facilities, all distinctions, all. And this is a problem that you can actually fix in using THIs to analytics by eventually having 100% completeness. So then the next issue is stockouts. So if you have a stockout, yes I'm sorry, I skipped one slide. So stockouts, all of the screenshots you are seeing are actually taken from the Niger Thrive 360 database which Barnabas presented and configured. I will not explain all the charts in detail but basically THIs to Thrive 360 allows you to identify facilities, vaccines which are out of stock and the corrective action, the preliminary corrective action is to restock those vaccines as quickly as possible. So you have it at a high level. You can see the districts that are out of stock. You can basically determine from a list what vaccines are out of stock. Where are they out of stock? So which facilities, which LGAs are low on stock? You can see how many facilities are out of stock. You can determine on the pivot table on the right side whether the number of facilities with stockouts is increasing or decreasing. So moving in the stock supply how long have they been out of stock? You can also see that from the pivot table and you can also have like on the right side the column chart you can have an overview. You can see the number of facilities that have stockouts of what vaccine and you can see there is a declining trend which is good because it means that the number of stockouts is decreasing and you can have a very good overview and you can do that analysis at a strict LGA provincial or national level. So this is something that can be done immediately and that has good value because every stockout should be basically treated as an emergency and should be resolved as quickly as possible. So I mentioned that in DHIS2 that's mainly for symptomatic analysis. Here in the small font you can see that if you have isolated stockouts at a few facilities and they're very limited in time then probably that is like a temporary problem. Maybe there was a large increase of the number of vaccinations where there was a glitch with the delivery. If you have several stockouts at several facilities over prolonged time that indicates an upstream supply problem. You could also use the Thrive 360 analysis to kind of locate the problem either in terms of what vaccines are mostly affected, what times were mostly affected. It was a period of the year where there were many stockouts and what geographic level was mostly affected. So was there a certain district where there were many stockouts and other districts are doing very well. So that might help to narrow down the problem. And what you can do is principle simple. So you need to expedite open orders. So if an order was made and not yet delivered requested it is delivered faster than according to the regular schedule. If there are no open orders then you need to place additional orders and ask for immediate delivery. And the other possibilities to redistribute from other facilities which I will present in a minute. And to come back tomorrow to our logistics performance management framework it is a symptomatic treatment because if you don't take corrective action treat the underlying root cause then those stockouts are likely to persist or recur. But it's still necessary. Of course, if you have a stockout you need to do something immediately. Correcting the inventory control system might take longer or analysis but so you need to do something immediately of course. Okay, I think I... I'm sorry. So stockouts, the next measure is preventing imminent stockouts. So as a supply chain manager my objective would be first to resolve all the stockouts. So that means that there are currently no stockouts. And my next concern would then be even if all the facility if there is no stockouts at any facilities then I'm still concerned that there might be some imminent stockouts and the way to look at that is compare the number of stocks number of weeks or months for which stock is left and then compare that with the expected next delivery. So for example, if the facility is making monthly orders and it's the beginning of the month and they have one month of... less than one month of stock left then probably they will have a stockout before the next delivery arrives. And in the table on the left also taken from Thrive 360 you can see the stock coverage time so you can determine. Let's say you're looking at Abahara Primary Health Center has 0.21 months of stock left so that's six days of stock. So unless there's a plan to have a delivery within the next six days it's going to run out of stock by the end of the month. And that helps you same measures corrective action, expedited open orders place additional emergency orders and redistribute from other facilities. So this is a short-term prevention because a stockout has not occurred yet and if you take the measures and they are effective quickly enough then you will actually prevent the stockout but still keep in mind that stockout is likely going to recur if the root cause is not treated. So the last issue I want to present is the redistribution of stock so already mentioned. So here you can see on a map again that is actually taken from Thrive 360 so you can see that the dark red facilities are those that have a stockout currently for PCG vaccine and you can see on a map the geographical vicinity of facilities in purple that have more than 12 months stock. So those facilities are likely to be able to spare some of their vaccines to alleviate the immediate stockout or the shortage and you can see that on a map easily then you can go on also on the pivot table and see how much stock is available to spare between the facilities. So again something very practical but very useful to have it on the map. So you identify nearby facilities with stockouts and shortages identify facility with excess stock and then try to organize to redistribute the stock to resolve that immediate stockout. And keep in mind that the quantity that facility can spare depends on the average distribution so the average consumption the higher it is the more stock they can share. So this is short term prevention as before for the expediting necessary. And finally the root cause analysis is still necessary so this is the analysis of what leads to stockouts actually there's not that many aspects that lead to a stockout so you have the lead time you have the inventory control policy and the irregular orders that is what leads to ultimately to stockouts and those are also the factors the parameters that need to be addressed. So the immediate short term measures as I mentioned is detecting reporting stockouts expediting open orders emergency order redistributing and for treating the root cause root cause is to prevent stockouts from in shortage from recurring you need to analyze the lead time so that's why you need a national LMI system and as a distribution system very important you need to study the inventory control systems or the parameters that you're using like the lead time the review period the safety stock levels to identify if they are adjusted to the requirements and if not then adjust them as I mentioned for determining and implementing the curative treatment so the ultimate therapy requires a professional national LMS system that ideally is indicated with DHS2 and my last slide is just to mention that as Augustine already presented there's other metrics indicators that you can measure with DHS2 that I'm just listing here and you can go to our sandbox we can explore our tools and we have also created some visualization libraries at the facility district province level same as for 3.360 and you can have some inspiration on visualizations if you are missing some including the offline analytics that is available on mobile devices and with that I hand over to Preno for a few minutes for some questions and wrapping up comments thanks for your attention great thank you so much George thank you to Augustine and Barnabas who began the presentations I think we had quite a lot of engagement questions in the chat I think more or less all of them were replied to you can always reach out to us at any point through the community of practice there's a section specific to supply chain and LMIS you can also reach us at LMIS at DHS2.org I don't see any specific question that we need to clarify further beyond what we've written in the chat but I think to highlight maybe from each presentation that Augustine really showed what currently is available in those specific countries that were analyzed and what's possible in terms of improving the data and maybe doing further deep dives to see exactly how the configurations are being done and what's being used I think it was interesting to note that under half the countries actually had dashboards configured a lot of potential to actually make use of that data for Barnabas everything that was presented was native DHS2 using the Nigeria NHMIS instance it didn't go into any data collection procedures or any forms there were no changes made to the underlying structure already existing it was simply building visualizations with existing data and bringing in a few additional data sources so what we're really curious to see is how the operationalization of those dashboards will be happening in Nigeria working closely with the UNICEF team with the National Logistics Working Group and the different teams working on immunization supply chain to see how can they connect to actions that will actually improve stock availability which then I think George really went in detail on how this data can help with both symptomatic and underlying issues and again with the focus on improving vaccine coverage and improving overall health outcomes and I think it's really a lot of good work that we can build on in the future do not hesitate to reach out to us for any more questions and the presentations and recordings will all be shared with the participants also in the community thanks again to the presenters and we'll be in touch with all of you bye for now