 day for you. We're going to first get started with a presentation from myself on the brand new WHO standard data elements and reporting criteria or definitions for supply chain or LMIS data at facility and lower levels. We were actually working on this all the way up until this morning. This is brand new. You're the first folks to ever see it. So that's quite exciting and I'm looking forward to sharing it with you. Then after our break, we are really lucky to have Sacky Boo and Clement joining us from HIST West Africa. They're going to take us through a case study of what they've been able to do to develop a new application and support supply chain reporting in Mali and Burkina Faso. Really a lot of great innovation coming out of that team at HIST West Africa and having some fairly profound impact in supply chain surveillance and reporting in those two countries. After their presentation, they will stay with us and if you would like some additional support, you have questions, maybe you want to even know how that you can use some of the tools and innovations that they're going to be presenting to you in your own country or projects. They will have an experts lounge. We call these expert lounges. These are just opportunities for you to have a conversation with the folks that have the answers. Sacky Boo and Clement will be joining us for an additional hour in the experts lounge. The experts lounge is on Slack. I'll just go ahead and drag Slack so you can see it. We have two Slack channels. We have Expert Lounge for Africa and Expert Lounge for Asia. The Expert Lounge for Africa is today. This is the one that we're talking about now with Clement and Sacky Boo where they will be here. You can type any questions that you have for them and they will be able to respond to you here in the Slack channel. You can of course ask questions at any point here in the Expert Lounge for Africa. You don't have to start necessarily later today. You can go ahead and start putting questions in now and they can be able to get to those right when they get started with the Expert Lounge. If you're joining from Africa or you're just interested in knowing more about what they had to say or understanding more about the tools and innovations that they will be presenting, then feel free to join the Expert Lounge for Africa and get your questions out to Sacky Boo and Clement. I will drag this back over. We do expect possibly some additional program changes maybe for Thursday or Friday. We're still nailing down a couple of presenters but so just hold tight on those but then just kind of looking a little bit at tomorrow as well. Today we are going to go through the data elements like I said here a use case from Molly and Burkina Faso. Then tomorrow we are having our local resident supply chain expert George McGuire. He's going to be taking us through some of the indicators that we can actually build from these data elements and reporting forms that I'm going to be presenting today. So that's kind of going full circle here a little bit and then after George we have another really excellent use case presentation from Moghini in his Malawi and how that they have started to use DHIs too as their main analytics platform for supply chain monitoring. All right. Martin, Alice any announcements that I'm forgetting? No, all good. Okay. So I will start us off just to remember a bit of a warning that we do have a word of the day coming up. The word of the day will actually be at the end of this presentation so you got to sit through the entire thing and then I'll give you the word of the day and you can go in and mark your attendance there. Okay. So let's get started. We are going to be talking about some supply chain data models and some of the data elements and again these are the WHO data elements that have just just been finalized just a matter of a few hours ago. Okay. So what do we want to do? We want to make sure that you are able to understand the LMIS DHIs to system architecture options. There's quite a few of these we'll go over them in a bit of detail but we want to make sure that these are quite clear to you and make some general recommendations on which ones are probably the better just based upon what we've seen through quite a bit of experience but also supporting you know what the DHIs to platform is able to do and what we hope it will be able to do in the future as well. Then we are going to understand some what we call reporting maturity models. You know yesterday I mentioned that we have different countries doing at different stages some are very very basic simple data you know maybe Excel forms or paper records just aggregated monthly or something like that very very simple which is perfectly fine while other countries are much more advanced where they are tracking every individual child and they want to be able to barcode scan every vaccine every child gets to be able to connect those two and have that automatically update the supply chain or this the stock fund hand data and then many countries in between those two extremes so we want to kind of cover those a bit of those different maturity models as we call them and and then let you know a little bit more about what DHIs to is able and is not really able to support currently. And then the last point here is let's just go through the aggregate facility level WHO standard data elements and we have a demo database for this setup as well so I'm going to point you to that and this is going to be a place where you can go in and take a look at it on your own time and actually play around with some of the analytics and some of the indicators as well. Alright so system architecture options and this is actually a slide that I stole from George presentation that he gave last week so George please feel free to chime in if you think I need to clarify something or I'm misspeaking about anything but what we have here are trying to illustrate to you essentially four different system architecture options. I'm going to start at the very top and then go one by one down through the list and let you know kind of why they're showing up here in different colors. So at the very top you see our different levels of our supply chain system as well as typically our HMIS so our health system as well so we have facility level then most countries above facility level have a district level some may have a state or a region or something like that and then above that many countries have a provincial level again some countries may call this like something like regions or areas or something like that and then finally we have a national level we can even say this may even go further to the global level to say something like reporting to a donor like WHO or Global Fund or GAVI or something. So what do we have well again yesterday I presented that we actually have quite a lot of data on what countries currently are operating with we have the global fund assessment and WHO assessment and a couple of GAVI reports that kind of give us a little bit of a clear idea of what countries are currently working with. Some countries and this is the very top one this is the gray arrow double-sided arrow going across they have what we call an ERP an ERP is essentially kind of an end-to-end electronics report electronic requisition and reporting platform you can think about an ERP as what they use in like grocery stores or retail shops where they are scanning and they are and that scan is automatically deducting from a stock balance and then that's going and routing information all the way up through to reorders and acquisitions and that kind of thing. ERPs are typically very very expensive to implement we've heard countries implementing and projects implementing ERPs as efficiently and cheaply as they can for upwards of you know 10 to 15 million euros again these are systems that are typically designed and utilized for large corporations large retailers like large grocery stores that kind of stuff and so they are very expensive sometimes they are flexible depending upon who you are getting to supply your ERP. Other countries Rwanda specifically has reported that their ERP is extremely inflexible they don't get the changes that they want made in it it just depends on which ERP you go with who is providing it and how much money you're paying them so but the long story short is most countries that we're talking about and that that you know we're speaking to right now an ERP is not necessarily a sustainable solution meaning that they they don't have the the resources the human resources their financial resources to be spending 15 million euros a year and it's and the inflexibility of it oftentimes means that they're not actually getting the changes and updates to it that they actually require. Again in the case of Rwanda they've actually dropped their ERP entirely and so they're not going that route anymore so the next level is a fragmented LMIS and so the fragmentation here you see these multiple red arrows we have a different one at each level we have a LMIS A at facility level, LMIS B at district level, LMIS C at provincial and national level now a general LMIS a logistics management information system is designed to oftentimes do most of very a lot of the similar functionality as an ERP but it probably is not as highly transactional and what I mean by that is that it's probably not necessarily incorporating a lot of barcode scanning and doing a lot of automatic or updates of stock tallies. So oftentimes an LMIS but we do see a lot of similar functionalities between an ERP and what we call an LMIS and sometimes the words can be kind of used interchangeably but an LMIS is typically thought of to be something not as complex as an ERP although certainly very still very complex but not as complex as an ERP and not necessarily always for a corporate setting or a business setting but getting back to the level two here this fragmented LMIS what we see in a lot of countries is and I'm pointing this out yesterday is that we they have a different LMIS at different levels so they may have say a paper record at facility level you know is it electronic no but is it certainly still an information system yes so they have like a paper trail at facility level and then that gets put into say maybe an access database at district level and that access database at district level then may be used by a warehousing system at provincial level and then at national level they may have just some kind of like standard report that the warehouse system actually produces this is actually was the case in Malawi for quite a long time where they had different systems at different levels and did the systems always talk to each other no not necessarily some oftentimes what we see is this a very manual data transfer between these two systems the systems are not necessarily connected like you know they're they're not having some kind of interoperability where they're just sharing data automatically that the data has to be manually transferred from one to the other to the other to the other and then what does that mean that typically means that the data flow is very slow that means that folks at the provincial and national levels don't know about issues on supply chain and stockouts at facility level until several months later in the case of one country that I worked in that will will remain nameless they actually didn't know about issues of supply chain until 18 months at national level they didn't know about issues at the supply chain at facility level until 18 months after that initial issue at facility level was reported because the supply chain because the information system was so slow because it was so fragmented and it required so much manual transmission or transferring of data another issue with manual transferring of data of course is it opens up a lot of opportunities for data quality issues every time someone has to manually touch or key in data it means that it's prone to human error and we are all human we make lots of mistakes it's very easy to put in 40 000 when you meant to only put in 4 000 right but that's an order of magnitude difference you know saying that I have a supply of 40 000 it's way different than saying I have a supply of 4 000 and that can throw off national numbers even uh so this fragmented approach is problematic in that it has very slow data also introduces the opportunity for a lot of data quality mistakes there is another kind of fragmentation that I also alluded to yesterday that's worth mentioning now and that is the fragmentation between multiple programs so we often see that different projects and programs will have their own supply chain or LMIS system and you know for example we may have uh I can say again maybe uh use Malawi as an example where the there isn't there's a separate there was a separate system for HIV separate system for malaria and a separate system for immunization well over the course of a lot of a lot of hard work and a lot of good efforts they've been managed to harmonize quite a few of these into one centralized system but you still have a separate system for immunization um and the question is can that also be merged in together into one supply chain system this is not just unique to Malawi this is very very common in many countries um and the reason that this kind of fragmentation begins in the first place is that uh oftentimes different donors multilaterals NGOs and projects they don't have the resources to develop an entire LMIS for all projects maybe they're coming in to just provide ARTs like HIV treatment and that's what they have resources to do they have resources to distribute those drugs and to monitor the use of those drugs and that's so then they what do they do they go and they they look at the national supply chain system and they say well this isn't actually functioning as well as we need it to let's just build our own and oftentimes they just build their own so that it is only monitoring just the drugs that they are or just the diseases that they are working on again like I said yesterday this becomes problematic for countries to usually maintain and it becomes quite unsustainable to have every single program with its own separate supply chain system and it becomes difficult to harmonize because usually they're offered they're often operating off of multiple standards different standards different kinds of indicators different data elements and also becomes extremely difficult for folks to at the facility level and the lower levels to be able to capture all this data because they're going between different reporting forms different apps different processes to report on commodities that are all being stored in the same room right it's all in one medical stores typically but you know each different pro commodity or each different programmatic you know all the hiv commodities all the malaria commodities tb all of these have a different reporting process becomes extremely difficult for folks at facility level becomes a big burden for them to go to work through all these different processes ultimately unsustainable typically for countries to maintain okay so that's with the fragmentation going down one more level we have some countries that are using dh is to as their entire lmis you know from facility level all the way to national level now if you kind of maybe you're picking up on the color coding here a little bit the erp gray is just too expensive most it's not really relevant to most countries the fragmented red that's not good red is not red is bad most countries find this completely unsustainable the using dh is to end to end we've kind of labeled it as yellow yellow is like you know caution why are we saying that well if you have a very simple supply chain system if you're just capturing very basic aggregated data at the lower levels and you just need to see that that that aggregated low data automatically aggregated up the hierarchy and your existing health information system if you're you're not looking at having some kind of parallel system that's just for warehousing and routing and and complex supply chain functionalities and requirements you're just saying i just want to see the basic supply chain data aggregated up through my dh is to instance like i see my health data right very common request especially to many countries that are just getting started with this and is dh is to well suited for that i would say yes dh is to at its kind of its its bread butter its base functionality is to just capture aggregated data or even disaggregated data at the lowest levels and automatically aggregated up through the hierarchy for you so can you use dh is to as an as an end to end uh supply chain system in theory yes as long as you keep the requirements fairly limited um because remember again yesterday we talked about there are some things that a a a functional supply chain system requires a highly functional supply chain system requires warehouse management it requires it typically requires routing and fleet management it requires the generation of pick lists to support kind of warehouse um uh operations and workflows dh is to does not do this you know dh is to is not the tool for those things and if you need those particular tools then dh is to is not going to be able to get you all the way there for a supply chain system at least not yet and so that being said that's why we've labeled this yellow you can keep it simple if you don't have complex supply chain requirements you can use dh is to but as soon as you start to get into things like warehouse management and needing systems for that dh is to stop spinning your it cannot be your go-to tool you need to have something else okay and then we have then the last level which is what we level green this is good this is what we will hopefully want to encourage countries to be able to adopt and this is to have dh is to at the lowest level the health facility level community health worker level maybe small um district hospital level um and have that as your tool that when you capture all the data at those levels and then dh is to we'll probably need to feed that into uh uh uh l o my s that's managing warehouses and larger health facilities or you know like larger like regional or district hospitals um and and um and and that carries it up through the rest of the the hierarchy again if you remember the the example i showed about barundi and what medecis is doing it's not like dh is to just ends at that level at facility level both systems are communicating and sharing data constantly so the folks that are looking at dh is to dashboards are able to see what their supply chain data at district and provincial and national level are but the people who are actually working within the supply chain the folks sitting in the warehouse who are filling out the orders you know uh and shipping those out they're probably interacting with a more specialized supply chain or logistics management information system that has those functionalities and again dh is to currently does not so um so right now as of today we think that this is probably the best option the the most appropriate it does of course bring in some complexity of interoperability but again as i mentioned yesterday we are trying to address some of these interoperability issues bilaterally with the other platforms so uh george i think i would like to invite you to speak up and say anything if you have anything that i missed about this you can just unmute yourself okay thank you very much just a small comment that came up during a meeting we write health facility just for the sake of simplicity but this would apply to any stocks could be a community health worker could be community based based first aid otherwise i have no other comments thank you yeah thanks very good point yeah health facility also yeah it goes down lower it goes down community health worker community health post it can even go lower anyone who's actually managing uh commodity sometimes these are even like mother's groups or traditional health workers or traditional healers even in some countries uh just depends on yeah it can definitely go lower thank you for that point okay moving then right along let's take a quick look at some of the various maturity models for supply chain data capture um again we are building these kinds of maturity models in partnership with close collaboration with gavi and who especially gavi uh who wants to really push the push the limits on um uh some drug monitoring especially well immunization monitoring and being able to connect those immunizations to specific children so that they can monitor adverse events and reactions to to different immunizations if they come up so again we're not just doing this in isolation we're really working with a large team of global experts to try to develop these things all right so we kind of have what we call a three tier approach we have a basic aggregate reporting model and then we're going to go into this one in quite a lot of detail then we have a tracking stock approach which is a little bit more advanced and then the third and much more advanced and this is really what gavi is pushing us for is being able is tracking individual patients giving them medications and then automatically updating the stock data from the medication that we've given to the patients so let me just go into these a little bit more so the the point to make uh we're going to go through the tier one and i'm going to demonstrate it to you here in a minute but the point to be made here is in this tier one which again is very basic just aggregate monthly or weekly data coming in about each one of the commodities we are able to actually calculate some fairly advanced indicators from this on friday we'll be i'll be showing you uh quite technically how to actually do these indicators in dhis too but we can get things like closing balance numbers stock out days adverse consumption stock status so we can see which facilities are stocked out under stocked adequate stocked over stock we can even calculate some resupply if that's appropriate for the commodities and and the and the system of pushing stocks out and we can also calculate some order fill rates so that's the um uh the order requested over the total that was resupplied so how much do we actually get based upon what we requested um so these are actually if you if you're reading between the lines here you can actually see i we put in how these are actually configured in dhis too uh but we're going to go into more detail later i'm going to show you this in just a few minutes as well uh some examples so again tier one just very basic aggregate monthly weekly or some in some countries daily stock reporting uh aggregated data stock tier two is where we start to actually put some data into uh what we call dhis too tracker if you're not a dhis too expert um there are again three modalities at which data can be captured in dhis too three different ways the first one is aggregated data so this is routinely captured data um again you can think about like a monthly health facility report something like that um the second modality is event capture so data that's associated with an an actual single event and these events are not meant to be connected or related to any other events you can think of like a um mass drug administration you can think of like a male circumcision campaign uh an itn distribution an insecticide treated um nets distribution campaign um individual events also like surveys like demographic household surveys and that kind of thing um these are just one-off health interventions that are deployed somewhere um and one event doesn't connect or is not necessarily related to any other events then the third modality of data capture is what we call dhis too tracker or tracker data and this is where we actually monitor something over time right this is the more transactional data so you can think of like a a uh an expecting mother going through antenatal care appointments excuse me or you can think about maybe an hiv patient coming in every every three months uh for uh um testing uh counseling and treatment or like an immunization campaign having a newborn followed through the third the first uh thousand days of life going through and getting you know growth and weight monitoring and their immunizations completed so these are like actually tracking typically an individual person we can extend this to also be tracking some stocks as well it doesn't have to be just tracking people it can be tracking whatever you want and we can actually configure this to be tracking some stocks um and that allows us to start to build in some more advanced functionality um i think one of the better ways to illustrate this is um with uh uh an actual example here so yeah i will yeah let's just skip to here so instead of talking about it in theory i can talk about it in actual practice in kenya about two years ago we were approached by uh university waslo was approached by the jsi johnsono institute about adapting their cstock system that was originally developed in malawi which monitors community health worker supplies uh into dhi's too they wanted something that was a little bit more sustainable than what they had developed in malawi although what is what they're they're still using an in malawi for the record but they wanted something that was a little bit more universal that other countries could adopt and be a little bit more sustainable than how they had developed this system originally in malawi so they came to us and said can we put this in the dhi's too we we worked with them for quite a long time trying to break down the different the data model and understand the requirements and we came up on kind of a two uh a system that works quite well actually um that has two different kind of reporting pathways okay so what happens let's talk let's start at the top and this again is for community health workers so on the 28th of the month the community health worker will report their stock on hand okay and the community health workers in kenya are treating things like uh kind of your your your iccm so your your integrated case care management which is diarrheal diseases malaria um uh fevers and some basic infections they're also trained to pick out warning signs like like a bloody diarrhea or a hemorrhagic malaria and and be able to refer those particular patients to secondary treatments or like health facilities okay so anyways so community health workers have quite a lot of commodities that they carry with them um and on the 28th of the month they were they they report their stock on hand and they do this uh every single month in an aggregate data set okay then dhi's too automatically calculates the average consumption and they calculate the and resupply value the average consumption is calculated over the last three months um uh taking the difference between what they actually received and the end at the beginning of the month their stock on hand at the end of the month factoring in that as the consumption then taking the average of that and then also say this is your average consumption of the last three months this is how much then we think that you need to be resupplied next month based upon your average consumption um so between the 28th and the end of the month the community health worker supervisor which is a person at the health facility they see a dashboard and that dashboard provides them the resupply values for all of the commodities for all of the community health workers okay and then they have about a week about five days typically to fill those orders to make the resupplies to actually have a bag for each community health worker and make sure that community health worker is getting uh adequate number of the commodities for uh uh or added number number of each commodity based upon the resupply resupply value that was calculated um and then on the fifth of the month the community health worker uh comes and actually collects the resupply and they record then how much they actually received so dhi's too is automatically calculating the average consumption and the resupply value and sending it to the person who can provide it and then when the community health worker actually gets that resupply they capture how much they actually received and that actually allows us to know what is the order fill rate what's the difference between what they should have received versus what they actually did receive and that order fill rate becomes very useful in kind of forecasting um down the road uh stock availability low stock or potential stock out situations okay so that's very basic aggregate reporting that's kind of that tier one but then we also built in something that was a little bit more transactional using the tier two tracker reporting um and how this works is if a community health worker has a stock out if they have a stock out or a critical low stock then they go into the tracker and they say I have a stock out or I have a very low stock and that is what we call the enrollment so now I'm looking at the second line here second line here the enrollment and the once they enroll that they have a stock out or a critical low stock that will send an automatic SMS message to their supervisor at the health facility saying hey this guy or a woman is about to run out of stock or they have already run out of stock you need to put an emergency resupply so the community health worker will then in stage one acknowledge that they have um uh received the the um uh the the emergency request for drugs then they will in stage two res fill the order and record that the order is ready that will then send an automatic text message to the community health worker saying your order is ready come pick it up and then the community health worker on the final stage three will come and collect the resupply and record how much they received uh from here we are able to calculate a few key indicators the first one is lead time lead time is the time between uh when the community health worker sends the notification to the supervisor that they have an um they have an emergency uh low stock or a stock out and the and the time that they actually come and collect it we are also able to calculate stock out days so the time at which they report um how many days it is from the time at which they report that they have a stock out all the way until they report that they have uh filled the stock and they've been resupplied um so this is actually using dhi's to tracker it's not using aggregates using tracker uh if you're familiar with tracker you recognize what enrollment stage one stage two stage three these are dhi's to terminology but hopefully all of you can kind of appreciate the workflow here okay then we have stage uh tier tier three tier three again is where we're actually tracking each individual patient and uh recording the drugs that are being given to them and that number from each individual patient is automatically updating the stock tally again this is what gavi uh is is encouraging us to be able to develop and and supply to countries we are still working on this this requires some still requires new functionality uh that we're still developing and um so this is not really quite ready yet but it is kind of a place that we're that we're going uh so if you ask me today can dhi's to do this i think the answer is no but if you ask me this next year the answer might be yes um so it's something that we're currently working on okay so now let's talk about the who data elements and again we have just been finishing these up this morning so you're the first folks to see them what we are talking about here is kind of the lowest common denominator so we're talking about tier one tier one which is basic again routine aggregated data and this routine aggregated data can be captured daily or weekly or monthly okay um and it's captured at hopefully the lowest level so facility level community level and we have a simple reporting form a very simple reporting form uh that works also in the capture app in the on android so what are the data elements that we are recommending if you want to use tier one you want to follow who standards you want to be able to calculate all of those indicators that i pointed out earlier what are the data elements that you need to capture what are the data values you need to capture for each commodity every month this is what we are saying so it's just um six values per commodity okay and the first one is received so how much stock did they actually receive during that reporting period okay sometimes this is referred to as like stock receipt receipts then the next data element is distributed what's the quantity of stock that was distributed as part of the patient services right this can be given this can mean any stocks that were actually given out to patients okay the next one is stocks that are redistributed redistributed means that you have kind of a lateral transfer or a facility facility transfer of stocks so what is the quantity of stocks that are redistributed back into the supply chain so again something that you send out redistributed then you have a discarded the next the fourth point here is discarded so how many stocks did you have to discard that can be caused of spoilage wastage expired any reason that you're just throwing away stocks because of you know some reason it could also potentially in some cases mean pilferage we do know that there's quite a lot of places around the world that have issues with stock pilferage you might be able to consider that also discarded and then you have stock on hand and we recommend that this is captured on the last day of the month what this is is just a physical count of the available stocks that are sitting on the shelf and we recommend that you count to the lowest denominator of the stock so you're not necessarily counting packets or bottles you're counting number of pills right what's the lowest count that you can have and then we have number of stock out days so in the course of that reporting period how many days did you actually have zero stock right and again george you've put a lot of work into dividing these do you have anything to add not at this stage thank you okay great so we recommend that right now as of today these are the only six data elements that you need to that we recommend you capture per commodity now i know some of you are worked a long time in the in the logistic space you might be saying no no wait about these others hopefully we can answer all your questions we certainly appreciate that many countries are capturing much more than this and there is some there's there's quite a lot of justification for that but just to be able to calculate all of the key indicators that we know of that have been given to us from various projects and and and and the multilaterals and donors these are the six data values that need to be captured per commodity so let's take a look at this we don't have to talk about it in theory let's actually just go in and take a look at this i'm going to give a quick demonstration here so let me just exit out and i am going to go to we've set up what we call a sandbox which is a demo instance for this academy and you can go in here and play around with this yourself so it's who-sandbox.dhi2.org backslash lmias so i'm just going to copy this and paste it into my web browser and you'll see that we have provided you a username and password here so you can log in as guest you see that i have a lot of different dhi2 instances that i have the password said for and the password is district with a capital d the number one exclamation point okay when you log in here the first thing that you're going to see is this bcg stock status dashboard this is some of the an example of some of the analytics that we can build from some of the indicators that i mentioned earlier we're going to come back to this on friday we're going to go over these kinds of analytics in a lot more detail on how to calculate these indicators but for now let me just show you how we propose that this data be entered we go to the data entry app okay we have to select our org unit of course i'm just going to go down to the very first health facility and again this is uh this is our training land demo database or demo country this is not a real country this is all demo data and i'm going to go to the data set that's called facility stock report zero uh 2.0 go to october select my month and here you go so we just have a very simple example for just two commodities are really maybe um uh yeah just just two examples here one for the first one is for rdt's so again we capture number received number distributed discarded uh redistributed stock on hand and the number of stock out days and the health facility will just come in and enter this data every month for each one of the commodities as simple as it really can be we've we want to minimize the number of data values that they have to enter to a bare minimum to be able to calculate the key indicators we don't want to capture more than what we actually use and this is kind of a cardinal rule the major rule for um any kind of information system do not capture more than what you actually use if you don't use a a data element or a data value that you're capturing in any kind of indicator don't capture it it's very easy to add indicators and data elements to systems it becomes very difficult to take them away so what I recommend is keep it simple keep it as simple as you possibly can keep it as minimal as you possibly can and so that's why we're making the recommendation for just these six data elements currently to be captured these again these give us all of the data that we need to be able to capture all of the indicators uh that that that who has defined so um that's that's that's really it um you can come in and take a look at this you can enter some data if you want this place is this sandbox here is built for you uh just a fair word of warning you are a super user here so you can't potentially destroy this if you want so I highly recommend well if you do destroy it just let us know in slack and we'll we will reboot it but um if you want to come in here and play around with this and just enter some fake data this is here uh for you and again we're going to come back to this in a lot more detail later in the week when we actually start to build out some of the analytics and indicators okay so I'm going to go back to my powerpoint here Scott there are some comments from the audience that it's hard to read the screen when you're sharing I believe it's from the HS2 I don't think it's possible to zoom in further is there it well it is a little bit possible to zoom in it could also be a bandwidth issue right but you but all the slides the slides are available so if you are having an issue seeing what I'm presenting just go in get the slides and then you can go to the demo database yourself you're just going to go into data entry and you're going to you're going to select that facility in stocks data set number two I can probably put a little bit more information in the slides on that as well okay so we just have 10 more minutes so I'm just going to kind of try to get ahead of a few questions here as well so what can we also do with these basic data elements or these basic data elements whose data values that we're capturing in that reporting form that you just saw we can capture or we can calculate very easily a few things and usually and sometimes these are actually values that are captured that that are often captured in the same reporting form but we're saying that we can calculate these values because there's because they can be calculated based upon the the six data elements that I just showed you they don't actually have to be captured themselves so for example closing balance DHIS2 can calculate a closing balance and the closing balance can be the opening balance plus received minus issued minus redistributed minus discarded why why have someone put in a closing balance when DHIS2 can calculate it for you now of course it could be necessary to put in a if if the calculated closing balance is different than the than the actual closing balance but again what we're calling the closing balance in those six data elements is the stock on hand so but if you want a closing balance indicator DHIS2 can calculate that for you there's no reason to ask that a question again we also can do unaccounted for stock loss so we can look at the discrepancy between the physical stock on hand and the calculated closing balance right and this is actually to appreciate if there was any kind of issues in if the you know it can it's often more of a data quality check it helps us identify if there's any missing stocks and so maybe some stocks have been lost and again so DHIS2 can calculate the closing balance we're actually asking for the stock on hand which is essentially a closing balance that's a physical count and then we can take the difference and measure any kind of unaccounted stock loss the opening balance is also something that DHIS2 can automatically calculate may seem a little counterintuitive but really the opening balance at the end of one month or sorry the the stock on hand at the end of one month should just be the opening balance next month right how do we actually get this data to transfer how do we get that stock on hand to just move over to being the opening balance in the next month we can actually use something called a predictor for that and that's a tool that we have in DHIS2 to move data across multiple periods or to calculate values across from data across multiple periods so we can and we're going to be talking about that on Friday in quite a lot of detail so you can get something like opening balance if this is an indicator or a value that you absolutely need in your country you can still get that value from the six data elements that I showed you you just have to configure a predictor to produce it for you so why are we not you know most many countries are capturing or want to capture a closing balance in the reporting form well in essence we still are because we're capturing the stock on hand but we're not calling it a closing balance because a few reasons and and again George I pulled these reasons from what you've said over the course of this quite a few conversations so please chime in here but essentially we it's it's highly advisable that the folks that are operating or working at health facilities the lowest levels they should just be counting the values that they have on hand we do not necessarily want to show them what their calculated closing balance should be and the reason for that is that it becomes something that they feel that they have to match if DHIS2 was showing them their their calculated closing balance which it is capable of doing of course then they there would be some pressure for them to try to match the numbers and we don't necessarily want them to do that well not even necessarily we don't want them to do that we want them to just say it's the last data month how much do I have of each commodity just count not don't worry about what the the the calculated closing balance should be and again if we have these two values in the system the calculated closing balance and the stock on hand report then we're able to calculate the difference and we're able to say how accurate is the data data quality checks ideally they should be a one to one meaning that the stock on hand and the closing balance should be the same number ideally so but if they're not then we can build out different data quality checks and alerts and notifications and indicators to help us to to to alert us to this situation so we can use like validation rules we can build out the the the different types of indicators and we can actually visualize these on dashboards in charts and that kind of stuff so we can so that we could maybe like you know pinpoint the facilities that have a difference between a stock on hand and a closing balance George anything that you wanted to potentially chime in and add to this thank you so I think this is quite amazing functionality that was in principle a simple tool you can have this level of checks and make sure that all the the stock figures add up I think it would also be it's also important for storekeepers you know that missing stocks is a very sensitive issue because you could be quickly accused of mismanagement or even stealing I think it will be also good to document when you find discrepancy so there's a clear audit trail of any discrepancy at the end of the month thank you right that's a really good point so actually in DHS2 we can see an audit trail of any time a data value is changed so we are able to actually go in and say okay so this was the original stock on hand reported and then see if it was ever changed after the first value was recorded and and monitor that change over time and and and potentially investigate it if there's any kinds of suspicion or or necessity there so a couple of optional data elements that some programs within the WHO kind of asked for but we haven't necessarily included we're not saying that these couldn't be added but we just don't see them as completely necessary now but you may want to use them yourself you could potentially include a yes no answer for a stock out so you could you know we're actually able to calculate currently from those six data elements stock out in two different ways we could say have they ever recorded any stock out days you know if the number of stock out days is greater is one or greater then there was clearly a stock out there and so then there was then we can make a notification or visualize that on a map saying this is where the health facilities that has stockouts we could also say that if the the stock on hand is zero then they have a stock out as well right so there's kind of already two ways that we could do this the malaria program in the WHO thought about it might be useful to have a yes no answer so if you're stocked out yes then that could be something that we could monitor and count as well it's a little bit redundant and that we're already actually able to get that data from two different sources but but it could be something that you could consider you know workflow wise it makes it very simple of saying am I stocked out yes and you don't have to enter the rest of the data also ordered so the WHO TB program wanted to include a a quantity of stock items that have been ordered during the reporting period different countries do different things some have a push system where health facilities don't have to order anything other countries do have a system where health facilities actually have to order how much they want because of this difference it really hasn't we haven't really put in this ordered value because it's not necessarily a universal thing so and it may not even actually be that useful because oftentimes what they order is not actually what they're supplied so so we can have we have excluded it currently um George anything to add about these yes thanks a lot so just a quick comment on on stock out so I've been working as a humanitarian medical logistics station for 25 years but this is really a dream come true because we routinely monitor stock availability in the upstream stocks but of but we have never been connected to the health facilities until now so that's really great and just one comment as you have seen in the basic model this is based on monthly reporting in the with a tracker you could consider having like real-time data but you can combine those two in the sense that you could have a system where even though we are of course not expecting anybody to count the stocks every day or every week you could still go into DHS to when you actually encounter stock out hopefully that's not very often on the 50 or the 15th day just just change either the indicator or the stock and then have an upstream dashboard so that the the logisticians who are supplying your health facility they will not have to wait until the end of the month in order to detect the stock out so they could have like a real-time monitoring let's say a daily monitoring of stock outs and react very quickly so this would be really a great improvement in service levels where basically the health facility doesn't have to you know send an email or call or somehow notify but ideally somebody would a planner would be like screening for stock outs on a daily basis and picking up whenever there is a change that needs to be addressed immediately thank you yeah that's a really good point George this data can be even though you know we're saying it could be monthly you could capture this data weekly or actually in some countries they they capture this data daily so if there's that zero put it in at any point you can DHS to can detect it and send alerts or flag it on a dashboard and again we'll show you how to actually produce those kinds of analytics on Friday okay so that is it for the first session for the day and so now it is time for our word of the day or in this case we have words three words of the day and the words of the day are I love DHS to so even if you don't love DHS to you still have to type it out and and put it into the attendance so please make sure that you are recording yourself as being here in the attendance form that's on the google drive and that's I love DHS to is the words of the day and you need to put that in in order to be counted as attending and you have to have your attendance to be able to get your certificate so please go in to the attendance on the google drive and I did the link as well in the chat link is in the chat great and record yourself as being here all right so that gets us the end of the first session after this we are going to have a use case presentation from his west africa sakebu and clement will come and join us so now it is time for our break we will take a break for until 1215 oslo time so that's just 12 minutes from now 1215 oslo time we will come back and have a use case presentation from his west africa so take you know refill your coffee cup grab a cup of tea take a bathroom break and we'll see you back here at 1215 oslo time