 My name is George McWire and I've been working as a Humanity and Health Care logistician for the last 25 years and very honored to join a vibrant and innovative team in Oslo as an LMIS technical advisor. So today I will build on what Scott started introducing yesterday on the WHO meta package and go into some more details on the LMIS. Just to be sure, logistics is my specialty so I usually refer to LMIS but of course we are always talking about HMIS and LMIS integration and logistic indicators concept for end users so I will often use the term health facility but in fact the concept is designed for any kind of end user as Scott mentioned yesterday could be a community health worker, could be a traditional healer or birth attendant or anyone who is using, storing, administering dispensing medical supplies to patients. So I will present some general principles some of which Scott already touched on yesterday which are rather abstract but the details will become more clear when I present the actual indicators. So one of the general principles is all the data should be entered directly into DHIS2 so I think this was mentioned during the presentations yesterday that one of the constraints in the current LMIS is that they are often paper based and then reports have to be sent to the district level where they then entered and there's long delays Scott mentioned yesterday up to 18 months for the data and reports to GRIPLOPS so entering directly into DHIS2 ideally on a mobile device allows to share all the data in real time. All the metrics are eventually all the metrics to be measured directly in DHIS2 so basically storekeepers or health staff or logisticians enter data and the calculations are then done automatically in DHIS2 so that there's no need to perform any calculations with a calculator or in a separate document. What we noticed during our extensive work with mainly with the World Health Organization on developing this this framework is that there's often confusion on terminology so we distinguish between logistics data and actual performance indicators which I will explain in a minute. It's just I think useful to keep these two terms apart conceptually. And the health facilities only measure metrics at the facility level so Scott explained this in detail we don't want to develop DHIS2 into an enterprise resource planning tool or a tool that covers the entire supply chain so basically the measurement the data recording should be limited to data that is actually collected at the health facility at the end user level and that is not available upstream. Ideally all the data if you are integrating DHIS2 with an LMIS will be available in that system and then if you collected a second time at the health facility level publication that's elite time normally that should be routinely measured by any upstream LMIS. One thing that we noted there was quite a lot of duplications different formulations different calculations of indicators so it's important to have a single indicator for measuring a specific for specific measurement let's say shortages so stock outs there's many variations to avoid redundancy because otherwise you risk measuring five or six different indicators but at the end the their meaning is or the consequences of action is basically the same. This Scott already mentioned I think this is an important point that we we're striving to standardize the concept and the indicators across all items commodities usually cost of health care goods and across health facility or also programs so currently different programs have different ways of measuring indicators maybe with a similar result a similar meaning and it would of course make it much easier for for everyone if there could be an agreement on some standardized indicators and metrics while you're still free to develop specific indicators if that is needed I will come to that later. We want to keep the simple the system as simple as possible so you will see that we basically cook down a large list of indicators to a very few it's also important for the performance management in general to realize that of course logistics is the main actor in providing logistic services in terms of making sure that all the goods are available at health facility but all health actors all health staff have an impact on logistics performance and can contribute to improving it but it's also important to have a clear separation of responsibilities and actors of uh accountabilities and actors in terms of who measures which indicator and who takes which action on what indicators so we'll come back to that in more detail on my second presentation next week and also to keep in mind that we all know that we do a lot of reporting and that will not go away but ultimately the objective of the performance measurement and of implementing or expanding LMIIS is to improve health services to patients and not only to generate more and more sophisticated reports so we want to maximize stock availability that's basically the ultimate objective of any logistic service to make sure that all the items that are needed at health facility community level are available at all times so we have basically two concepts a basic and advanced way of measuring recording data and measuring indicators so the basic package is the WHO metadata package that Scott already showed yesterday and that is already available or will be available very shortly and in the basic package you enter aggregate quantities by item so by stock keeping unit and that is usually done on a monthly basis at the end of the month so you record your stock on hand or your stock issues once a month and all the data is logistics data is entered manually that means you have a record or you physically count your stock and then you enter the numbers into ideally a mobile device directly in DHS too and for the I will explain the distinction of manifestation root cause indicators in a minute so the indicators like the stock out days and the discarded stock all those have to be entered manually so all the data recording data entry will be done manually I lean in a mobile device either directly or from from paper records that will still be necessary but we also already building for the advanced system that will also be possible thanks to the advanced functionality of LMIS and which we have tested in Yemen I will come back to that later so that's not too far future it is in principle already possible but it still has to be built out and refined in this advanced package it will be possible to manage medical products at the product level with individual batches and even with serialized numbers if you want to do that Scott already mentioned that there's a push to record vaccines at the batch level or even at the serialized number level for vaccination campaigns and the other feature of the advanced system is that the logistics data will be brought into the system exclusively or almost exclusively by digital entry so that means by using a barcode scanner for all the stock issues for example and then ideally the receipt of stocks rather than having to type it it could be entered into the DHS2 by electronic data interchange that means by an electronic file that is downloaded or uploaded from the upstream LMIS directly in DHS2 of course you still have to check your stock but once you confirm that it is complete basically press a button and all your stock is added to the DHS2 database so that you don't have to enter this manually if you are managing let's say ARBs or malaria stocks or vaccine stocks and you have five or 10 or 15 items the manual entry works well but if you have a larger facility and you're managing let's say 250 300 or 400 items in the hospital then the manual entry is quite cumbersome and prone to error and since all the data is brought into the system automatically that will also then allow to calculate all the indicators fully automatically and create and create a dashboard and of course if you enter the data as you receive goods or as you issue the goods you can also have real-time data so this is not monthly reporting but at any time you can go into the system and you can check your stocks what is important is that we are planning to have both options so every no one is forced to use advanced system soon or everyone can basically decide how you want to progress from the basic system once it's implemented to an advanced system and whether you want to do that for only some facilities or all facilities or only certain places so we are aware that there are different needs people are moving at different speeds and we will accommodate that by building a system that allows to do any of those implement any of those two modes and switch between them at any time so the overview of the framework as I mentioned we distinguish between logistics data and indicator so logistics data can further be differentiated into stock management and stock replenishment so those those are just numbers statistics and facts and then for the indicators I realized that in DHS to indicator has a particular meaning we differentiate between manifestation indicators which basically indicate the symptom I will explain in a minute and those are fully standardized and then the root cause indicators which try to identify the root cause of a problem of the symptom which will allow to correct it and those are derived so if you have a certain manifestation indicator like stock out then of course you have only certain causes that can cause a stock out so you're not free to choose your root cause indicators anymore and then you will have program it will be the possibility always remains to have program specific indicators so we try to standardize but nobody is compelled to limit the indicators to only the few that I will present in a moment so there are certain needs for programs to have specific indicators and those can still be customized to individual countries or national protocols or individual programs as needed but we still hope that the standardized basic indicators will be used by everyone so that you have a shared basis across programs and items and the issue of terminology will come back it's coming back again and again so we are still working on double actual to complete the the terminology so if you have any comments on this I highly appreciate it so logistics data as I mentioned those are just facts those are just numbers they are indispensable for any LMIS so I'm not saying that because they are not indicators in the strict sense that they are not needed then on the contrary they absolutely need it but they're not performance indicators so if we take for example the stock on hand stock on hand is just a number so if I tell you that I have 15,000 tablets of paracetamol in stock it doesn't give me an indication on whether I'm managing my stock well or badly you could argue that if you have no stock that indicates that you have a stock out but only if you actually want to keep this item in stock so those are just numbers that basically have to be interpreted in order to have a meaning or to derive an action from that so those numbers don't give as such don't give an indication on the performance or the quality of logistics services that are provided and those in the basic system they're all entered manually or they could be calculated so I'm presenting the the logistics data in detail so that you which you have already seen from the screenshot from Scott yesterday from the DHS to Sandbox to which you have access so the opening balance is a stock on hand from last day of the previous month and this correct on the first day of the month then you will add the stocks received so those all the quantities that you have received at your into your stock usually from upstream logistics services from a from a medical warehouse but it could also be redistributed from another health facility and the stocks distributed those are all the stocks all the quantities that are leaving your stock so they could either be given directly to a patient like in case of a vaccination campaign or they could be distributed to a ward or service such as such as in a hospital then it's inevitable to have some damage stock from time to time which has to be discarded so it could be either expired in stock or this damage because it was dropped or it was damaged because the cold chain was not maintained and those stocks then have to be discarded and have to be recorded of course then this was already discussed there's a possibility that a redistribute stock could be sent to another facility or returned to the sender and then eventually I will come back to that later in more detail on the second presentation next week that ideally once you have recorded all these numbers you will have a correct stock on hand count but you may have discrepancies so you could have stock that is unaccounted for unaccounted losses let's say pilferage or a mistake was made that you cannot identify and usually it's a loss but you also have real cases where actually you find that you have more in stock that you should have because there was a calculation mistake or because more was shipped and indicated in the packaging and at the end of the month you will once you have reconciled major stock correction then your stock on hand should be the same as the closing balance and that becomes your opening balance for the for the next month I think these are numbers that we have all used that are used everywhere it's just a matter of like harmonizing and eventually agreeing on the terminology so this is the screen that Scott has shown already if you yesterday if you go on the DHS to sandbox facility stock reports too then you will see receive distributed discarded redistributed stock on hand and then there is one indicator will come to that later that needs to be entered manually that is the stock out days so I will just briefly touch on the second category the inventory control which I don't want to go in detail today but we are fully aware that of course in order to have stocks at the health care facility you have stock replenishment and ordering is absolutely critical so there's different way of doing it it could be done at the health facility usually it could be calculated at the district level there's different variations that's why it's quite the complex issue so this includes certain parameters and settings eventually the inventory control the stock replenishment will be determined by national policies often there's guidelines how often to deliver having minimum order quantities and so on ultimately it will be a management decision in the country or in the health facility how which policy to adapt adapt but this is also recurring theme that it is important to measure logistics performance especially the stock availability in the stock outbreaks but at the end of the day if stock outs persist then the way you replenish your stock is absolutely the key issue on ensuring that your stock availability increases so we have this is not part of the first of your package but this will be a part of a wider discussion next year also depends on how you want to use THAS to in conjunction with other upstream LMIS systems as Scott pointed out yesterday so that's a more complicated discussion I'm just listing the elements that you need to have for any inventory control system at the bottom but of course I think this is is an excellent strategy you can't manage stocks without having the basic logistics data so you're not able to calculate your order unless you know your stock on hand and your your issued stock and your stock received so the first step is to implement the first WHO package which will basically lay the foundation on which then the on which then the advanced packages can be built or which will allow to integrate THAS to which other with an upstream LMIS which then will provide those additional data settings parameters and and data so as I mentioned earlier we distinguish between manifestation indicators and root cause indicators is maybe a bit of a strange name if you have an alternative title to propose most welcome so the way I look at this is basically a symptom for example a stock out so the purpose of the manifestation indicator is to monitor a certain aspect of logistics services for example stock availability which is of course the obsession of all the logisticians to determine what is the quality of services so obviously if I have no stock outs I have an excellent service what is important for manifestation indicators to have a clearly defined target range it's very difficult to maintain 100% stock availability you can always strive for continuous improvement but you should there's a need to define in a policy that you're aiming let's say for ideally 97 98 if you can 99% stock availability because you need to compare the result of your monitoring with the target range in order to determine whether you need to take any action or not the manifestation indicator so basically answers two questions is your problem do we have a problem and how big is the problem so if you have if you have if every item was out of stock during the past month obviously the problem is much bigger than if you had stock outs only on a single item so it allows you to have an idea of how urgent it is how urgent the issue needs to be addressed and the importance of having the target range and the monitoring is that if you determine that the the measurement result of your manifestation indicators below or outside the target range then you need to find out what the root cause of the problem is so the manifestation indicators we have I mentioned earlier we want to have a very simple system we have basically determined only two that you will have in the first WHO package that is stock outs usually measure this number of days of stock out I put shortages in brackets because a stock out is just an extreme way of having a shortage you could have just a few tablets left and it's a near stock out so that's also a problem but stock out is what is easy to measure and what is very visible and then the second manifestation indicator is discarded stock which are stock losses again if you have a large number of big quantities of stock that are lost that have to be discarded it still doesn't tell you why you have these losses you just know for a fact that those goods have to be discarded have to be disposed of and are not available in stock which is a problem but then you still have to go back and to find out what the reasons are and then as I mentioned earlier you could have program specific indicators will come to that later because as always DHS2 wants to keep the flexibility we don't want to lock anybody into these two manifestation indicators but we believe that basically everyone will be interested in these two indicators and you will measure them and then in addition if there's program specific indicators who can still be added so I think it's important that we are all in logistics familiar with with measuring and reporting but there's a risk that you submit reports and you monitor but at the end of the day the logistics services are not improving so the measuring in itself is almost useless if you can't or if you don't want to improve services so basically this is just to make the point if you don't want to improve services don't measure it's like carrying out a diagnosis on a patient where you know that you don't have a treatment for them so there's there's limited value in doing that so that's important in that's kind of one of the fundamental underlying principles of the of the performance management concept is that the ultimate purpose of performance management is to improve services and not only to generate reports that make donors happy or other people happy so the root cause indicators are the second group and so the purpose of the root cause indicator is basically the diagnosis so you have a symptom like somebody has a fever and now of course if you determine that the patient has a fever you want to know what is the reason for that fever so the purpose is to determine the the causes the cause or if there's several causes of the problem so that you can take some corrective action. Manifestation indicators determine root causes so what it means is that the root cause indicators is not something that you can choose freely I mean you could choose that you want to measure stock outs or you don't want to measure them but if you want to determine the root cause for stock outs there's not that many and therefore the root cause indicators follow from the manifestation indicator so we'll explain in a minute so a certain symptom like a stock out can have several root causes could have one or several root causes but as I mentioned already we want to have only one indicator for every root cause so we don't want to have multiple measurements that are measuring the same problem or same phenomenon. We only want to measure something that can be changed where you have can take a management decision and an action to actually improve to correct the root cause and therefore improve the services. Same for as for the manifestation indicators you'd have defined thresholds for taking action maybe you have different thresholds so you could say that if you have one or two stock outs one or two items are out of stock out of 10 items or 20 items you need to look into it but of course if you have if every item was out of stocked in the last month then you have to take urgent action so ideally you would have a kind of action plan determining the urgency and the type of action. It's also important for indicators to be measured by one actor so to be very clear on who is doing what who is responsible for measuring and reporting and even more important is to have specific actors taking corrective action to clearly define that otherwise you risk the blame game that people are just blaming each other and as in the end nothing changes as I mentioned often there are several actors which impact the quality of logistic services and everyone needs to improve on her his side in order to improve the overall performance and again the mantra is fix the problem don't only measure but fix the problem and the ultimate objective is to improve health services by maximizing stock availability so on the right side you can see the the root causes for the manifestation indicators come back to that also next week for stock outs there's not that many reasons surprisingly when you start to think about that there's not like dozens of reasons why you have stock outs so you could have been adding new items that were not used until there because until then because the protocols which were changed obviously above for a safety stock scoop is set to law where you have not used an impris system we're not going to that today but let's say that the inventory control system or the stock replenishment system is not set up well and if orders are made irregularly and likewise if you have to discard stock if you have losses there's not that many reasons why that would happen so you could be overstocked you just have too much stock with relatively short expiry dates you might not be respecting the first expire first out policy and having correct levels of stock but still stock expiring because they're on the back on the shelf and they're not properly rotated or you could have damage but poor storage conditions so typically by damage by if the cold chain is not correctly managed then as mentioned you could have program specific indicators which we have been discussing with WHO with the various programs so for example in EPI you need to monitor open closed while wastage which is a mandatory reporting requirement and again DHS2 is not blocking from adding additional indicators to the to the standard indicators HIV uses multi-month dispensing so the number of patients receiving three months of supplies of ERV is doing the visit and typical TB program they have also special requirements to to monitor TB tests okay that concludes my presentation leave the floor open were there any questions in the meantime yes we do have a couple of questions from Slack first of all if you look at the technical support channel there's a questionnaire from Tantilly are you able to see that George okay read it up hello team would it be possible to integrate the WHO LMIS package basic or advanced within an existing DHS2 instance with already the existing needed data elements thanks okay I think Scott can answer that question okay yeah so it's a good question I think that none of us you know if you're at this academy you probably already have some DHS2 deployment in your country you may already have even some supply chain monitoring going on in that I think it's would it is actually quite common that countries will adapt the WHO standards packages into their existing systems so that you are able to take what you already have maybe modified or tweak it a little bit to to the WHO standards and then also more importantly start to be able to calculate these indicators that George has been talking about yeah so I think it is that's kind of the standard practice to be honest it's very rare that countries have no DHS2 setup already and they're starting from scratch thank you Scott there's also one question that asks about logistic channel from Sophie what is the stock outs are defined by WHO is there any differentiates between STPs or warehouse okay we're working on the on the final wording but basically we are defining us suggesting to define a stock out the stock out occurs if if an item is unavailable on the shelf for any period of time doing a certain day so regardless of whether you had a basically in theory a stock out for one hour or for 23 hours then that is one stock out day and if you had a stock out of a drop product on Monday and you receive a replenishment on Tuesday then basically that would constitute two stock out days but if you have any other suggestion I would be very keen on that and I think that for for warehouse or for a service delivery point the the definition will be the same so we would also like to to harmonize the use of indicators across facilities so if you have another definition of a stock out and if you can just drop me a line I will appreciate thank you George and yeah questions are running in now is there also an Android app for the LMS so I think that there the answer is yes in that there is an Android app a generic Android app to be able to capture data into DHS to it's called DHS to capture you can go to the Google Play Store or onto GitHub which is where you're going to find the most updated version naturally and get the application now is the app this is something we keep coming back to I think folks are struggling to understand this a little bit is the application specifically for LMS the answer is no nothing in DHS to is hard coded for any specific purpose DHS to is a generic platform you have to configure it to be able to capture any kind of data whether supply chain data or whether it's immunization data or whatever right and so what we're saying is that we are developing the packages with WHO of already configured metadata now metadata is the data about your data it tells is defining the what the when and the where components of of your data that's captured so you have to define the metadata WHO is defining we're working with WHO to pre define that for you so that if you want to use DHS to for supply chain then you can grab this preconfigured metadata approved by WHO install it into your existing DHS to deployment and then you will have the basic reporting system that George and I have been talking about you can then of course use the Android application to start to capture that data at the lowest levels and to point out that the Android application also captures data offline so once the user logs in of course they have to log in online they have to then also be online to pull in the configuration but once they are once they have loaded the configuration into the application of the various reporting forms then they can enter that data offline but then you have to remember that that data has to be uploaded at some point there's no magic here they have to eventually re-establish mobile data or a wi-fi connection or something for that data to be transferred off of the phone into DHS to I just want to chip in that I'm working on the on the project in my organization for the ISRC to integrate DHS to with Oracle I will come back to that later but this is I think this is absolutely critical to have the possibility to use a mobile device and to use the the app if that had not been available in DHS to we will not be looking at the tool and I just confirmed yesterday that on my tablet PC know the the data entry and the sharing of data with the database works perfectly I think that's one of the great strengths Monica thanks for insisting on the question I understand what you mean I didn't want to go into that but for our project we have also discussions in principle you could use the application even at the word level let's say you could have many you could manage the stocks at the operating theater level and at the pharmacy level so we are looking into that into that possibility but in principle the the current concept that I presented is based on the assumption that every health facility has a central pharmacy that is supplied directly from an upstream medical warehouse and that that central pharmacy then will distribute goods to to the services thank you thank you George that's all the questions I can see for now but I don't hesitate to ask more I think well I think I think Monica sorry Monica is also asking a little bit more in saying does the location does the stock out definition factor in the location either at the facility store at the service delivery point is there a situation where the store has no stock while service delivery units still have available stock yes sorry for me sorry for missing that exactly yeah I think that's it's not part of of the concept it's a matter then of how you would manage in my opinion if you are if you are in charge of a central pharmacy and you have a stock out then that is kind of that is by definition of an emergency now you could argue that if all the words have stocks then the patients are not suffering that is true but if you have a stock out in in a central pharmacy and the the words or the services the service delivery points are not overstocked then that still clearly indicates that you have a management problem because you should not be running out of stock I would say as a logistician you're lucky because the patients are still being treated but it's still urgent to remedy the situation but it's a valid point you could argue that instead of counting only the stock at the central pharmacy you should include the entire stocks in the health facility but on a practical level that can be quite challenging in a big hospital where you might have 40 or 50 different services where you would have to all count and aggregate so if you wanted to go to this service delivery point level then I think one option would be to actually record the stocks in in the at the service delivery points for example in the world and we're actually looking into that whether we could have a system where the central pharmacy would have an automated system of replenishing the operating theater for example based on their on the stock reports thank you. The last thing that I'll add to that is that if you are capturing the stock out data at any individual location then of course in DHIS2 analytics you are able to visualize that stock out at that location so you can see it on a map you can see it on a chart a pivot table and then we're going to see that I think in the Malawi use case later today where they're able to see in fairly you know close to real time where the current stock outs are or at least where the stock outs were last month for their various commodities on a map specifically. Okay I think that Monika is touching on a really important point that the framework is basically providing a tool but there's quite there's some management issues behind that I mean I fully agree that if you have stocks in a world then technically you could say the health facility has not stocked out but just to to make the point if you have a big hospital like we are supplying in in south of Afghanistan with 50 services and you a single farm a single ward still has you know 20 tablets of paracetamol you could argue that there's no stock out and there's no problem but then of course some patients are still suffering and you would have to redistribute the stock so I think the strategy is to make sure that all the stocks are all the wards of service delivery points are replenished in a systematic way they have the stock levels that they need and then the stock out in the pharmacy will reflect that there's a relative shortage in the hospital that needs to be addressed thank you okay Scott I think it's over to you now I don't see any more questions don't forget the minty meter yeah we are yeah are you ready for that yeah let's go ahead and do the minty meter and I can also do the word of the day after we do the minty meter okay just give me one second so that I can share my screen here it is can you all see my screen yes great so you can go on minty.com and use the following code 6917968 let's give a few seconds to participants to join and then we'll start okay let's start first question are you currently recording the following regardless of frequency and whether manually on paper or electronically stock received stock distributed stock discarded stock redistributed stock correction stock on hand so yes or no George do you see the results we have so far yes thank you yes a few seconds again obviously there are no correct or wrong answers for this one so 65 participants are recording stock received 61 stock distributed 60 so stock discarded 51 participants recording stock redistributed 43 participants do not record stock correction and 57 participants are recording the stock on hand then next question are you currently using dhs2 for recording stock data stock replenishment data logistic indicators just a few seconds left 44 participants so far are using dhs2 for recording stock data 64 participants using dhs2 for recording stock replenishment and 50 participants were using dhs2 for recording logistics indicators let's move to the next question are you currently measuring the following indicators stock out days or similar indicator stocks discarded because of expiry or damage okay we have 56 participants um measuring stock out days and 50 participants currently measuring the stocks discarded let's move to the next and final question what is your first impression on the presented concept ready to adapt usable but requires modifications total crap try again so we can say that measure it to the participants think that the concept is usable but requires modifications yeah Alice I was one of the folks who voted total crap just because I thought it was funny but um so I'm not surprised yeah right um I think that what's important to understand is as folks are looking at this and there seems like there's a lot of logistics experts attending this academy and like you know real experts so I we want your feedback this is still very new just I was saying yesterday is this list of data elements and indicators was just finalized yesterday morning um this is all brand new uh very much still ongoing process with WHO tell us what you want tell us how you think it should be modified we can do it so uh this is meant to be used useful for you and if it needs modification to be useful for you then we will make those modifications so please don't hesitate to communicate how you think the modifications need to be done and be very specific tell us exactly what you think needs to happen and we can do it all right thank you for the quiz is it time for the word of the day yes all righty folks the word of the day is well the words of the day are I think to myself what a wonderful world I think in this time of COVID we all need to be reminded and think about what is wonderful Alice was quite insistent that this be the word of the day yes sorry for the select notification so please in your attendance and again we use this word of the day to make sure that you were here you were paying attention uh and this is how you get your certificates so uh please go in fill out the attendance and you are going to type I think to myself what a wonderful world and it is a wonderful world despite all the news what we are now about to go into break um George there was one question from Sophie in the ask about logistics they are asking if we could explain a little bit more about the district level logistics perspective so maybe that's something that you could respond to via type during the break other than that again please go through the and fill out your attendance and we will take a break now a little more than 15 minutes we will come back uh at 1215 so 1215 the Oslo time please come back and we will see you on the other side of the break