 So now I'm going to go on to the e-registry, which is the most complex form of a track implementation. Some of you might be familiar with the immunization e-registry, which is, I think, the first one that we at the NIPH worked with UIO to produce, and that is probably the first e-registry that's now available well, you know, through the HIST due to any country that wants to implement. So an e-registry involves electronic data entry, right, of course, as a tracker, and we usually prefer that it is a point of care during consultation, and you have to enter raw data. For example, you say temperature instead of saying fever, right? We want the system to determine what is fever because the the categorizations of things can change from, you know, when they change the clinical protocols from one to another. For example, maybe some t-count for where HIV was initially categorized one way and now it's categorized the other way. Now, if you want to go back to your start-call data and recalculate how many people were in one category for another, you can't do that if you've been, you know, just putting in the category. So what we do for e-registry is you enter the raw data, whether it's temperature or blood pressure or whatever, and then the system will then say this person has fever, according to the guidelines of your particular context. In longitudinal formats, which is we are following a uniquely tracked entity, a uniquely identified track entity, longitudinally, right, and so the data is structured so that we can see that person's every visit that the person has gone to and it's all linked for you, not like the events data that we have where it's just what happened today, how many people showed up today. You can't see, you can't connect the person showed up today with that same person showing up another time and the tracked entity has been uniquely identified. If your context does not have a national ID system, you can have the option of tracker generating a uniquely unique ID for you, and you can use that to track the entity. In the Bangladesh implementation that I worked on, we also use biometrics, palm biometrics, and it, you know, one of our colleagues in Bangladesh was actually able to integrate it into the DHIs2 app, the tracker app, and so when the client comes and she just raises her palm, we scan it, and then it just brings up her record, and that really was helpful because, you know, in a lot of places you have, you know, people don't know their birth dates, there is no national ID, even if you use the name, the name can be spelled differently because in Bangladesh, you're going from Bengali, which is a completely different alphabet to English, and so somebody might translate it into English one way and another person can translate it into English another way. So we needed a way of uniquely identifying people and we used biometrics. An e-registry has to be for a specific health area, so this is not, you know, for outpatient data. You can have people coming in, somebody is, it's a child who's coming for immunization, another person is a pregnant woman coming, somebody has a headache, but it's not pregnant, so, you know, another lesson who is coming for HIV care, you can put all their data into the same system, into the same program because the way an e-registry works is we have protocols, right, which are based on the country's protocols for pregnancy, for antenatal care, for immunization, for child health, for HIV, and whatever information that you enter, we will then give the health worker how to proceed based on the fact that you're treating these HIV patients. If an e-registry patient comes in with fever, the recommendation that you will give for how to treat the person is different from a child coming in with fever, so if you put all the data, you know, outpatient data into the same system, the system will know what to do, we don't know what guidance to give, right, so what we do is you can have all the different programs in the same instance, but it has to be a different program, you have to have a program for antenatal care, you have a separate program for child health, for vaccination, for HIV, for TV, etc. So that the digital health interventions are done appropriately. The e-registry also has client data-driven digital health interventions, and we refer to the digital health intervention nomenclature from the WHO. Unfortunately, I forgot to add that here, but if you just Google it, WHO digital health interventions, there are different categories of digital health interventions, and they tell you in great detail, you know, for each category of what options there are. And so with an e-registry, we make sure that we have digital health interventions, which are driven by the client's data, because our focus is giving very specific information targeted to you, the client, to make sure that you actually behave the way that is healthy for you, for your particular health concern, whether it's pregnancy, child health, vaccination, HIV, etc. And so these include decision supports, which the clinical worker will tell you, hey, you have hypertension, you should do this, that, client communication telling you specifically, you have an appointment on such and such a date because you have chronic hypertension or gestational, you have developed gestational diabetes, and so we want, we need you to come in and you can have it signed off by the health worker who registered her, you know, because we have that information in the system. You can also have feedback dashboards that I talk about, which is given to the health workers to help them track their own progress, so that at the end of the year, at the end of the month, when they are doing, when they are assessing how well the health worker or the facility has done, it is not a surprise, you are giving your progress in real time by a feedback dashboard that is targeted to you. So some key features you're going to, I'm going to show you a bit of a video, which unfortunately goes a bit fast. So it's, this video is going to show you some of the features I've talked about. You're able to search for records, and you're able to look at the risks, the health risks that are generated. You can see data entry validations. I think there was one for, for problem that pressure again. You see how when we enter the raw data, it generates risks, it generates management, and it recommends for referral. And then how you schedule the next visit, and there's also an automated SMS to the pregnant woman and a quality feedback dashboard. So the last three will not be shown in the video, but these are some of the key features of the registry approach. So here goes with. Yeah, I hope you can see. So this is what the registry looks like. You have your creating an internal care visit. This is what the visit looks like. This is from Palestine, so you see the Arabic select the date of the visits, and you can see the previous values at the top. There are different sections for different portions of the visit. These are some information that we can use to generate risk. For example, do you have chronic hypertension, and you see the red that shows up right now. It says high risk pregnancy, which is interesting because this person is called plus for Mike, you have a man who's pregnant. So then you see these are the management's the person has diabetes the person has chronic hypertension she needs preferred you select the clinic, and you've done your referral. And if you see at the top we have a high risk pregnancy that has not been managed you now see that there are risks that have been listed that this will follow the health the client throughout and every health worker will see it. There's a lot of pressure. There is an error that tells you that is outside normal range maybe there's an issue, and it generates a risk of chronic hypertension. If you see the risk related to crime pregnancy. Here we have the dates that that risk was generated. So this is this is an interesting thing with the registry approach. It's listed and as I said, regardless of the dates of the client comes in this section will remain so that throughout her pregnancy, you just every health worker interacts with the with the woman will see this list and know that hey this person has been identified to have diabetes this person has identified to have chronic hypertension. And then for the particular date that you enter the blood pressure which also generated the chronic hypertension. It tells you the date and the gestational age while when this was this happened. So for example, if the person had fever, you will see every date where she had fever here so that if she had fever once and it was some time ago you know you can just ignore it. But at least it gives you a record a running record of all the times that the woman has had incidents that are out of the ordinary. It also gives you a list of the previous values of key things so this is not each and visit. If the information is not is not entered, you know, nothing is seen here. One of the great things is also the top tab which gives you key information about the person. This says that is an unmanaged condition because the person has been referred for the chronic hypertension we don't know if he's been managed yet. This is a person's name 18. The person's a high risk pregnancy and I think that was developed because she has diabetes and chronic hypertension, and this will stay throughout her pregnancy so anybody who opens up a record that red label is there so you know how to treat the person differently. It gives you the EDD based on the LMP you've you've measured it gives you the date the current days gestational age, and then this is gravita para and number of proportions so this woman has is currently pregnant and this is her very first child. Then there is additional complexity in the form of feedback dashboards and this is what this is the feedback for anemia for example if you need a particular context anemia is an issue for example in West Africa. We have high anemia issues because you have malaria. And so but in some parts of the world they might not need to check anemia and so if there's a particular health issue you want to check you just create a dashboard for it. And you want to look at how are percentage of women who are screened for anemia at booking it should you really want to get as high as possible the numbers going up at 65%. The president woman who had no anemia at booking who were screened at the 24 to 28 weeks visit that has gone down by 55%. So we have a target we have the patient count and this is for a particular facility right and this is your performance versus other clinics. This is the top percent clinics score so that you can actually see how you are performing compared to your other clinics. Now these are certain actions that we can recommend to help you improve your your statistics based on the data that you've had. So and he said remember the importance of recording the data entry. Now we have maybe you just screen everyone right but then you just didn't put the data in and so the system things you are not screening people. So then we are reminding you that hey just record the data even if you do it and there's no problem because if the data is not in the system you know we assume that it's not done. Then since you're having all these people have anemia please check the iron supplement stocks and you can check with a responsible stock person to find out if there are low stocks and if there are low stocks follow up with your supervisor and document the follow up here. Because hey we're seeing that people have anemia and let's make sure that if you identify that people are anemic you have the stocks to do so. So this is very targeted right so if you have you know some very specific program like like an HIV program or TB program or you know ANC where you have certain specific actions that you want people to do. You can check you can give them this dashboard to see are you doing those actions. If you're not doing them we tell you how to do it and actually if you're doing them we often can tell you we can also tell you something like hey great job you know keep keep doing it in the action section. And this makes things very personal to the health worker and they know exactly what they are supposed to do so that at the end of the month at the end of the quarter at the end of the year there are no surprises when it comes to performance. Now how do you develop such a program. You have to plot out the data entry workflow. You have to review data from all sources. This includes the notebooks and registers of all the relevant cadres of health workers. How do they register new clients. How do they find existing clients and actually their record. This is all part of figuring out how to create the uniquely identified entity. Look at how they record their data and in what sequence because you're doing real time data entry so you need to make sure that when you provide it to them online. You are not disrupting their workflow because then that means there's more you have to train them on. Look out for notebooks and other informal or personalized record keeping. Because sometimes these registered are printed a long time ago and they are not up to date with the current the the current data and data points are supposed to enter. What fields do they discard in their registers or do they disregard in their registers you can actually look at discarding those data points and not load up your system with data points that nobody's using. What fields do they put dummy data in because sometimes people just put something there just to fill it out and that may be a cue to provide equipment or training to enable them measure that data point otherwise this might be a data point that you should discard. So in developing such a program you actually you know you don't just take the register and just take paper and put to put online. This is an opportunity to review the data entry and the workflow and discard things that are not necessary. Please consider if you are completely getting rid of paper if you're completely getting rid of paper are you capturing all the clinical information that they are writing. For example, you know more EMR type information notes that the person is writing. If you're not capturing all of that you may need to decide and design a simpler paper form to capture the clinical data that you don't need for the tracker right so that the health worker can keep her note somewhere else. So this is some work I did for an implementation in Bangladesh. I looked at these are four different types of what three different types of health workers. This was a clinical worker in a facility. This was also a clinical worker in the facility who was a midwife. This is a field worker and then you also have to look at client cards. These were my sources of information right so for the FWV. She had an ANC registry she had a PNC registry register a delivery one and she had the mother's ANC and PNC. The client card had an injectable funding planning card and they also had a TT card and we wanted to keep track of to status for clients. So you first of all, identifying all the data sources and then I went through and basically just listed all the different data points that I saw, you know, and then I created the system and if, and I just do a check mark right of who has all these different ones because what was required to do was you have to these two different health workers who are actually doing the same job of doing ANC. And so let's we're designing one system for all of them. So let's see the different how they enter, or how they categorize the same data points, and to compromise and have one data point for the same thing that they're entering. So this is a very tedious job but it's really helpful to do that you get a good overview of the data that you're collecting and then you can read things out if not necessary, update things if they are new protocols. And then I included you know the type of data that it is it's free text, you know, etc, etc. Then you have to plot out the clinical workflow, you have to review the clinical guidelines and it's good to upgrade to global standards if it's possible. You should also observe a health worker with a patient and see you know how they they start do they start with identifying the patient how do they identify the patient. Maybe they're using phone numbers when they look at the patient. That's a key to you that when in your identification process use phone number because people's birthdays are not, you know, constants and people don't know their birthdays. Also have a health worker treat you like a patient. And that also, you know, kind of turns another part of your brain so that you understand the workflow. Sometimes it's different observing versus you participating. What part of the clinical guidelines do they follow. And if they don't follow everything notes that and then check with the clinical people, are they supposed to not do this if they're not let's just discard it because you don't want to waste people's time with data points that are not necessary. What do they not follow which is due to lack of equipment or training, then that's a potential to address these things. Right. And when do they refer at what points within the workload do they refer because maybe at some point you know some data points don't need to be answered you can just tell the person stop right here. This is an emergency situation just refer and get the person out. Um, so we have this you require somebody with with clinical knowledge to do. We had a doctor and with us our team includes medical doctor who then looks at what are the infant interventions the kind of data that is collected during the first anti-natal care visit. So you have the personal history of the profile and she wrote out the data points. She had the complications in previous pregnancies this and then she wrote out what is being collected. We also need to think of family history complications in previous deliveries, etc, etc. So you need a clinical person to blot out these are the key things that you are supposed to record. And then within it, you, you know, you write out all the data points, because what we usually will do is once you're reviewing the clinical protocols, we want to present it to the country's clinical experts so that they verify that hey, this is actually what we want. This is actually what we want to do. So you have to present it in a way that is, you know, user friendly for them and so this we felt was a bit good. I left all the formatting, etc, for you to see, you know, the process that she goes through to try to decide something she has questions something she's discarding and so she highlighted them different colors. So then she's looking at how to generate the program rules, right. So if somebody has a risk related to a previous condition. And the person is, for example, less than 18 or more than 35, we consider this person high risk to refer to the health to the complex. So she rise these things out. And I look at this this is easy for her to do you know this is a workflow that we can present to to the clinical. We usually have a clinical meeting workshop where we bring in the clinical experts from the country from the district from, you know, whatever target area and walk them through the clinical interventions that we are going to do. So we needed to find a format that works for presenting it for the doctor on our team to be able to to kind of the other thoughts. And also for me, the person who is doing the configuration to be able to take an easily, you know, configure. So we felt that this was a good. A good compromise. When I look at this, I see that when I when somebody enters age and is less than 18 or 35, I need to write a program rule that generates a referral to the post that I help complex. Then, and then let's go of course when she presents to when she leads their presentation to the workshop of clinical staff, this is also very easy to follow. But this is very tedious work. She has to look at the global standards versus the Bangladeshi standards and look at where we have risk and we have management. This is an example. You know, this is the level of detail that is needed. I'm not going to go through very specific. But you can see, you know, what's work is involved and it's good to do this level of detail. It helps to be clear. It helps the person doing the configuration so that they give you the configuration that you want. And if you have somebody testing the system, they can look at this put in these data points and see how the system reacts. It makes it very easy for you. So I take that these workflows and then I generate this this Excel sheet and I think quite a lot of you do this. I think some people call it the data schema, maybe where I have a column for the data elements where the form name is, is it auto generated? Is it required? Does it have a description? What is the data entry format? Do we have validation where, you know, is there a particular range it's supposed to be in? Are we supposed to show it or hide it? And then we in eRegistry has specific referral types where there was an urgent A&C referral, a non-agent A&C referral and we also had the flag for like, you know, is this person high risk pregnancy? Is should this person do facility delivery, etc. So for example, if I take this number of babies born one, it is the eye button text says in this pregnancy, you know, how many children, you know, were born for a given pregnancy that you entered and then we have a a drop down so that people, you know, to try to reduce errors and then you show this data element if you have the multiple pregnancy saying yes, right? So if I look at this, it makes it very easy for me to configure the system and write my program rule. So this is very, very detailed. And I also give this to the person who's testing this system and they just go through and, you know, when they finish testing it, they can just tell me this is, I've done this, I've done that. Yeah. Yes, yes, I can do that. I can share the Excel sheet with you. And let's see, going on. Now this is a lot of change for this health worker. And so they're like, what is in it for me? This is you coming and bringing more stuff. So you have to give them, you have to design a system that takes in the user and makes things good for them. Nobody likes change. We usually start with keeping the paper while we add electronic, but that creates a data entry burden, unfortunately, but you know, you can just dump the paper and go into, into electronic. They have to learn new technologies, new work processes, etc. So you have to recognize that this is a challenge to your users. So how can you motivate them? You can use, you know, the data entry users can also get reports and they actually should. You know, everybody likes things that are designed for them. So they should get reports so that they can check, you know, am I putting the data incorrectly? Is there something that's not, you know, working? You can also tell them that, hey, this reduces your reporting burden. We all know that the big issue with health worker burden is reporting. They spend so much time of their work during reporting. But here you enter the data at point of care, for example, and that's it. At the end of the month, your reports are all beautifully done. There's data validation. You saw that when the, when the black pressure was entered incorrectly, the system showed you that, hey, we think there's a problem here. So, you know, it's not now that at the end of the month when the clients is gone and you're trying to make a change, how are you going to figure out what the, what the data was supposed to be? So, you know, we give you real time data validation so that you can make corrections. When you're doing reports, don't just think about reports for donors in the national level, and this is a problem that we have, but that is because that is kind of a remnant from our aggregate system. But now we're doing tracker, which is individual level, you know, the health worker who deals with the person at the individual level, they can get reports on that. When it was aggregate, fine, it could go to national level, but now that we're doing tracker, which is individual level, you can give reports to them. So you can design reports that are relevant for the facility and the individual health worker processes, like I showed with the dashboards, where it helps you keep track of your progress. Also consider adding appointment scheduling so that you can help the health workers generate working lists of who is supposed to show up today. Who was supposed to show up yesterday, who didn't show up? Now I can just call them so that they don't have to now look through their records and figure out who was coming, who did not. If they just, you know, start up in the morning and they have that list, you know, they kind of sign one health worker, this is your job. You wake up in the morning, you know, your customers work in the morning, you generate the list of people who didn't show up, the list of defaults or whatever it is you call them and call them. And that makes life very, very easy. So the appointment scheduling, you can use the system to schedule appointments for our registry for antenatal care. We have the WHO has a recommended number of antenatal care visits. Originally it was four visits, now they're recommending eight, but with the four visits, it tells you within which gestational ages that you should come. So if you come for one visit, the system is able to generate your next appointment based on the WHO protocol when you're supposed to come again. And the health worker doesn't have to calculate, you know, what is your gestational age, when is the next appointment, you know, that makes life easy for them. It can send reminders to clients of the appointment that gives the users burden for the health worker because they don't now have to follow up with people. So if enough people get reminders of the appointments, you're going to have more people coming in and it reduces your burden of going to call them to come. It generates a list of clients due for care and you're able to track clients who have missed appointments. So this is the end of my presentation, you know, just our interventions. And we are supported by the Norwegian... What is Norwegian? I always forget. But it's kind of like the equivalent of the Norwegian USAID. And we are given, we are supported to offer this, you know, eRegistry design support to countries that want to do eRegistry. We can do workshops with you. And yeah, so if you are interested, just email me at akdo.fhi.no. My other colleague, Brian, who you probably saw yesterday, is also involved in this and we offer these workshops. So any questions? Thank you. It's the Norwegian Agency for Development Corporation, NURAD, which I guess, you know, spells with R is something else in Norwegian.