 Last time the format was basically a lot of question and answers about your priority needs. Today we're going to go be a bit more structured. Both from beginning to end through a number of the training slides. I won't go through all of them in detail there is not time for that. This is designed as a one day training, or is it to a day training. But, you know, I will start from the beginning using some of these slides. I will now go through overview. I don't think I will do who nets installation. I think you've already done installation. You don't need me to teach you how to install software. There are people there who can help you install software. If there are any problems. There are questions on installation. Let me know. But I plan on doing who net one overview, followed by who net three laboratory configuration. I will also try to go a bit more slowly. But please let me know if you have trouble understanding. I'm going to who net one overview. You don't need to do anything on your side for this presentation, this, this, this first presentation, simply just watch and listen. So who net is surveillance of antimicrobial resistance, but not only antimicrobial resistance. It is surveillance of routine, usually microbiology data to study evolving microbial populations, including antibiotic resistance. So if there's an outbreak, it doesn't matter if it's sensitive or resistant, you still want to know. What is the software for managing microbiology data, especially most commonly routine data. So I work with Dr. Thomas O'Brien, who started this work. If I go back he started his work really around 1964. 1962 he trained with Kirby and Bauer. So he's been working this for a very long time. He had a vision when he was head of the microbiology laboratory. He saw that the results of the microbiology laboratory were valuable for patient care. Of course, that's why we do the test. But in addition, the microbiology laboratory provides valuable data for epidemiology and quality and knowing about new threats in a way different from chemistry and blood bank hematology. In those labs, the focus is on the patient. In microbiology, there are two areas of focus. There's the patient that must be treated, but there are also the bacteria and viruses and parasites. It's important that routine clinical microbiology laboratories generate routine data every day that could and should be utilized to provide a detailed view of evolving microbial populations in real time, in addition to patient care support. However, the data from routine laboratories remains largely untapped and underutilized. So his idea was that the use of a common software supports local, national, regional and global collaborations to support a number of objectives. The recognition, tracking and containment of emerging threats, both resistant and non resistant cost effective care and treatment guidelines, public health policy interventions advocacy research. And importantly, in our early discussions is to improve laboratory capacity. In the first one or two years of most surveillance initiatives, the most interesting findings, the most unusual findings are often not true. They are often due to errors in laboratory testing and results or simple biases. If you only collect samples from intensive care unit patients, you will end up with very good data for intensive care unit patients, but those data are not generalizable to the community. So when we get later to data analysis I will always start with a review of the data from a quality control perspective, followed by an epidemiological and microbiological perspective. We developed who net with two objectives in mind. First of all to improve the use of local data for local purposes, and secondly, to promote national and international collaborations. The most important use, they are the local uses. Why are the local uses so important. Well, we need local, we need the local data for local outbreak detection, local antibiotic use policy, improving the local data capacity. It is at the local level that the patient sees a physician, the physician works with the laboratory to make a decision about an antibiotic for the patient to receive. So in terms of the control of resistance, we really need to think about that level of the patient and the physician and the local pharmacy and the laboratory. That is one reason why the local is so important. The other reason that local is so important is for sustainability. If the reason, if the reason all of you are doing who net and data entry and data management is because WHO asked you to do it, or IDDS, or EPHI is not going to survive in the long term. For this to survive in the long term, there needs to be a value to the data entry people. If we do it, if it's a short term, they will do it often because people ask them, but for this to survive for two years, five years, 10 years, there must be some value to the people who are doing all of the data entry. So that's why we feel local is so important. And at the national level in international are additional benefits, benchmarking, mentoring, looking for national pictures of resistance, national pictures of outbreaks and antibiotic policy. Internationally, WHO uses these data for a number of objectives, one of which is gap identification, which countries have data, which countries don't have data, which countries have good quality data, which countries have obvious mistakes. So I need to update this, but we have the who next home page. The software is free. So go to the web page and you can download the software, the tutorials, and some other information. This slide, we also need to update this is an older picture of who net registrations and use from around the world. You see a lot of who net use in Latin America, because I started 30 years ago. 30 years ago. East Asia 30 years ago. Other countries started 20 years ago, like the ears net project for remaining European countries. Caesar, for Central Asia Eastern Europe started about eight years ago. And now because of the Fleming fund and because of glass, and because of the initiatives of ministries of health, and the WHO World Health Assembly resolutions, we see more and more use now in Africa, South Asia, which is relatively new for some of the countries. We will talk about two softwares, the who net software and the back link software. Who net is for data entry data analysis and data sharing. That's the more important software that the reason we are doing all of this is so that people can benefit from who net. We now have about 28 languages of the desktop version has been available for 30 years. The web version unfortunately has been on the back burner, because I only have one programmer and we, you know we don't have time to do everything we would like. We had a very significant advance last week, where we completed the final final final retirement of some of our old technologies. So I'm very pleased to say that in the months that come, you will start to see much more quickly, new features, new possibilities. We will not start with the web we will start with standard reports, more outbreak things and other features along those lines. But after we do some of those low hanging fruit, high priority needs. And then we will finally get back to the web version back link is useful. If a laboratory already has a computer system throughout the high medium and low resource world, a large number of facilities have a system. This is for lab microbiology laboratory data. Sometimes those systems are very expensive, very complete. The commercial system such as certain and many tech and some quest. You have mid range systems, which focus more on final results. And then you have the low end systems like Excel, people just put the data into Excel, or access. You also have data in a machine like a vi tech or a micro scan or a phoenix. The purpose of backlink is to avoid double data entry. If the data have already been entered into your computer system. We would like to use backlink to transfer the data into who net in a semi automatic manner. So you can drive all of the benefit from your data, but without the need to manually re enter the data. We do that once a year, once a month, month, once a week, my own hospital, we do this automatic daily, we download the data at one o'clock in the morning at 115 backlink automatically runs with out as being there 130 who net runs a series of analyses. In the morning, nobody looks at who net and 130 at seven eight o'clock in the morning are infection control staff and our laboratory staff. Look at the output, they look at the Excel files, or the access files. So all of that and backlink can be automated. When you install who net backlink installs at the same time. So you do not have to install them separately. The two softwares are distributed together. You can look at the three who net modules. Most interesting module is the last one data analysis. That's the reason we are doing this. You can look at interactive analysis or standard reports. So, we really want to focus a lot in our training on data analysis. However, before doing data analysis, you must do data entry manually or using backlink. You have to do clinical reports. That's the boring part of who net data entry. But before doing data entry, you have to prepare who net, and that's laboratory configuration. And that's where we will begin our training. What you do in laboratory configuration, you basically tell who net, who you are, what into bodice you have your locations is basically configuring the data entry screen. So you will do configuration, followed by data entry, followed by data analysis. And at some point we will also discuss backlink on how to get the data transferred into who net. I will show you screens from all of the softwares to just to give you a little orientation. So here in laboratory configuration we start. There were, I thought there were red lines, red circles, an animated version, but this is not this this software this slides I'm showing you are not animated. So at the top of the screen, you choose your country. We have a special country for demonstration purposes called WHO. You put your laboratory name, your laboratory code. This has always been a three letter code. But next week, we're changing it to a six letter code, so that you can make the code longer if you would like. I would say that if your hospital is predominantly human data you choose human. But if you work in a public health laboratory, a food laboratory and animal laboratory, you might want to choose the second option for human animal food environment. Who net for who human animal food environment is almost identical to just for human only. It just gives you additional features, what kind of animal. What kind of animal food. So I will show both, but they're very tiny differences between the two. So that you answer those questions at the top. Then you have four boxes for buttons at the bottom of the screen for command buttons. The first one is usually required what antibiotics do you test. Do you do CLSI, do you do you cast, do you do just diffusion, do you do so. Do you do E test, do you do a combination. Most labs do a combination assumes to some discs, some E test for example. You can also tell who that your panels panels what do I mean by that panels is to make the life of the data entry person easier. If I'm entering a gram negative organism, I want to see my gram negative antibiotics. If I'm entering an enter caucus I want to see my enter caucus antibiotics. So this feature for antibiotics allows you to define the complete list of all of the antibiotics that you test, and then you can customize it by telling who net which antibiotics go with which organism grouping that is called panels. Then we can enter the locations. What does location mean. Well, it means what you wanted to mean. In a hospital, the location typically means the name of the ward or the name of the clinic, diabetes clinic. I see you medicine male medicine female. So for a hospital location, typically is the name of the hospital ward, or the name of the hospital clinic. And if you work in a national center like a PHI, the location might be the name of the hospital, or the name of the town, the name of the district, the name of the region. Or if you work in an animal laboratory, the lab, the location might be the name of the farm. The name is a market, the name of the veterinary clinic, the name of the slaughterhouse. Or if you work with food, the name that the location might be the name of the restaurant, the name of the market. So the word location is a very general word, you define location in the way that is most interesting for you. If you work in a hospital, usually want to know the location with sample was collected inside the hospital, but at the national level, sometimes you just want to know the name of the hospital or the name of the region, or the name of the city. Finally, there are data, well, then there are data fields. When you start HUNET with new laboratory, HUNET gives you a predefined set of commonly requested data fields. Patient number, patient name, gender, date of birth, location, location type, like inpatient outpatient, simplified. Specimen date, specimen number, organism, beta-lactamase, the antibiotic results we discussed, comment, these are the normal HUNET data field questions. A data field is simply a question that you will see when you get to data entry. HUNET allow, and most hospitals are happy with that list, and they don't change the list. There are good reasons to change the list. Some people want to add more information. What was the diagnosis? What was the name of the doctor? What was the Graham's name? So HUNET gives you a standard list, but using this feature in front of you called data fields, you can add more questions that might be valuable for your hospital or for your country. You can also remove questions that are not needed. For example, if you are working in an animal laboratory, you do not need the animal's first and last name. If you are working in a food laboratory, I don't want to know the age of my food or the gender of my food. So if there are questions you want to add, you can add them. If there are questions that are not relevant, you can remove them. So for data fields, you can just customize it to match what you want to see in the data entry form. Finally, on the screen, you see the alerts. It's a very valuable feature, but most people don't change it. If you want to change it, you can. But in alerts, we give you about 190 predefined rules. There are high priority alerts, medium priority alerts, and low priority alerts. For example, the high priority alerts, we have NYSERA gonorrhea is important. Vibrio cholera is important. Salmonella tithy is important. Those are important species. Or CRE, Carbopendium-Resistant Intervectoration. CRE are important. VRSA, Vancomycin-Resistant Staph aureus are important. So these are high priority important species, important resistance. Then you have medium priority. MRSA, VRE, ESBL, they are important. They are also relatively common. And then you have low priority alerts. Low priority alerts are primarily quality control. For example, Klepsiolonamonia sensitive to ampicillin. It's possible, but it is rare. Most Klepsiolonamonia 95, 98, 99% of Klepsiolonamonia are ampicillin resistant. If you do find a laboratory result, Klepsiolonamonia ampicillin sensitive, it might be true, but it also might be a mistake. It might be a mistake in different entry. It might be a mistake in the antibiotic result. It might be a mistake in the organism. Maybe it is ampicillin sensitive, but maybe it is an E. coli and not a Klepsiola. So these low priority quality control alerts can help the laboratory to find problems and address them and investigate them quickly. Just because you had a question, please, after this slide, can we pause for a question? Of course, thank you for reminding me of that. So who that gives you about 190 alerts? And most people are happy with those alerts, but you can add more alerts or you can delete alerts that are not useful for you. And then you click on save, and that is your configuration. You have saved your antibiotics, your locations, your data fields, and then you are ready for data entry, and then after that you will be ready for data analysis. If you want to make any changes, simple, you just click on modify laboratory, you come back to the screen, and you make the needed modifications. So I will stop there. And what are there any questions? I think you're on mute. There's a question in the chat box. Suppose we have data from sites having their own configuration. Does who not allow us to a new lab code when we're interested in merging them. So typically, in a country, every lab has their own configuration, their own antibiotics, their own locations. So who not at the national level has no trouble with that. So for the things that you care about, like inpatient versus outpatient, a minimum core set of antibiotics, you would like to introduce some degree of standardization. So in answer to your question, it is not a problem that everybody does something different. But for the things, the minimal things that are most important, where you want standardization, I would like to recommend standardization. This can be standardization data entry or standardization and laboratory testing. For example, the names of the medical wards, like medicine male medicine female fifth floor or fifth floor north, those the lab should do exactly what they want to make it most customized and useful for them. On the other hand, at the national level, there's certain things we would like them to try to standardize. You know, they're also just testing issues, if maybe nine of the laboratories test any panem and one of the laboratories test Mara at the national level, it would just be easier if everybody tested any panem. So in a later session, I will show you about national data merging, national data management, national feedback, and then facility specific feedback. So yes, it is not a problem of everybody does something different, who net has no trouble managing bringing them all together. On the other hand, for the things that you do care about those things I would like to try to introduce standardization. So when people send me things. Sometimes I say that's perfectly fine I don't need at the national level, continue to do what you're doing. In other cases I say can you please change this. Okay, great. For example, who net as a field for the medic the patient medical record number. Somebody removed that. And they replaced it with something called national ID, which is what they were using in their country. The problem is some of the labs at the patient number in this column, but this lab at the national medical record number, the mass the national patient, the national citizen registration number in this column, because they were in different columns it was So, mostly if they do something different I'm okay with that, but sometimes I do want to standardize it. If it is something I plan to analyze other questions. Yes, john, can you hear me. Yes, I can. Yeah, this is good morning and good afternoon for all of you can make it thank you. So my first question is, is a new version of phone it. Can we access data from sites and also, can you communicate with scientists online. Is that possible. Okay. I'll answer that in two parts. The version of who net that we use and distribute and support is the desktop application. The software is not online. You can share data online by email or secure file transfer. In Vietnam they have made a nice web portal where they can upload the data. They did not use who net for that. And they do collect the data through the web, using a solution that they develop nationally. So who net is a desktop application. If people have data from who net, the new who net 2020 who net 5.6 who nets to every version of who net is still compatible. So if you have any who net data from 1989 who net can still read it. So, even if you have different versions of who net, don't worry, the current who net can still read them. So don't worry about the version of who net we can still share the data. But it is a desktop application it is sitting on your hard drive or on one of your server drives. We have a web version, so that you can do data entry online and data analysis online. We have a reasonable demonstration version, but it is only as a demonstration is not ready for use. I want to mention the example in Vietnam they have some very good it people there I've been working them for many years. So they use who net for a number of things. They also made additional utilities, and one of them is a web portal. In fact, Mikkel was very involved in this. They have a web portal for uploading who net data into their national DH is to platform. So there's a web portal for where a laboratory uploads the data automatically into the national database. They have the ability to analyze the data on the web platform. So the analysis are not as complete or as rich as the who net analysis. So they can still use who net for doing outbreak detection and other things, but they can also use the highest two for a number of things. So this is an example where there is not a web version of who net, but they have made a portal for using who net data on the web. So this is a presentation and a training over the last few weeks with Sri Lanka. And that's been very interesting for me, because, like many of you, he knows who net use for facilities very well for, he goes to hospitals, he trains them and who net, he teaches them who net. He knows very well. He knows individual laboratory use of who net well, but there's a lot he did not know at the national level he did not know how to merge the data from the different hospitals. And so part of this training was very useful for him, because I usually do not give training courses to national data managers, national data managers is usually more one on one question answer. When I go to the country we give a formal training course for a lot of people and often there's not enough time for the national data manager. So similarly in Ethiopia, based on your questions, you know, many of you do know who and add very well for individual laboratory use, but I hope as part of this training, we can also give you some tips and shortcuts and value for in addition, working with network data at the national level. There are two countries, Argentina and the Philippines where we have a project where we are also using softwares called WinSCP and FileZilla to do automatic secure file transfer protocol that is abbreviated SFTP. So FTP is file transfer protocol. SFTP is the same thing, but it's secure. This is one of many technologies to automatically send data in a secure manner. So in fact, they at the national level every day, they do have the who net data files. They do not use who net to do that they use normal windows utilities at the local level the people manually enter data into who net and automatically every day. They use one of these softwares I mentioned to send the data to the national level. So this is an example of an online automated daily up to date system that uses normal windows softwares but it does not use who net. So in the future we will offer an option within who net, but you can already do it now you don't, you do not need who net to send files securely in an automated way. This could also be a value to Vietnam, you're right now they were interested in that right now they have a monthly manual data upload, but that manual data upload could be replaced by automated daily with a secure email a secure file transfer. So there's a lot that could be done for sharing data in a secure manner that do not rely on who net in the future I want to put some of those into who net but you do not need to wait for us, you can already use many windows softwares for doing this. So what do you advise to for us to use any kind of software that can secure our data because currently data is coming through email, which is not really secure. That's a very important question, I would say in the majority of countries, unfortunately, the most common way to share who net data is just by sending a normal email once a month. Two problems with that it's not automated so somebody has to manually send the email at the other end somebody has to manually open the email and save the attachment. The other issue is not secure. So, but now countries are starting to move in a better direction where they're starting to require patient protections. In the United States, this, this has always been true, we have a law called HIPAA, so you cannot send an email, you have to do it in a proper way. So how can we send the data, you can send the data using a secure email, or you can send the data with a normal email, but with the password protected secure attachment. So they could send you a normal email, but they could actually put a password onto the attachment. So that's a secure email, normal email with secure attachment. But those are both email approaches, for example, once a month. The other or there's the web portal that's what Vietnam does. Someone goes to a website and they upload the who net files. So these are three ways that you can have a secure transfer once a month. So one of the things that Argentina and Philippines is they use these softwares called file Zilla, or acute FTP, or when SCP the name of these softwares they are secure file transfer protocol softwares. So there are many options for you. And one advantage of these is that they can be scheduled. You can do this automatic every day at one o'clock in the morning. So let's try to come up with an appropriate secure solution for you. So should we do the secure email approach the web portal approach, or should we do the secure FTP approach. The answer to that really depends on how often you want the data. If you plan to analyze the data monthly, they could just email you the data monthly. And that's a good way to start there's so much going on there's so many things that we would like to do. So just if you collect the data monthly in a secure way. This would be a big improvement on what you are currently doing. You will still get the data monthly, but it will be in a secure way. On the other hand, if you do want automatic daily analysis automatic daily feedback automatic investigation of local or national outbreaks, then one of these SFTP solutions would be appropriate for us to discuss. Okay, thank you. Maybe can you send me all this, you know, the type of software so that we can use because currently we are looking to solve this problem. We collect data every month, but it is through email. So I am not comfortable with that because it's not secure. Sometimes they send it in their personal email, even though we gave them a different email. So just summarize this. I mean, send us the type of software that different countries use so that we can go for that. Thank you for that explanation. We have to solve it. Yes. And I just request that somebody send me a specific email to request that my life is controlled by my inbox I turn on my inbox and I will. I will send you separate email for that. Perfectly. Thank you. The other question is I have also another question. We have also a plan to capture both human and animal data in one location as a pilot. So, can you tell us how we can configure it to capture both the data. Yes, definitely. So right now we are discussing the who net overview at the high level with this first slide set. After that I will continue with a live demonstration of who net laboratory configuration. So I will answer that question during the demonstration of laboratory configuration, and also during data entry. Okay, thank you. I already mentioned on the screen because I don't know when you join the call, you see those two options there at the top human versus human animal food. That is the first step. If you want human animal and food, you choose the second option. If you're a hospital laboratory that does 99% 99% human and a little bit of food a little bit of animal. I would still just use human. I do a lot of animal a lot of food and I suggest the second option. So this is laboratory configuration is basically preparing the data entry screen so that you can later do let data entry and data analysis. Moving on to click here great data entry. First of all at the top the first question is this human or animal or food. So to change all you do is change from human to animal to food. It's environmental and feed but those are not developed. So the three working priorities are human animal food and very simple if you change from human to animal. Some of the questions you see here will disappear. We do not need to know the animals first and last name. We do we we do not plan on throwing the animal of birthday party so we don't need the animals date of birth. If you change the question to food, the questions again will change. So that's really the only difference. You change the origin from human to animal to food, and then it will change the questions on the screen. The location questions will stay the same the specimen locations the microbiology results, but the quote unquote patient origin questions will depend. Great. So the data entry screen, you choose human animal food. If you choose human, it will ask you the number first and last name sex date of birth age, age categories basically adulterapediatric. If you have it the date of admission is very useful. I do recommend it, but it's often not realistic it depends on if the laboratory has it. Many laboratories say we don't have it, but we can ask for it you know next year. So they have a data entry form. Sometimes the date of admission is on the data entry form but nobody fills it in. Or sometimes it is not on the data entry form and they can add it. So maybe this year you don't have date of admission, but something often it's something that could be added in the future. We'd like date of admission because it helps us to separate in patient infections from outpatient infections. If you have somebody who becomes very sick on Monday and goes to the ICU on Monday and has an E coli in the ICU on Monday. That's still a community infection. They were hospitalized because of their infection. So if a patient has an infection a sample and E coli on hospital day one on hospital day to most of those will be community infections. If somebody has an E coli or a staff or is on hospital three four and five. It's more probable that it is going to be a hospital infection. There's not a perfect definition but it's a simple definition. If it's hospital day one or hospital day to, we're going to call it a community infection. If it is hospital three four and five, we're going to call it a hospital infection. It's a simple rule. It is not a perfect rule, but a perfect rule, you need the patient's medical chart, you need somebody to read every, and you cannot do that. So date of admission is helpful to helping you to, on average, separate the inpatient infections from the outpatient infections. As I said, it's not realistic for many places in the short term, but hopefully in the long term people can consider it. You have the location information that could be the ward or the clinic or the farm or the restaurant or the veterinary clinic. So location is simply what you want it to mean department would be like medicine and surgery that's not really relevant for animals or food, but location type is inpatient outpatient farm restaurant. Again location is very specific you the laboratory puts exactly what they want. So in specific type we tried to introduce some standards, just so at the national level, you can have some degree of separation of inpatients and outpatients, and a reliable consistent way. Next specimen number, specimen date specimen type reason most people leave that empty in a hospital laboratory 99% of the work is diagnostic, you are taking care of sick people. You should be doing some research samples or some screening samples. So, a lot of people just leave reason empty, but if you want you can enter that as well. Finally, the micro about the microbiology results organisms your type beta lactamase, these are all optional. You can add more questions if you want like Ramstein, or you can remove questions. If you're working with Salmonella, you don't need the beta lactate you don't if your food laboratory with Salmonella beta lactamase is not a relevant test ESBL is but not the normal beta lactamase. At the bottom we see our antibiotics disc MSE test. This is the full list this is all of the antibiotics. But if I put an E coli. Most of these antibiotics will disappear because who never will only show me the gram negative antibiotics. I said earlier is the panels. So if I choose a staph aureus that list will become focused on staph aureus and staph lecocci. If I put nice you're going to read it will show me the nice you're going to read. So that's what I talked about the full antibiotic list, which is what you see here, as well as the panel list, and the panel depends on the organism. If you have any questions at the top right of the screen, you see it says save isolates you click on save isolate. When it will ask you do you want to save the isolate and start a new isolate, or do you want to save the isolate and continue with the same specimen. You know for example you have you do have a blood culture with an E coli, but it also has a staph aureus. So in that case I don't want to retype the name and the date of birth and the specimen number the specimen date. It's the same sample. Yes, I want to continue with the same sample. You only need to put in the microbiology results. At the top of the screen it says you database. This would allow you to see like an Excel spreadsheet format, one row for each isolate. So after you enter the data you can still see the data later. Click on your database, but it will show you a database list for you to see all of the results. You can edit the table edit the isolates search. It's a lot of people leave who net open on their computers, waiting for phone calls. The doctor will say I need the results for Mrs Jones. So what you do is you go to view database you do search you find her results. You tell the doctor the results. So a lot of people do use who never clinical reporting is not ideal for that, but of course it's an important news. We didn't design it for that purpose. And we will, there are things we can do to make it better for that purpose, but it is a common use. A lot of people, you can also see print you can print out the clinical reports. So that data entry allows you to do data entry reporting, like on the phone just by reading it off the screen, or by printing it out and distributing it there. There is an option there caliper most people do not use it. There are these have one here I have them at the office. It's basically an electronic ruler. You know you can measure the zone diameter with the caliper. Do you see me I don't know if you see me. I don't see you. If you there's a lot of space on top of your head but other than that, we see you maybe if you move the camera, we tilt it down a little bit we'll see you more, but we see you fine and we hear you perfect. Yeah, that's great. Okay, great. So I guess I'm going I'm moving my hands. So, oh then this is better. Yes, thanks. So with the caliper caliper is an electronic. I can see you. Yes, all of you should be seeing the same thing. Yes, you are your camera is being seen who cannot see the who cannot see the camera. Who's that. Can you see the screen. You don't have to see me, but can you see the screen. Yes. So you can. Yeah. You can set up go to meeting to see both cameras and the screen, the screen. I see I see both. We can see you if we set it up. Yeah, you're good. Well, it's not so important to see me. I moved my hands when I talk. Yeah, measure zone diameters. And then you can automatically put the zone diameters into who net these, these are not in these are not inexpensive. The caliper is about $150 and the wire, the wire is also about $150. So these digital electronic calipers are about $300. There's also Bluetooth version, which is even more expensive, but I'll just leave it at that very few people use calipers. So that's the data entry screen. And the rest of this particular presentation I will focus on data analysis. And again it's okay this one is animated. So basically in front of you, we're basically made in India. I gave them one of my presentations, and they used my slides, but then they put more slides so it's a mixture of what I did and what they also did. Okay data analysis that on the screen the data analysis screen there are three required fields on the left. What kind of analysis, which organisms and which data files, those are the required questions. On the right. There are additional questions I just want certain isolates and certain options and one for patient. I'll start from the beginning. So first we choose the analysis type. Do you want isolate listing do you want percent resistance, do you want multi resistance do you want outbreak alerts. So analysis type you choose the kind of analysis you would like to do. Then you choose your organisms. You can be very specific like E coli or staph aureus, or you can say all Graham negatives, or all organisms. When it also allows you to do viruses and fungi and parasites, when it also allows for no growth, normal flora contaminated specimen, grand positive caucus. The third required question of the data files, do you want the data from this year from last year, do you want the data from hospital one, the data from hospital 10 when it allows you to analyze one file, or 5000 files. So if you have data from 20 hospitals, you can choose the data from all of the hospitals at the same time. In other words, we're not combining the data files, the data files are staying separate, but we can combine their contents during the data analysis. So we'll talk about that more when we talk about national data analysis and national data management. So if I have answered these three required questions, which analysis which organisms which data, then you can begin the analysis. But there are other options. For example, maybe I do not want all of the staff or isn't all of the E coli. Maybe I just want the urine. So I can say specimen type equals urine outpatient. It's resistant pediatric. I can say, from a certain time period a certain room just from the ICU. So, if I want all of the staff or isn't all of the E coli, I do not use this option called isolates. But if I want to put in a patient filter, or a location filter, or an organism filter, then isolates allows you to do that. It does have other features and other options that we will discuss later. The most important one is one per patient. If you have, of course, all of you have very sick people who are in the hospital for a long period of time. And sometimes you will have a patient with step or is five times. They have step or is in the left arm, right arm, urine, blood. They will have step or is on Monday. We'll have step or is again on Friday. Some people ask, should I just put the first one in. I said, no, enter all of them. We want all of them in the database. If the patient has five step or is put all five into who net. It's also easier for the data entry person because they don't know if it's the first one or not. If you have five staff or is from the same person, put all of them into who net, because maybe some of them are an ssa maybe some of them are MRSA. Maybe some of them are in the ICU from Monday. Some of them are in the outpatient on Thursday. So I do want to see the five results. Did the resistance pattern change? Did the location change? Did the patient have it in blood and urine? And this is important obviously for clinical reporting. If the doctor sends you five samples, the doctor wants the results from the five samples. So, for data entry enter everything. Also, if you're downloading the data from your laboratory information system. Of course it's going to download everything. But in data analysis, sometimes I want all of them, like I just described, but sometimes I just want the first one or the most resistant result. And that's what the purpose of one per patient is. If I want to see a list of people with CRE and their movement around the hospital, I want to see every CRE. If I want to calculate the percent resistant to CRE, I want to discount one patient at a time. Sometimes if you have a patient with CRE in the ICU, you might have five or 10 CRE from that person. And I want all five or 10 in my database. But if I want to tell my pharmacy the percent resistant, I don't want to count that patient 10 times. I don't want to count that patient once. Otherwise, I'm going to bias my statistics to the sickest people with the most resistance with the most complicated situation. So when it allows you to analyze all of the isolates if you want, when it also allows you to analyze one per patient, both of these options are useful. These are examples of some of the outputs. This analysis is called isolate listing. And I simply asked for a list of people who have MRSA. Okay, I'm going to show you one slide for each of the analyses before I do that. Are there any questions? It's easy for me to talk and talk. Yes. John, can you a little bit explain how we can identify hospital acquired infection and accommodate infection. This is important for us. Yes. Okay. I will go back to the screen called isolates. So here in the United States, the CDC, the US CDC has a project called NHSN National Health Care Safety Network. And they have two different ways to report multi drug resistant organisms. So the CDC wants to know the MDRO organisms, the multi drug resistant organisms. But they allow the laboratory to do it in one of two ways. One is called laboratory defined event. One of them is based purely on laboratory data. And who in it allows you to do that. So I will tell you how to do inpatient versus outpatient using the CDC strategy of just purely using laboratory data. The CDC also offers a different module, which is called clinical reporting. This is much more work. So first I will describe to you the clinical reporting. The clinical reporting, if you find somebody with MRSA, what you need to do is to find that patient's medical chart. Look at the risk factors. How long have they been in the hospital? Did they have surgery? Were they hospitalized because of MRSA? Or were they hospitalized because of a hip surgery or diabetes or blood pressure, and they picked up MRSA during the hospitalization. So this is very manual. And it requires the patient medical record. It requires a judgment call. The infection control person looks at every detail and decides, yes, I believe that this is a hospital infection. So this is the detailed clinical report module. There are two problems with it. It's very tedious and a lot of work and you need the medical chart. That is one problem in doing this. The other problem in doing this is if you ask 10 different infection control people to tell you if this is a hospital infection, they won't always agree. Sometimes they disagree. Because the truth is, if you give me the information, there are three possible answers. Yes, this is definitely community. Yes, this is definitely hospital. But there's a lot in the middle where we are not sure. And when you're not sure, some people will say it's a, some people will say it's hospital acquired. Someone will say it's community acquired. So even though the clinical reporting module is the most thorough and the most detailed, it also has a lot of variability between different people. This is especially true if the hospital cheat. Some hospitals, you know, if you get MRSA, the hospital doesn't want to look bad. So if it's clearly a hospital infection, they will call it MRSA. But if it's maybe a hospital infection, they might call it a community infection so that they don't get penalized in the national picture. So the two problems with the clinical reporting is that it's very detailed. It requires a lot of time and effort and knowledge. And in addition, you end up with variability. Different people will have a different judgment call. So now I will describe the other CDC way, which is possible in who not, which is the laboratory data approach. What they have. So who not specifically does the Debra Joe approach, the Debra Joe approach is based on the CDC approach, but simplified. I will describe the CDC approach, which is more complicated. The CDC approach says, if it's hospital day one, or let's see, if the location type is outpatient, we're going to call it outpatient. That's usually true. It's not always true. For example, if the patient was discharged on Friday, patient was discharged Friday, and then they go see the emergency room on. They go to their own private doctor a week later. They might have a hospital infection, but the hospital infection was not clear during the hospitalization. The hospital infection became clear after the patient went home. So this this is I'm highlighting some of the deficiencies of the CDC and the Debra Joe approach, but it's also easier and at the high level national level, we want to try to keep it realistic and sustainable. So the CDC approach, if it is a sample from the community, we will call it a community infection. That is usually true, but it is not always true. If the patient had a recent discharge. If it's a if it's an infection from if it's a hospital and if it's a hospital sample from hospital day one, or hospital day two, we're going to call it a community infection, because the patient just arrived. They were very sick with fever and cough and diarrhea and sputum, they were hospitalized, and they had a sample taken on hospital day one or two. So even though this is a hospital sample, we're still going to call this a community infection. But if the sample was taken from hospital day three hospital day for hospital day five, we're just going to call it a hospital infection. That would be true. You know, for example, if a patient is an abdominal abscess, and you don't take a culture of the abdominal abscess until hospital day four, it is a community infection, but you did not diagnose it until hospital day four. So in short, if it's a community sample we call a community. If it's a if it's a hospital sample on hospital day one or two, we call a community. If it's a hospital day three, or four or five, we call it hospital. What I just told you is the WHO definition. That's the definition that I put into who not the CDC definition is a little bit smarter. If the patient if it's a sample from hospital day one or two. But the patient has been recently discharged from the hospital, we will call it a hospital infection. In other words, if the patient is the patient spends a month in the hospital. When it comes to the outpatient clinic three days later. The CDC will call it a hospital infection, because the patient spent a month in the hospital, certainly scientifically that makes sense. But the laboratory people around the world do not know that the patient was hospitalized last week. So the literature just takes a little bit of a simple approach. The CDC definition, you do need to know if the patient was recently hospitalized. And that's not realistic as a simple flight definition. So, I hope those that definition is clear. How do we do that in who net. The way that we do that in who net is we go to isolates, and there's an option there. If you have date of admission, there's an option they're called hospital day. The hospital date three or later. I'll just show you I'm going to show you who net. I hope you. Where's my new one. Oh, oh, that's right. This. This is my old laptop. So I have not updated this who net yet. So I'm going to who net. And I'm going to the Denver test hospital. I'm going to data analysis. Going to data analysis. Whoops, hold on. Particularly database does not have date of admission in. So let me just go to clinical information, date of admission. Just ignore what I'm doing and just getting this ready for what I want to tell you. Okay. So here you see there's a question called date of admission. After that there's one called hospital day. So hospital day admission is a real data field. Hospital day is a pretend data field. It's not a real data field in the database is generated during the analysis. So if I want to look for specific date of admissions, I can ask for specific date of admissions. But that's not what we want for our discussion. I go to hospital day, and I can please show me all isolates that are starting at hospital day three or later. hospital day three to nine, but you don't have to put anything there. So if this is the hospital day three or later, it's we're going to consider it to be hospital. If it's hospital day one or two or outpatient, because there should be no, there should be no date of admission, then it's going to be called inpatient, then it will be called community. Is that clear. Does that help. Yeah. So to use this definition, you need date of admission, and you need it entered. You do not need to put a date of admission for the outpatients, of course, but for the inpatients to use this feature, you need the date of admission. I would suggest the date of admission as a longer term goal. You need to discuss it with the laboratories. As I said, some people will tell you yes it's very easy we have that. Other ones will tell you well, it's impossible. We don't have it and we're never going to get it. I think a lot of your places will tell you well we don't have it now, but you know we could start a next year. Hopefully we can routinely do the date of admission. So, when you're getting started there's so many things to start with this discrepancy this distinction between inpatient and outpatient infections is not one of the first party. There are a lot of other priorities. I'm going to show you one other thing here. There's another field here. There's one location to you there's one called location location is very specific to a for B for East for West ICU East, that is very specific for this location, which is great for them for the national level. There's another field called location type, which is the same thing but it's been simplified and standardized inpatient outpatient nursing home, farm, you know, restaurant, etc. Okay. So, I told you how the WTO allows you to do inpatient versus outpatient, but to do that you need the location type and you need the date of admission. A lot of people do not have that. So they should do something which is reasonable. It's not perfect. But if you do not have the date of admission, what you can do easily and say, I want my outpatient samples. You can also say I want my inpatient samples. And let me just put it, I see you now I see these are different kinds of inpatient samples. So if I do this it will give me all of my inpatient samples. So that is a nice easy way to say yes, here are my statistics for my outpatient samples. And here are my, here are my results for my inpatient samples. Just keep in mind that the inpatient samples reflect a mixture of inpatient and outpatient infections. That's an important distinction. An inpatient sample is very easy. It means you took it from an inpatient. Infection, that's that's the more complicated question, because many community infections are diagnosed during the hospitalization. So if you do not have date of admission, you can use this. This is not as good as using the date of admission, but it is still meaningful. You know, for the inpatient samples, how many of them are hospital day one and how many are hospital day three, four and five. So in short, if you have, if you have the date of admission systematically, use the hospital day. If you don't have that, you know, if you don't have that just use the location type, and that does allow you to easily separate the community samples from the hospital samples. Other questions, please in a bit over one hour so far and just watching the time. Okay, may I ask you one question. Yes. Regarding one per patient. Yes. Okay, suppose we are interested to analyze for intro bacteria that means there are a lot of organisms within this group. So in that case, you know, one per patient. Suppose a patient may have a source and equal eye. So, I don't know how it consider which organism, because both of them are intro bacteria. Yes, I understand, I understand. And one, when I said, when I slipped for patient first I slipped for patient. I lied. Look here at the top of the screen. Which isolate of each species. So when I said first I slipped for patient. What I really meant is first isolate per patient per species. For example, if the patient has E coli and Cepciela and pseudomonas and staff. When I say first isolate I mean, first E coli first Cepciela for pseudomonas. So does that answer your question. It's the first isolate of each patient per species. Okay, thank you. Okay. So what you can do on the screen I can do it by isolate. I can do it by patient. I can do first isolate, or first isolate with antibiotic results, because obviously you don't, you don't do antibiotic testing on everything. Or you can do average resistance most resistant most susceptible. So the pharmacy for the CLSI recommendations for annual annual statistics and this is the same as the European, the Europeans don't have a document about this but that's why I'm referring to CLSI. So for annual anti biogram annual statistics preparation, the city, the CLSI recommendation is first isolate for patient per species. Or first isolate for patient with antibiotic results. But for infection control purposes. We often use this other option called by time interval or resistance phenotype. So it let's assume that patient, let's assume that a patient has MRSA in January, and then MRSA in October. For my purposes of annual anti biograms, I don't want to get the patient is two episodes. Patient is MRSA in January, patient is MRSA again in October. When I am doing my annual statistics, I only want to count the first MRSA. Because otherwise I'm counting the people with multiple episodes, multiple times, which introduces a bias patients with multiple episodes tend to be long term ICU sick people with complicated medical histories. So if the patient has MRSA in January and in October, for purposes of annual antibiotic statistics, you take the first one, you take the January one and you ignore the October one. But that's the pharmacy group. But if you're working with infection control group. If the patient has an MRSA episode in January and they have MRSA again in October, the infection control people often want to count that twice. That is two episodes of MRSA infection. The MR, if the patient MRSA bacteria in January, and MRSA bacteria in October, the infection control people want to count both. So when it allows you the flexibility of doing it both ways. For example, and here, there is no standard agreement so people do what they want. For example, I would like the first isolate. Every 90 days, if the patient had MRSA in January and February. I just want to count the January isolate. If the patient has MRSA in January and June. I want to count both isolates personally for a lot of our own research. We take the first one per year. The patient has MRSA in 2017 and 2019 we take both. If the patient had MRSA in December of 2018 and January 2019. I just want the December one, because it's within my time window here. So for pharmacy, normal clinical statistics for normal statistics, I recommend by patient. The first isolate only your first isolate with antibiotic results. But, but if you're working with an infection control audience. Often they would like to date. They're not so interested exactly in the patients, they're interested in the episodes. And so these are different, these options are useful for different people. Okay. Okay, maybe additional question. Yes. If I want to count the number of carbapene and resistant organisms. I don't know how can we do this one specific to let's say for intro bacteria. Sure, sure. Okay. Before I answer that I just have a few more comments about this one isolate per patient. In order to do that who needs a patient identifier. If, so let's see. Maybe my medical record number is 12345. I come back next month, I come back next year, I come back in five years. My number is still 12345. That is a medical record number. The medical record number depends on the hospital. Usually. So if I go to different hospital, I get it usually a different medical record number. But it's always because there are some countries where there is no medical record number they use the national identification number. So if you go to any hospital, they will always use the same national identification number. So these are useful numbers. And most of the time we're not, most of the time we're not trying to track patients between hospitals. It makes it's difficult when they get a different number. So I'm focusing now on removing repeat isolates inside of a laboratory. I really like a patient number, because what happens if you don't have a patient number. It's hard to know if the same if it's the same person at its extreme and this is very common there is no medical record number. If there is no medical record number the only thing you have to go on is the patient's name, or maybe the date of birth and the gender. This would be important for you to explore in Ethiopia, does every laboratory have a medical record number. If they don't have a medical record number are they leaving it empty. Are they putting in the patient's name. One thing I do not recommend and I do see this many times is, if there is no medical record number. I put the specimen number as the patient number and that is not correct. The specimen number identifies the specimen, the specimen number does not identify the patient. So the patient comes back next week. They're going to get a different specimen number. So please do not put the specimen number as the identification number. If there is no identification number leave it empty. If that sees an empty patient ID, who that will automatically use the patient's name instead. Using the patient's name is not perfect. Of course, different people will have the same names. And of course, even the person with the same name. Sometimes they'll type it differently. Like my name is John Michael Stelling, but a lot of times I'm just John Stelling you don't always type my full name. The best thing is to use a patient medical record number if you have that great. If you don't have that the patient name is pretty good. It's not perfect of course different people might have the same name. But it's not often that two people will have the same name and the same bacteria in the same year. If you have a patient, you know, Mary Jones in January with Serratium marcessons, and you have a patient Mary Jones in June with Serratium marcessons, it's probably the same person. Once in a while you might be wrong, but it's not going to change the statistics in any important way. So it will be important to explore and maybe you know already and if I stop talking maybe you can comment. I want to understand whether or not you have a reliable patient ID. If you do not have a patient ID, then who in it is just going to use the patient's name, which is pretty good that it is not perfect. A lot of people have something else which is sort of useful it's not perfect but it is helpful. A number of people do not have a patient ID, but they do have a hospitalization ID. The patient is hospitalized in January. All of the January samples will have the same hospitalization ID, which is very useful for getting rid of the repeats during that hospitalization. But the patient comes back in October, they're going to get a different hospitalization ID. The patient ID is the best hospital ID is not bad it's not perfect hospital ID is good, because it allows you get to get rid of the repeats during one hospitalization. So I'm going to stop right there on this question. Do you do you know what you have. Do people have medical record numbers. Do they have a hospital ID. I don't know that they heard your question but do you came through your, yeah, you're good. But I don't know. I'm going to continue then. If you have a question say it or go to the chat window. Okay, so I'm describing you these wonderful features about by isolate by patient by time interval, but they only work. If you have a reliable patient ID, or a semi reliable patient ID like the patient's name. So now going to continue. And you had a question about CRE. Okay, I'm going to show you data from the temperature test hospital. And I already know that these are old data and they're not, there are no CRE. So the first thing I'm going to show you there will be no results, but I will show you. And then I'll change to a different antibiotic. So I'm going to go to analysis type. It depends on what you want you said you wanted to count them. I go to isolate listing and summary for 100% resistance among CRE. We do. I don't want that right now. I wanted to see the people who have CRE. And I want the list. I want the summary. I want both I want both. And I'm going to leave it at that I want and I click on okay. Organisms. I can choose E coli soradia, or I'm going to organism groups. There is an option here called all into bacteria C. So you can do all the into bacteria C at the same time. And I click on okay. I click on data files. And we give people one month of sample data for teaching purposes. So that's what I will show you. I'm not looking on okay. So, and I'm just going to begin analysis. So this is a list of everything. It's so here, let me go to the me Penham column. So me Penham, 27 millimeters, 29 millimeters, 26 millimeters, all of these are large sensitive results. I'm going to click on me Penham. And you can see the smallest zone diameter is 18. So actually, you know, at that time, I forgot about this at 18 at that time was considered sensitive. In fact, I think 18 is now intermediate or resistant. So in fact, this will work. But so this is one way I don't recommend this particular way. But what I've done is I've gone to me Penham and I've sorted it. So I see 18 1920. So these here are going to be resistant or intermediate. I think one way that I do not recommend. Why don't I recommend it because I have all the sensitive ones I have the resistant and the sensitive ones mixed together. I'm going to click on continue. Here is my summary, I see that I have 86 E coli from 71 people, but that is a mixture of sensitive and resistant ones. I'm going to click on continue. So basically I did not answer your question yet, but I'm showing you how to do isolate listing for all introductory ACA for the sensitive ones, and the resistant ones. I'm now going to go to isolates. And here I see patient name age at birth but at the bottom I see the antibiotics. And I see in me Penham. I'm going to go to me Penham. I double click, or I clicked on define criteria. And I can say resistant. Or I can say non susceptible non susceptible would be the resistant isolates and also the intermediate isolates. So it depends on what you want. Do you want resistant or do you want the intermediate. I'm sorry was there a comment. Okay. I will do the resistant ones. Also, you can say greater than less than equal. You know so you can look at high level resistance, less than eight. You know from. So there's a lot of options here. You can also say tested or not tested. But so resistant. Yes. Yeah, that's okay. Great. I'll click on okay. And, and that's it I'll just say in me Penham. I can click on okay. And I will now begin this analysis will not show exactly the same results, but only the isolates that are including the minimum resistant. So in this database, there were only six isolates. Those six isolates came from six different people. Here I do like to see the repeats, because a lot of times, it will be three times from one person. So I'd like to see, I'd like to see all of the isolates. In this example it is six isolates from six people, but it could have been six isolates from the same person. There's one reason why I like to see all of the isolates. I already mentioned some of those reasons. I want to see what rooms were they in did the patient move to the isolate going from sensitive to resistant or resistant to sensitive. There was a blood urine. So I mentioned these reasons why I want to see the repeats. There is also another reason. The CRE was very, very, was very, very rare. It almost did not exist, except as a laboratory error. In a stable drug, in a hot tropical environment, the imipenem discs have a tendency to degrade. So we used to see a lot of CRE, but it was false CRE. It was just the imipenem disc was bad. So if I see imipenem resistance from one person, it might be a mistake. It might be true, but it might be a mistake. If I see imipenem resistance four times from the same person, that just reinforces that this is a real finding. Okay, so by having the repeats, it allows me to just re, re-confirm that this is probably not a laboratory error. If it's the same patient over and over, it's probably true. And that's another reason why I like to see the repeat isolates. Okay, there is something very interesting about Morganella morganii. So there's something called intrinsic resistance. And for some reason, Morganella morganii, they don't call it intrinsic resistance. They call it intrinsic decreased susceptibility, which is basically the same thing. So I'm actually not surprised to see the morganella's here, because the morganella's have always been a little bit resistant to imipenem, to meropenem. So this confirms to me that the system is working. It is, morganella is more resistant intrinsically to imipenem than other enterobacteriaceae. So here, Proteus Mirabilis and Proteus species. I completely forgot about this. This I'm going to Google right now. Proteus also has intrinsic resistance. Proteus, Proteus imipenem resistance. I won't go into details on this, but both Proteus and Morganella are organisms where there is some resistance, which is ancient. And I'm not sure if their E is modern, but for these two organisms, it's always been there to some degree. Okay, I didn't want to spend too much time on the microbiology. And you asked me how to do something mechanically, but I am interested to see that it's only the Morganella and the Proteus that came up here as resistant. I'm not going to click on continue. And I now get the summary. And I see there were three Morganella isolates from three people. Two from two, one from one. So it is counting the number of isolates separately from the number of patients. In our example, it's one to one. So the numbers are the same. But what you might see are six isolates of Morganella coming from three different people. And then very valuable, I can see January, February, March, these are the number of people. So this allows you to look at the growth over time. And then you have more CRE. So as we start to get into feedback reports, standard reports, you can do this on a monthly basis. I'm working now with the Vietnamese on this kind of thing. They already have these nice monitors about data volume, the data volume went up the data volume of total samples went up down data completeness went up data completeness went down. And they did these not for epidemiological reasons, it's more for project monitoring. Are they doing data completeness well, are they do they have is the data volume, if the data volume goes from 600 isolates to 620 to 580 to 6. That means something's wrong. So the data are missing. So the Vietnam have started to do this, but so far they're not doing it for scientific resistance issues. So we're working now on that. But that's what you can see here. If you did this kind of analysis monthly for CRE, it can help you to find a possible outbreak. Okay, continue. How do you measure data completeness. Okay, great. Let's see. That's the next question I want to continue a little bit on here. I'm going to ask for any pen and resistance. But there are a lot of people who do any law, a lot of us are doing any pen and other laboratories do Mara Penham. The laboratory I'm showing you here does not have Mara Penham, you see it goes in the pen and mesla salon. I'm going to take a quick little detour and robotics, Mara Penham, Mara Penham. Good Mara Penham, good, good, good. Let's put it together antibiotics, Mara Penham, Mara Penham. And I'm going to move up. And so I want to put it in alphabetical order. Good, that's perfect. Okay, save data analysis. Okay, so here, I can say I want the isolates in the pen and resistant. And then I could put Mara Penham resistant. And let me just repeat the same analysis. I go to isolate listing and summary, okay. And I say, EBC, all into factory ACA, I choose the same ones of sample data. I say, okay. And it says it found none. There are no isolates. But what happens to the what happens to the Morgan Ella. And that's what I wanted to explain here. EBC, in the pen and resistant Mara Penham resistant. I want to draw your attention to this option at the bottom of the screen. Include isolates that satisfy all of the selection criteria. In other words, I am now asking for bacteria that are resistant to any Penham, and also resistant to Mara Penham. We didn't test Mara Penham. And that's why there were no results. None of these isolates are any Penham and Mara Penham resistant, because we didn't test the Mara Penham, or more precisely, they are Mara Penham resistant but we didn't, we don't know that. So, so here I'm saying I want any Penham resistant. That's fine. I want Mara Penham resistant. We don't know because we did not do the test. And that's why when I choose this feature, there were no isolates because it's looking for bacteria resistant to both of these drugs. There's another feature here, include isolates that satisfy at least one of the selection criteria. What this means is resistant to any Penham or Mara Penham. Now when I run this analysis. I see the six isolates. All of these bacteria are resistant to any Penham specifically, but there isn't any Penham or Mara Penham, but but of course there were no Mara Penham results. So I'm going back to and that's the summary the same thing here at the top isolate satisfy at least one of the following criteria, any Penham resistant or Mara Penham resistant. It's important at the national level, because some of your labs will do any Penham. Some will do Mara Penham. Some will do sephraxone, some will do sephotaxin. Some will do Cipro, some will do levo. So, I do recommend standardizing it because it's just easier for you if everybody just test any Penham. If everybody, if everybody just any Penham, it's a lot easier, you only focus on any Penham. But if some lives to any Penham, some lives to Mara Penham, who net can still handle that, but you do have to do some more work to combine them together in the way that for example in the way that I just did. Okay, so you asked and this often. Yes. Question. Yeah, can you use the same analysis using specimen type urine. Just these analysis but using urine specimen. The first thing I'm going to do is let me get rid of the Mara Penham. Sorry, just make this clear this criterion. Okay. Okay. The reason I get rid of the Mara Penham is I want it. Okay, well, I want to include isolate that satisfy all of the six solution criteria I want urine and any Penham resistant. So, so this is resistant. And now I'm going to go to specimen type specimen type, and I'm going to say urine. There are in fact are several kinds of urine. There's urine urine from ladder. So I'm just going to choose all of them. So this is all of the different kinds of urine. So now I want bacteria that are in urine and also any Penham resistant. I removed the Mara Penham. Maybe if you have, maybe, maybe in without moving, you have actually only in Penham, but in our country, some hospitals, they have Mero Penham. EPHIs are using in Penham and sometimes Mero Penham. We do have actually all, all Carbopenham drugs. In that case, maybe I don't know how can I explain for you. You know the conditions all and end. I don't know we can use. John, can you hear me. Yes, I hear you perfectly. You have a very important question. And I can tell you the things that when I can do now. I can tell you the things who that can do in the future. So, um, I removed the Mara Penham because of the point that you just made. Because if I say I wanted to be specimen type will specimen type equals urine. And resistance to any Penham or Mara Penham, who that gets confused. It doesn't know what's and it doesn't know what's or. Um, so if I want to do any Penham or, I'm sorry, if I wanted to, let me start with the simple case. This case is going to work. Specimen type equals urine. And also any Penham equals resistant. This is going to work. Yeah. And it's on three. So these are all urine. And these are all any Penham resistant. So here it's this. And yeah. And if I try to do. Okay. But if I try to do. Okay. If I go to Mara Penham, and I say resistant. This analysis will work, but it won't give you what you want. This analysis is specimen type equals urine and any Penham equals resistant and Mara Penham equals resistant. Or, but they won't find anything because we don't have any Mara Penham results. So what you want is this and this or this. And so I'm going to try to do that. And, and, but as you can see, I have the choice of and and I have the choice of or was it not clicking. Okay. And if I click on this, it's saying I don't know what you want. What do you want and what do you want or. So I know exactly what you want. And there is a way to do it, but not in the way that I showed you the way that I showed you what we really would like to do. I'm going to take a little detail on macros macros are wonderful. I'm going to click on macros. My mouse is frozen. I must have accidentally clicked something. So my mouse is now frozen, but I can still use my finger macros. So no more mouse today, or I don't know if I have to reboot or something. Okay, so I'm going to go to new macro. And I'm going to call this urine and CRE. I'm going to click on save and save. Here's a reminder, what is a macro macro means if I leave who net and I come back next week. I can go to data analysis, go to macros, I can go to urine and CRE and load. So it simply is a way to remember what I was doing already. So if I do the same analysis every week, the same analysis every month, the macros make it very easy to do that. The macros themselves are very simple. I'm going to click on urine and CRE. I'm going to click on edit. So here what you see is isolates equals urine. And here, and I see any Panama is resistant. And here you see the or aspect. So it's this or this or this. Basically, what I would like to do in the future is allow you to mix and an or. Okay, so if you want to do and or as one step. Who net does not currently permit that. But we can do it as two steps. And I do this a lot for my own projects. When we were looking for outbreaks. I want to look for outbreaks on the CRE on the in patients, hospital day three and later. So I wanted this and this. But any kind of or merit venom so I do exactly what you have described, but I do it in two steps. I will now show you how I do this in two steps. So I'm now going to show you how to do and and or as a two step process. Step one is to get a data subset that I'm interested in. I'm going to go to analysis type. I'm going to isolate listing, and I'm choosing isolate listing. Okay, good. I click on okay. Organisms, I say EBC, for all into bacteria, I say, I say, okay, data files, I choose my one month of sample data, I say, okay. And then I go to isolates and I say any pen and resistant or merit venom resistant. I'm going to run this and it is going to give me the six. Okay, and I forgot to change this to or. So we're going to see the six results here. So great. These are the six CRE that is step one, I find my CRE. The secret here is I don't want to and again my mouse isn't working. I don't want to output this to the screen I want to output this to a who net file. I'm going to use debase. I'm just going to choose debase. And I'm going to call this a file called, you know, enter back to your CRE 1995 or something. Dot TST or dot DBA dot TST is fine. Who net has just run that analysis that I showed you on the screen. So instead of showing you on the screen I outputted it to a new file. That new file indeed is a who net file. Oh, but it's in a different folder what folders it in. Okay, you see here it says Vermont that's because I'm in the middle of something else. Let me just change this to who net. I was doing it I'm doing a manuscript so I changed my output folder. Let me redo this and I'm going to output to the who net folder. And now here in the who net folder output. So here you see where is it. See, there it is. So you see this file. I'm in windows when it output. You see this file here called ABC CRE 1995 test. That is the output of this analysis of the isolate listing analysis, but it is also a who net file, because the listing is basically the who net file is just a list of the bacteria. So instead of choosing my one month of sample who net data. I can choose this ABC CRE. Let me say isolates I want all isolates. Let me put this on the screen. Let me change this to all. So I shall listing all organisms with this new data file begin analysis. So what you can see is the complete content of this file, or the CRE enter bacteria. That is step number one. Does that make sense. We took a very large who net file, and we extracted a small who net file out of it. I could have extracted blood. I could have extracted. I could have extracted ICU. I could have extracted any pen and resistance which is what I did. So step one is to make a file that has what I'm part of what I want. That's step one. And now that I've done that. And I say all all all, it has those six isolates. I will now go to isolates. And now I will go to specimen type. I will say, you're in. Okay, okay, begin analysis. And now I just see the three. So this is exactly what you asked for, but I had to do it is to stop. Does that make sense. Yeah, thank you. Yes. In the future, it would be nice if we had a nice macro editor and and or or and. And I would like to do that eventually, but you asked me now and this is what you can do now to answer that question. Okay, thank you very much. Sure. Next question and we have about, you know, little less than 20 minutes left. I'm ready for another question or you can give me guidance on what you would like to see next. And I had a question on how you measure data completeness. Thank you. I forgot about that. Okay, great. Good. That's an excellent question. And I'm working now. So the Vietnam. So, so Vietnam took a lot of good ideas from who that they copied them. They expanded them, and they made the formatting wonderful. So they started with who not but then they went beyond what I did. Now what I am doing is taking some of their ideas and getting them back into who not. So let's see. I want to show you what we are doing with Vietnam. But before that I'll show you where we all started going to click on exit. So there's this feature in who not called data analysis, quick analysis. And then there's quick analysis. So here you can see there at the top are two standard reports that all who net users have. At the bottom, you see user defined reports VT is the code for Vermont. I'm working on a manuscript we just had it accepted preliminary accepted I just so so you see here I've done this special thing for Vermont. I'm going to spend our time on the standard report, but just to show you what I did for Vermont. I did one analysis for step forest resistance. And you can see here I did it for this for the full state. I did it for the long term care facilities. I did it. Overall, I did it by laboratory I did it by blood and urine and age. So, by doing these macros my user defined macros. I can say this is what I want. I'm going to set that up. But then I can do the exact same thing every month in a very quick and easy way. That's, we can discuss those later. I want to focus on the user standard report. So if you look at the unit standard report. The things I like about it. It's easy. You don't have to do anything. That's what I like about it. It's a lot of valuable information in a quick easy way. The two things I do not like about it is the who net standard report is not customizable you can't, you can't change it. It is what it is. It's a standard report that you cannot change that's one thing I don't like. The other thing I don't like is the formatting is not very attractive. So what I would like to do is merge together the features of the user defined reports with the standard reports. The difference is historical the who net standard report I wrote that 20 years ago. And the user defined reports I wrote around 15 years ago. So now with the Fleming fund as the new priority is to make a better looking configurable standard report, and based on a lot and expanding the report. That's the other thing I wanted to expand it. So let me go to the who net standard report and go to data files and as you can see under edit. There are different sections there's a summary statistics alerts alerts, etc. data files. Let me choose. Let me go change back to my other folder my data folder. Okay good. So this is the standard report. That is 20 years old, and it needs it needs an overhaul. Great. So, you see section ABCD a is the summary. It simply tells you. There were six, there was one laboratory in that data file. There were 622 isolates. The isolates in that data file went from January 1 1995. January 31 1995. That's correct it is one month of data. So you need to find typing mistakes sometimes people put the wrong year, they put the year 2029, or they'll put the year 2002. So this helps you to find errors in the dates. So section a is a high level summary section be it's giving you these ideas about completeness laboratory 100% complete location is complete department identification number one of the isolates, the identification number was missing. So this is one area where you can look for completeness. Then you have another column called invalid. You know, is it required or not a required field. The following fields have no data to a large degree that's because of confidentiality I got rid of them. And then we have the detailed statistics. 7% were cardiology 7% cardiac surgery emergency room ICU. So these are helping to answer your question about completeness. In Vietnam one thing I am recommending and also Sri Lanka when I'm having the same overlapping conversation with the number of groups. I think it also makes sense, not only to put completeness field by field by field this is what you see here, but to also give them a score to give them a score on completeness. But I want to give them a score on completeness of every field. I want to give them a score on completeness of the most important fields. If they don't put, and that's a, so if they don't put the beta lactamase, well that's fine it's not always relevant. Even if it is relevant you don't have to do the test. Things like, and there's certain things like medical record number I want that to be complete. And birth and gender, I would like that to be complete, but how realistic is that in the short term hopefully realistic. So I think that patient ID, and age and gender would be good fields for scoring. Something like date of admission is not realistic for most places in the short term. So I would suggest that out of the fullest opponent fields. Out of the fields that are relevant at the national level and try to standardize those. And among those fields relevant at the national level, which is the minimum core set that you want to score them on. And then you can say for that core set patient ID, date of birth or not date of birth but age. If we have patient ID, date of birth, age, and gender, we have, if we have 100 records, there are 300 possible values. You know, the ID, the age and the gender of those 300 possible values, they have entered 250 of them. So to do that, you would need to decide what are the core fields you want to use for that high level completeness score when it already gives you the field by fields completeness score. But some of these fields are more important than others. I don't want to penalize them if they didn't enter the first and the last name. I don't want to penalize them if they don't enter the beta-lactamase result. So does that, does that help? That's excellent. Yes, thank you. And the fact also applies to antibiotic testing. Of course, I would love the hospitals to all test 15 antibiotics, and I can recommend test these 15. The reality is, I want them to test at least these five. So if they have 100 E. coli, and I want them to do these five, I can do, who not already allows you to do the percent completeness for each antibiotic. You know, you had 100 E. coli, you have 80 ampicillons, means ampicillin is 80% complete, so who not allowed you to do that already. But we would like to allow you to define a core set for scoring purposes. So if you say, well, we want you to test these 12, because you know, 12 is a common number, because if you have one big plate, or two small plates, 12 is how many fit. But a lot of places only test one panel, one plate, and that's six. So you might want to recommend the test 12, but you may want to score them on the six. For E. coli, I want ampicillin, cotrimactis, all Cipro, genomycin, and whatever else. As my first line core testing, and then you can give them a score. Among these five antibiotics, they've done 90% of requested. So if you have supplemental drugs, like any panamamic acid, these are often second line drugs. There's first, so there are two concepts, there's first line testing and second line testing. There's also first line reporting and second line reporting. What we do and what a lot of people do is they will test 12 drugs, but they will only test, they will only tell the doctor six of the drugs. So this, if you have an E. coli and urine, sensitive to all 12, you have an E. coli sensitive to ampicillin and also sensitive to imipenem. We test imipenem, we always test it, we test all 12. But I don't want to tell the doctor the imipenem result, because the doctor might not know it's more expensive, it's a reserve agent, I want them to get an ID approval. This is an example of first line testing of 12 drugs, but first line reporting of six drugs. So I always test by to selectively report. There's a different concept of selective testing. Day one, I test six drugs. If it is resistant, then on day two, I will test some more drugs. I don't like that, because for epidemiology, the more data the better. So if I want 12 drugs, I try to test all 12 drugs in the first day. Anyway, that's different, the scoring issue. I don't want to score them on the five key drugs that I want them to test. And that they're going to be scored on that. I also may want to score them but not penalize them on the supplemental drugs that we recommend, but we do not require. If you have 300 E colise, and five core antibiotics, you have 100 E colise, and you have five key antibiotics, there should be 500 results. So you can score them on the minimum required sets. So I'm suggesting basically define a minimum score, a minimum set of data fields that they will be scored on, as well as a minimum set of antibiotics that they will be scored on. And that will be different for the different species. Does that help. Yes, that's good data fields and antibiotics. Yes. I'm now going to go to, and I will email this to you. What am I doing here? Who not develop wrong folder. And here, and going to who net development. This is our active programming area and standard. Oh, I'm on the wrong laptop. I had this laptop for repair because the monitor was broken. They, they said they could not fix the monitor. They got it back two days ago and the monitor was working. They lied to me they fixed it with some other fixed it accidentally. So let's see. And I'm going to go to email and emailing myself and standard report. I'm taking the time to do this because it's a very important discussion. So I'm going to email this and see drive and I go to development and I go to standard reports. And if I go to, yeah, I'll just email all five of these. Hopefully none of them are big. I don't want to slow this down. And what's the fun with the latest one year 11 of five feedback ideas for discussion. Okay. And let me click on that standard. And I very much welcome your input. First of all, into the content of the report, the formatting of the report, I'm not ready to act on yet. I want to make sure the content is good before we get to the formatting. So I've now emailed it to myself. Let's see how long it takes to arrive. And this, this, I just restarted the one of the files was big. So I just reset it. And hopefully it's been mailed. Good, good. It has been mailed. Now let me go on this laptop. Look, when that arrived momentarily. Okay, so in short, I think me kill anything for an I think I may have already sent to you. Some reports from Vietnam and Australia and Japan. Okay. So I'm going to go to the hood at standard report. One file I just sent. I think you send the Australian one and then the Vietnam one to I believe. Yes, you're right. Okay, so this is based on that, but I didn't send you further. I email myself the wrong document I sent a word document. I'm going back now to here and I'm just going to join our call from the computer with the bad microphone. And I will turn off the volume on that go to meeting. Okay, providing a time check here in about three minutes left. Thanks. Yes. Okay. I've turned off the microphone. Which microphone do you want to use the I'm going to make you present it on the other computer. Can you speak into your other mic please we can't hear you. How was this. Let's not discuss it. Okay, I left that. Anyway, there's not enough time, but basically I've started to put some thought into precisely. Oh, can you hear me. Oh, yes, we can. Okay, good. So I'll send it to you and I think this could be a very good discussion for what should be in the standard report. Different ways of scoring different measures of completeness looking at data quality, microbiology quality. So, the, so the Vietnamese took my ideas and expanded them. Now what I've done is I've expanded them further. We have not implemented it yet. We're trying to define what we would like to be in the new standard report capturing all of these ideas. We basically two kinds of facility report. One report would be a data submission report. Somebody sends you the data for March, you analyze thoroughly in many ways the data for March, and you provide them a feedback on their submission. Also for the facility we want a temporal trend to say well, the data volume has gone up the, the completeness has gone down the, the, there are possible outbreaks. So these are the trends and CRE. So we're looking for facility specific data quality feedback on a monthly basis, but also temporal trends with data volume and completeness and CRE and outbreaks. Those would be facility specific feedback submission reports and temporal reports. I would consider a different set of discussions about national benchmarking national feet national monthly reports and national time trend reports. So, the, I'll send you this file, which has taken the Vietnamese ideas and put more new ideas into it. And this would be a good subject of discussion in the future.