 Good afternoon. Good morning, everyone. Thank you for joining the call So as last time else, I will start off with some fairly standard didactic presentation But also feel free to jump in at any point with questions We might stay with the pure didactic for a while. Oh, I need to click on share Hold on Share great. Hopefully you can see my screen So I will start off with the normal lunette presentation But feel free to jump in because especially a lot of you already have experience and I want to make sure that your questions get answered a Lot of times when I do a basic presentation I do focus on the high priorities the most important points. There's limited time people have different skills But you're a lot of your past that so sometimes I might skip over points that might be important for you If I do please stop me and I'm happy to address those as well So I thought that we could start off with just normal data entry and see If we want to go through all of that or take a break and move on to something else and then I could start something similar with data analysis I like as much as possible the idea of working with your data Many of you have been trained with my data set. It's going to be the same points or Even if you have not trained with my data sets Your data will have different issues and so we'll I would try to incorporate that into the discussion of data analysis Great and the sound quality. I hope is okay Yes, he actually very good. Thank you. So continue But it's a problem. Everybody's on mute because sometimes I might talk for a long time not realizing that nobody's there Okay, I do need to move to go to meeting on my screen because it is obstructing my slides How do I move this? Okay, I'll put in the lower right-hand corner Let me try to move this out of the way It's a little ironic that you can see more of my screen than I can Okay, I have now moved the meeting to go to meeting controls out of the way Okay, let me go back to the beginning for slide So when a data entry there are basically two common ways to get the data into who net I'm who net has three modules interesting is data analysis Coring is data entry and before that you do lab configuration to get everything set up We've covered a lot of configuration on the last call. I will if you have questions I want to enter those but let's hold off so we can get started on data entry So it you need to get the data into who net the two ways to do that are by manual data entry And that's will be our focus for now somebody takes their paper notebooks They go to the who net screen and they just manually type the data out of the notebooks into the who net screen The other approach is to download the data from an existing system so if you have Something like polytech or meditech or Sunquest or a vi-tech or a micro scan or a phoenix or Excel or some other source of data In a sophisticated system or a basic system if you already have some of the system We do not want you to manually retype the data into who net Some people do because they do not know about backlink or because they don't know how to export the data from their own system But wherever possible we want people who have it systems to use backlink to get the data into who net in that way You can avoid the double data entry you can often automate the process so you can have automatic exports and automatic backlink So those are the two ways to get the data into who net manual data entry out of notebooks or Downloading the data from existing systems in this call We will focus on data entry and it's a later point. We will focus on backlink So we will cover five points part one is how do you create a new data file? Just like in word just like an excel if a file does not yet exist You click on new word document or new excel document or new PowerPoint document similarly in who net you will say create new data file Or if the file already exists you will say open data file Then we will go into the core of data entry the manual typing of the data Then we will view the database after you have entered five isolates or 20 isolates or 200 isolates You would like to see what you did Make sure the data is still there. Maybe do some editing Maybe the doctor has given you a phone call and you want to look at the patient results So you can you view the database to do those things? You can also use it for clinical reports Many labs use who net for clinical reporting back to clinicians They type the data into who net they click on print report and then they share the report with the doctors There are things we'd like to do that would make that better and more complete and more But we provide it as is and many people do find that useful On the other hand many labs just already have some way of reporting They have a lab information system that prints up clinical reports or they have paper reports the doctor sends an request for testing The lab writes the results on that request form and then sends the request form back to the doctor But if you would like to use who net for clinical reporting, it is one of the options It's okay, and we would like to make it better and then finishing up to move on to eventually data analysis Occasionally after several slides I will pause to ask if there are any questions. I will not ask at the end of each slide But of course jump in at any point if you do want me to stop and answer a question right there Okay step one creating a new data file you start from net and select your laboratory That could be the WHO test hospital that could be your data that could be some other data configuration So when you open who net you see a list of defined laboratories select one of them and click on open laboratory And then after you do this you will be on the who net main screen and at the top of the screen you will see Data entry data analysis help. So you click on the data entry option You'll see several features there such as new file open file Export data combine data. So we will go through some of these additional features later At the beginning of this presentation the first time we will choose that first option called new data file And here's the menu These slides I think I mentioned on a previous Slide on a previous presentation. We're prepared in India. We've ever made these slides. You're student 5.4 Which I have not seen in a very long time, but everything is almost identical We have added we have deleted a few things. So these slides do need to be updated but Right now we are doing so many big changes That we're just waiting to update these until the new software is Stable with regard to new menu features So you see the first two options are new data file and open data file Of course the first time you open it you choose native new data file Subsequently either choose opposite open data file like you see here Or at the bottom of this menu, you will see the list of the most recently open files Word and Excel do the same thing when you want to open an Excel file You can use open file But Excel and Word are also very helpful They do show you the list of the most recently open files just to save you time in in finding files that already exist I will now continue So creating a new data file a data file has every file you have ever made as two key characteristics What is the name of the file and where did you save it? Sometimes people write to me John. I lost my data and one of the most common reasons they quote unquote lost their data Is that just don't know where the data are. They're in some other folder. They're on some drive They're on you know, they move they change computers and they want to move everything from computer one to computer two So please just like any important word or PowerPoint or Excel document Please remember what you called the file and where you put the file Who net is automatically suggesting a location Who net default location is in the who net data folder? It's a nice simple option It's a nice simple approach that most uses around the world when they save who net files They go into the folder called who net or in this case who net five data So if you do that, it's easy for me to tell you where to look for your files But not everybody does that. Some people put them in who net data And then they create subfolders for different years 2019 2020 2021 so they are within the data folder, but with an additional subfolders that they have created or Many people put the data on to a common network drive I do like that if you have a T driver a P driver an S drive We'd like to have the data on the network drive for two main reasons One is that everybody with who net with the proper credentials and username and password is able to use exactly the same files I can be sitting at computer a in the laboratory and somebody else can be considering a computer B or C or D and The data entry person sits at one computer to enters the data But anybody else who has permission can use those same files So that's one of the reasons why I like a network installation of who net people are using the same files They can use the same configurations the same software So you only need to maintain the system on one computer and all the data are on that computer The second reason why I like data on the network drive is about backups Many people are base are bad about backing up their own data Sometimes people write to me John. I lost my data and sometimes the reason they lost the data is the computer got lost The computer got damaged or the computer got stolen The computer got reformatted I can't we cannot retrieve those data the data are physically gone Of course no matter where you save the data you should be backing up your data Backing back them up to a memory stick back them up to a CD drive Back them up to another computer So no matter where you decide to save your files make sure that you have a good backup strategy Many people do back up their data on memory sticks and CD ROMs Keep in mind that is not so secure because memory sticks and CD ROMs also get lost They get misplaced and then whoever finds it might be able to access your patient confidential data If the CD ROM or if the past or if the memory stick does not have a password So I please insist please have backup strategies when people tell me they lost their data and it is the data are lost It's heartbreaking. They put in so much effort to put all those data in and then the data have been lost Another common reason people lose the data is they had somebody there for five years They managed all the data and then that person leaves one year later Somebody else comes in to want to continue and they just can't find the data It's not that the data are lost, but nobody knows where the data was stored previously So please have a strategy for backing up your files. I Mentioned that I do like the idea of having data on the network drive The reason for that is network drive IT people are usually pretty good about backing up the network drive If you keep the data on your local hard drive, it is your personal responsibility to have a backup plan But if you keep the data on the network drive and if you trust your IT department, hopefully they have a good backup strategy So if you did lose something talk to the IT department and hopefully they can get it back for you So if you're using who not on a single computer Most people just put the data into the who net data folder or maybe they create some subfolders either by lab or by year Or both but if you are in a network, it does make sense to put the data onto a network drive So that's about the file location. Where are you saving your data? And then what name should you give to the data file? Give it any name you want any valid name for Windows when it does suggest a very simple name for example W for who net? 2020 for the year Ethiopia or ETH for the country dot Hospital 001 hospital BWH hospital, you know, whatever your laboratory codes are so who not suggest a nice simple name for You know for the file But you can give it a different name. You might want the data by month You might want the data by blood or by urine or by project. We have a two month project Let's put the data for those for that project in pneumococcal study into that one file So please leave names that mean something to you don't give a file a name like data Because you don't know what year what lab Try to have the name of the file be meaningful to tell you what years it is what project it is what months it is There's a small point. I'm going to come back to later I've often find it easiest to manage the data by year. You have one big file for the year In January, you have a file called 2020 in March You have a file called 2020, which is the same file, but it's bigger So at the beginning of the year you just create one file and through the year The file gets bigger and bigger and bigger. So at the end of the year, you have 2020 complete It's a nice easy convenient way that time every time they send you the 2020 data You simply replace the one that they sent you previously Having said that I personally don't usually replace it, but the one they sent me previously I just put an archive folder if they send me something. You don't like to delete it But I'll just move it away. I often have one folder for archive Permanent records of what people have sent me, but then I'll have a different folder for active analysis and cleaning That is my my the one I use for analysis Okay, a small point. I will come back to you later as many people do save their data by month They have a file for January a file for February a file for March. Nothing wrong with that It does mean you have a lot of additional files to track One small recommendation many people when they have data for January They will put the name JAN in the file for example 2020 January Or they'll just type the word January They will type the word January the word February or they'll put abbreviation like SEP for September I do not recommend that The reason is just out of convenience I have a simple question, but it's a trick question. What is the first month of the year? Of course, many of you will say January But that's not true in alphabetical order In alphabetical order the first month of the year is April Followed by I'm not sure August. So when you use the word January February March When you look at Windows the files will be listed in month order alphabetically So if you see that you have ten months and two of the months are missing It's easy to know which month is missing if you use numbers Zero one for January zero two for February. So if you're using the months then it's easier It's easiest to know which two months are missing. It's a small thing But in the long term I suggest avoid the word JAN and just put the number zero one Or you can involve avoid all of this completely by just having one big file for the year 2020. I Don't have a strong recommendation. There are good reasons why people do it different ways But you should come up to some standard and try to get the labs to comply with those standards It will simply make your work at the national level easiest I don't want hospital one to do one thing hospitals or two to do another thing hospital three So try to standardize it as much as possible. It just makes the national data management easier easier to find mistakes For example or missing data So in who net you give it a name you give it a path and you click on okay If the file exists already who now will warn you this file already exists. Do you want to replace it? So usually for new data file Usually you don't want to replace an existing file So if who not warns you you're about to get rid of something Please take a moment to think about if you want to delete something that exists already Yeah Great. So and here's what I'm talking about The screen actually has been modernized, but here it says W for who net zero one for January zero six for the year 2006 the WHO test country and the WTH You know WHO tutorial hospital as an example file name and you can save the data by year by month Or you just type something else by default you see the data are saved in the who net data folder Good so now you would be in the who net data entry screen, which is probably on the next screen here Yes, so this is what the data entry screen looks like at the top You see the first question is human or there is origin, but you can change that to human animal or food You put in the identification number last name first name sex etc You put in the location details in the second box Specimen details in the third box Microbiology results including the organism antibiotic results and then finally other If we were sitting down together in a big room during a live training course I would ask you to manually enter these data in so that you could actually just practice yourself with data entry We're not going to do that now because after the PowerPoint we will do a live data entry So so I'll just show you what they did here. Please type in the patient ID 1 2 3 4 5 Patient name John Smith mail birthday neurology specimen number specimen date blood Steph or is it'll act amaze positive and then the zone diameters for the distifusion results as It says here on the slide as you enter the zone diameters When that will automatically tell you if it is resistant intermediate or susceptible because who net knows the breakpoints Who net does not require zone diameters who net does not require m. I see values But we do strongly recommend them and I will come back to that in a few more slides When it does accept the letters are I and desk if you do not have measurements then who net has no trouble with that But the quality of the data is compromised There's less you can do if you do not have the measurements. We'll come back to that in a few more slides So here's the data entry screen as I said you just type that information in if we were doing an in-person Workshop I would stop at this point and have people you know enter in some test results Here at the bottom you see it says Steph or is the letters are SA you The suffocating result is 20 followed by the letter s. That's a sensitive result here on the right It has you know the breakpoint so with the breakpoints who net knows is it resistant intermediate or susceptible? At the end of this you click on save isolate Later we will talk about view database We'll talk about backtrack summary a little bit print clinical report exit caliper or clear just to erase everything on the screen But right now we have finished manually entering our data and we click on save isolate We'll talk about date formats later Do you want month a year day month year? We'll talk about that when we get to data entry Okay, so data entry in who net you can either enter the measurements or the interpretations But we strongly recommend that you put the measurements in for a few reasons For me really the most important reason is to get the correct result There is an alternate quote-unquote method that is extremely common called the eyeball method I ball method is terrible. You pick up the plate. You pick up the diffusion plate You look at the zones of inhibition and you say to yourself. Oh, that's a small zone diameter I think that's resistant and So much as I disagree with you. I've been here longer than you. I think it's sensitive They did not take out a ruler. They did not take out the tables Nobody really knows and this unfortunate eyeball method is very common Of course, if a bacteria is highly resistant or very sensitive, they should get the correct result Basically anything below 10 is resistant anything above 25 is sensitive But in between 14 and 25, it could be resistant. It could be intermediate. It could be sensitive Spends of the organism depends on the antibiotic. It depends on the year because these breakpoints can change So please please please measure So there are sick people at the end at the other end of this plate We took this sample out of a patient. We put into a plate if The the patient suffered from their disease Taking of a sample transporting the sample It is the laboratory's job to do the best job possible to give the doctor a reliable result And if you simply use your eye to give the doctor a reliable result, it won't be reliable and certainly Sometimes you make the mistake that the bacteria is really sensitive, but the lab says resistant That's unfortunate because it's an antibiotic that the doctor could have used But they didn't because the lab incorrectly said it was resistant That is what we call a major error There's something even worse called a very major error. A very major error would happen if the bacteria is really resistant But the lab said oh, it's sensitive This is the this is a bigger mistake than the first example I told you because the lab might say that there's ampicillin sensitive And the doctor decides to give the patient ampicillin And then the patient may die because it was really ampicillin resistant So if you're simply using your eye to estimate r i and s Um, it's you're not going to get the right result. There would be many cases, especially with low level moderate resistance A lot of resistance a lot of important resistance is moderate level ESBL resistance quintalone resistance Carbopenam resistance is often not high level resistance So please measure to give the doctor correct information for patient care That's reason number one Reason number two. I like the breakpoints. These breakpoints can change over time Uh, if there is a change of the breakpoints, you cannot compare your old data and your new data Maybe in 2015 the correct interpretation was resistant Or more commonly maybe in that year the correct interpretation was sensitive But sometimes breakpoints change, especially to become more resistant A good example of this was around 10 years ago with carbapenems There was very little carbapenem resistance in the world very little imipenem resistance and um And I worked with one of the states in the united states and they thought That they only had three examples of imipenem resistant e coli I will review that c re carbapenem resistant Intervector eca in the whole state. They thought they only had three But then clsi realized that the breakpoints were in the wrong place When I applied the new breakpoints to their old data because I did have the interpret I did have the measurements I found that they had about 25 So the problem was much more common than they realized Because they were using the older breakpoints and the newer breakpoints were better at finding The the the resistance So if I because I had the measurements I was able to compare the old data and the new data But if they only gave me r's i's and s's I I would just have to say yes, the breakpoint has changed But I don't know if it impacted my statistics or not Some people wonder why do breakpoints change? um, and also as along the same lines if I am analyzing data today in 2020 If I am analyzing data from 2010 I should not be using the 2010 breakpoints. I should be using today's breakpoints People say but we did the test in 2010. Why are we not using the 2010 breakpoints? The reason for that is because the test hasn't changed the test we did in 2010 is the same test we did today What has changed is our understanding of what is resistant and what is susceptible So for the test we did in 2010 is the same test we're doing today. So please use today's breakpoints for that um, so for example in years that if they are reporting data in 2020 They recommend either using the 2020 breakpoints or practically maybe the 2019 breakpoints because not everybody has the 2020 breakpoints yet But they want to use the point is they want to use recent breakpoints Either this year's most up-to-date breakpoints. That is the clsi recommendation or last year's breakpoints, which are pretty close um Let's see I'll also detour a bit. This is important more for data analysis than data entry. In fact, it's not important for data entry But since I'm on the subject of measurements Some people wonder well, if it was quote unquote sensitive in 2010, why would it be resistant today? The bacteria hasn't changed This comes to an underlying misunderstanding of what breakpoints are about The purpose of breakpoints ironically is not to find resistance Uh, many of you may think that's a strange statement to make But the truth is that the purpose of these breakpoints is not to find Resistance genes is not to find a little bit of resistance The purpose of these breakpoints is to predict the clinical outcome In other words, if the patient has this strain The strain might be a little bit resistant And there are many examples where there is low level resistance where the patient would still get better So that's why we're not looking for resistance. We're trying to predict the clinical outcome So in the example I mentioned with imipatum resistance, the word bacteria that were a little bit resistant And clsi said, you know, they are a little bit resistant But based on our pharmacokinetic pharmacodynamic data, we think the patient would still get better We think there's enough concentration to overwhelm the resistance And then in 2010 they realized they were wrong and they said, oh, you know, this little bit of resistance is important So it's not as if from the so basically even though we thought it was sensitive in 2010 What we really meant is we thought the patient was going to get better Whereas today we realized there's more clinical data. The patient will not get better So the best way to understand this is that we're not trying to find resistance genes We're trying to predict if the patient would get better or not And that's why these breakpoints change and that's why sometimes the breakpoints become more resistant Things that we thought were not important are important About two years ago the floor or maybe even last year the floor quintalone resistance breakpoints changed a lot Because we used to think that low level resistance was not important and now they realized it is important Okay, I'll continue So we like the measurements because breakpoints change over time Reason number three is the measurements really help you a lot to look at data quality We will talk about that more when I get to data analysis And by reviewing and then in a fourth reason here you can if you have a high level resistance or low level resistance The doctors doesn't care if because you're not going to give that drug to a patient You tell the doctor your patient has high level cipro resistance or low level of cipro resistance Either way the doctor is not going to use cipro for the patient decision But it is valuable for epidemiology Knowing about resistance mechanisms. It's valuable to infect control If you have two patients in the same room And one patient is high level resistance and the other patient is low level resistance And if I trust my laboratory, it's just two different strains. There's not a transmission They just have two different bacteria. One of them is high level resistant. One of them is low level resistance But if the two patients have the same organism The both of them have high level resistance and both of them have low level resistance I'm very worried about a transmission That maybe the patient in bed a Infected the patient from bed b Or a doctor or a nurse or the door handle infected both of the people So these measurements and high level resistance and low level resistance give us a better understanding of the underlying molecular epidemiology The best way to know if two patients have the same thing is to do molecular typing But molecular typing like sequencing molecular typing is expensive. It takes time It takes money expertise equipment, but we don't have that routinely Maybe in 10 years, but no, we don't have that today, but we do have for the zone diameter measurements So these are four reasons why I like the measurements One it's the correct way to give the doctor the correct results Two the breakpoints might change over time. So the measurements allow you to compare your old data and new data Three the measurements allow you to look at data quality much better And four it allows you to separate high level resistance from low level resistance as a way to do phenotypic tracking Now there's a fifth reason because of this concept of one health Now the people are looking at food and animal and human and environmental samples Um, there are different breakpoints if it's a cat or a sheep or a dog And in that case you need the measurement if it's a cat you give this person interpretation If it's a dog you give this interpretation and you need the measurement to do that If you want if you take a nice look from a cat But you want to compare it with the human results you want to use the human breakpoints So with the measurement you can apply all of the different relevant breakpoints depending on what your need is I'm going to hit enter I'll stop right there for a moment. Are there any questions on anything I've heard so far I'll just count to five myself internally Okay, I'll assume that's no questions I'm going to go back a little bit So all of you already saw this screen in the upper right hand corner. It says save isolate as soon as I click on save isolate You will get this you'll get this screen Do you want to save the isolate and start a new one? Do you want to have a question? Yes here go ahead Uh, I don't know whether uh, you see it our annual report Uh, I think the whole report is based on uh qualitative result It is only said the resistance Intermediate or susceptible Uh, do you think the test measure can improve our report? For the coming year because we have to also go to in two details Some epidemiologist's comment as you have to go to the patient level All these kind of things so really can you again explain if the test measurement is useful? uh to I mean to write our annual report our in marsary blance report Currently it's qualitative. Yeah Also, was there a second question? That's quite a no no this is the first question and I mean the second question is About data entry one of our major problem is the communication with the doctors Yes, do you think this manual entry can Facilitate the communication between physicians and the lab. Is there any mechanisms that we can? Uh facilitate without using other uh data source like uh politic or whatever we have Sure, let me answer the first question first about the annual report. Um, yeah, I'm mostly working now with vietnam Sri Lanka, uh Laos and a few other countries to develop standard reports for analysis And who net already has a feature called standard reports It's an old feature that we will be expanding. So now with the Fleming fund We are going to make a lot of modifications to expand who net standard reports We're distinguishing between a few kinds of reports And uh, so I'll tell you a few kinds of reports One kind of report is a data submission report If somebody gives me their data for january I want to give that lab Feedback about their data set. I want to say thank you I want to say you You you have put in The patient's gender 95 percent of the time excellent. You have patient of the patient's age in 20 percent of the time. So it's basically a feedback report. I have to sneeze. Sorry, uh, Hold on Usually three times Okay, I'm finished um So, um Okay, so, um, so one kind of standard report is a feedback report Thank you for your data. You gave me 100 isolates. It was 20 percent complete 80 percent complete. I found these strange results We did not test any pendulum. I would have suggested you do that So one kind of important report is is a report That the laboratory can generate on their own But also that you at the national level can provide to them Thank you for the january data This looks great. This looks unusual. This looks wrong. Please make the following improvements This kind of feed report is not useful for the general stakeholder It's useful for the lab to make to try to help them improve through testing on a month-to-month basis To data quality test quality report data volume report. That's one kind of report Another kind of report again hospital specific is trending over time You know, if they give me their june data, I will give them a detailed feedback report on the june data with june data alone But I also want to give them a second set of reports comparing the june data To the january to may data You know in june did the total number of isolates drop? Did the percent resistance go up? Did the data completeness get better or get worse? So these are two kinds of facility specific reports And for the facility specific reports You can also give them a score on how well they did data completeness Did they do measurements? It's a quality measure So I would for that purpose in the feedback give them feedback on that And then the time-train report to see if things are getting better or worse Either epidemiologically or data quality wise live quality wise Other kinds of reports are meant more for the stakeholders The pharmacists the epidemiologists the needs the news media general advocacy general awareness At that high level the zone diameters are not so important So many countries like philippines spend spend a lot of time and effort on the measurements But the measurements do not end up in the annual report Or they might end up in the annual report on a section dedicated On either one of two areas One section in the annual report might be data quality and completeness measures And you can have that you know that 80% of these were measurements and 20% of these were not measurements So they use it as a quality metric. That's one way the Measurements might end up in the annual report And the other thing might be something very specific like imipenem resistance Because imipenem resistance there are different mechanisms. There's carbapenemase production There's pour and loss and this will often be reflected in the measurement high level resistance moderate resistance Decrease susceptibility So for the general average stakeholder for most things there often will be no measurement no mention of the measurements It might be in there at the highest level for a quality metric It also might be there for a very specific molecular issue like low to medium to high level resistance But generally speaking we don't use the measurements at the highest level We do use the measurements internally within the network to improve the quality and interpretation I'll give you one example of this from one country in latin america I'll just simplify the story, but basically we found that some labs had a lot more aminoglycoside resistance than other labs When we looked at the zone diameter distribution, what we noticed is that The sensitive bacteria In some hospitals was shifted to the left for the zone diameters In other words, what i'm saying is around the world the sense the resistant bacteria around the world are different But the sensitive bacteria around the world are relatively similar So some of the labs had the sensitive bacteria in the expected range like 20 to 25 millimeters But some labs had the same sensitive bacteria Like 17 to 22 the whole breakpoint range had shifted was shifted down And they ended up with a lot more resistance Because a lot of their sensitive bacteria were incorrectly in the intermediate zone When we investigated even further we found that these labs were getting their medium from the same manufacturer And so basically we were able to use the zone diameters to find bad quality medium To have good results, especially for something like aminoglycosides If you're familiar with the word amino amino and h2 and h3 Is high positively charged so Some things like that are positively charged are very impacted by the medium of the ph the ph of the medium And the calcium magnesium concentration of the medium And so we went to this manufacturer and said the ph or something is wrong Your labs are giving smaller than expected not measurements for the aminoglycosides So this is simply an example where we use the measurements to find a specific issue in the quality of either the discs Or the quality of the medium so that's using the measurements for quality objectives Or as I said earlier if you're looking for outbreaks or if you're looking for high level resistance low level resistance So these are very valuable scientifically But often you do not mention these you might put these into a publication But usually they're not in my usual annual report for most stakeholders Is that a reasonable answer for the first question about the reports? Yeah, so you advise to use qualitative data That is a resistance intermediate or susceptible For the different reports for the typical annual report that we plan to share outside of the network Yeah, for them our i and s is typically sufficient Okay, thank you. I do use the measurements for other things Valuable things, but I just don't put them in the annual report for everybody. I put those in specialized reports Okay, second question about communications. I'm not sure exactly what you mean, but um, let's see The Hoonat is used by many laboratories for clinical reporting So you take the results from the paper notebook you manually type them into Hoonat You print out the clinical report and then you give that document back to the back to the doctor um So let's see. Um, so in that way it is improving communication. They have a nice standard printout um Hoonat Hoonat reports can be saved and uh something like a pdf file So it does raise the possibility you could save the report and email it to somebody So that could be useful You can do the clinical report with Hoonat, print it out onto paper or save it to a file and then email it to somebody So it's one way to improve communications Another way, especially if you have the data on a network drive Is you might have the people in the lab do the manual data entry But you might have a secretary up front Who is doing the clinical reporting? So I was at a laboratory in Bhutan and they were doing Before we set up the network drive They have a system. Um I've forgotten about this. They have polytech I forgot about this. They have polytech And in this similar situation to you they're using polytech for all of the labs Except for archaeology They just said it was inadequate for a lot of their needs So what was happening is for all chemistry and blood bank and other results patients would go to the hospital lobby And the person in the hospital lobby could look up all of the results Except for microbiology because they're not in the system So for microbiology, the patients had to walk up to the laboratory That's a pain in the neck for the laboratory because they're constantly being interrupted But patients asking for their test results So when I was there, I helped to set them up the polytech not the polytech Hoonat on the network drive So the lab would enter the results manually into Hoonat But the person downstairs in the hospital lobby could look up the results So this was a big change for them because it meant that all patients could get their polytech and Hoonat results in the lobby They did not have to walk up to the laboratory Uh, so this is a second example Eventually we do plan to have a web based version of Hoonat Uh, both an intranet version and an internet version. Well, of course the same version But you can put you can put the web version on Hoonat When it when it will exist on the hospital intranet Meaning that anybody inside the hospital firewall would be able to look at the results But if they have the username and password So this would be useful because right now doctors come down to the lab or they call the lab They come down to the lab. I've seen this many times in fiji They were using they are using a system called star limbs, but not for microbiology And so for all of the labs they could go on the floor and look at the results In fact, in fiji they have this wonderful system that around the country no matter what island they are on There's a web based system for looking at results Except for microbiology because the microbiology is done manually in paper paper systems again They just didn't like it for microbiology So I was able to help them resolve these issues to get Hoonat and integrate it with the web interface So, um, this is ways of improving communication so that the clinicians and I understand that that the Block lion hospital is they do have a system that any physician in the hospital Can look at all of the chemistry results and blood bank results, but not microbiology So regarding Block lion, I do recognize that it might be more work for the labs And start entering the results into polytech or Hoonat But it's not only about the lab. It's not only about their workload You have to consider the workload of the clinicians if clinicians are walking down to the lab bringing their infected Hands with them not washing their hands looking at the notebooks And then they go back to see their patients without washing their hands. There's a lot of risk is contagious in there Um, so if you can replace that and or the other options they call they call the lab Calling the lab is just a big interruption to laboratory work. Um So at the block line, there's a very powerful incentive. I would hope to get people to digitize the microbiology data Uh, so that physicians could simply look at the microbiology results Just like to do for the other laboratories So these are some different ways that you can use Hoonat Once we have a web version of Hoonat It will be even more valuable because on the intranet then people can look it up The block line doesn't need that because they already have an intranet system of that nature But if they use polytech and they interface it then physicians could do it that way So I hope that those are some some answers to your questions on communications Yes, uh Great. Yes, John, uh, I can understand uh save any mail to a physician. This is uh, uh, Very understandable and also the web based is also we're trying to have that web based but still as I said before We don't have polytech all across the country But the second option the second, uh, I mean Solution you suggested network drive. That's not clear for me Oh Save and email to a physician And the web based is also very clear because if it is interfaced everybody can access But the the last one is not clear the network drive how they can share Uh, this information or how communication is facilitated through this. Can you again repeat this one? Sure, sure. Um, and also I forgot to mention the new polytech web version Right now I talked to jeff fisher, you know the company owner president a few weeks ago And uh one focus of the web work is covet so that people around the country can enter their covet results But it's still the web platform Um and ideally with the web platform You can look at the other results and chemistries and blood banks and microbiology results So this can be a wonderful opportunity With the web version for doctors logging in to see all of the results no matter where they are in the country And it would be a shame just like the fiji situation It'd be a shame if they could log in and see all of the patient results Except for microbiology This would not be a problem for epi results because your poly your microbiology data are in polytech So it should be possible for doctors to look up their microbiology results But for black lion the doctors would be able to look up all of the Chemistry hematology blood banks serology, but not microbiology because it's not in the system So especially once the polytech web version is further implemented and further integrated That'd be an additional powerful reason to try to get the microbiology in that would greatly improve Physicians access to data Regarding your point about the network drive. Yeah, sure. Um, you know So basically if I want to give access To the microbiology results to 100 people The web solution would be the best because you have Basically you have the server that everybody is looking up You just need all of the computers around the hospital to have an internet connection So they can look up the results. So if I have 100 200 500 people the internet version would be great Is anybody with an internet web connection should be able to use that system For a desktop application like hoonet That's not a very convenient solution What they would have to do if you wanted 500 people to have access to it for a web version That's easy. You just give them internet web access Uh, you only have to maintain the server and the web connection But if you wanted to give 500 people access to hoonet as a desktop application You would need to go to all 500 computers and install hoonet separately So that's why if you want to give hoonet access to all the physicians The web version would be much better, you know, once we once we eventually get back to it Because I don't want to install hoonet on 500 computers and then maintain it just too much work to do that But if you want to install hoonet on five computers or 10 computers or 20 computers, that's much more manageable okay, so basically what if I have uh Let's just say that I have maybe three computers in microbiology two computers and the laboratory administration And maybe two computers in the infection control office And maybe two computers in the infectious disease department So there might be 10 computers where I want to give these 10 people access to hoonet So what I do is I go separately to each of the 10 computers. I install hoonet I have the hoonet look at the common network drive the t drive or the p drive So I can have I can put the data onto the network drive So if I am sitting in the lab at computer number one When I click on save isolate the screen that's in front of us when I click on save isolate It won't save it on the hard drive. It will save the data immediately to let me say the p drive Um, so it will save the data to the p drive But then somebody in infection control or in pharmacy They have access to the same hoonet. They have access to the same p drive So they can look at the results even though they're in a different part of the hospital So as long as they're on an intranet with access to the same network drive They can see everything that the labs people can see The disadvantage with desktop hoonet is if you have 10 computers somebody has to walk to the 10 computers walking to 10 computers It's not a big deal But walking to 500 computers is a big deal So so so for 500 people the webs version would be much better But for 10 people somebody goes to computer one and sets it up They go to computer five and set it up They walk over to the pharmacy and set it up They walk to the infection control office and they set it up So the data are residing on the network drive So one group enters the data Either a manual data entry or by using back length, but then anybody who has access to the network drive Um, you know would have access to the data Right now the data that we discussed would be on the c drive hoonet data folder But it doesn't have to be the c drive in the data folder It could be the p drive hoonet data folder So anybody who has given access to the p drive hoonet data folder can do everything That the lab people can do Does that make sense? Yeah, yeah, thank you I'm going to take a little detour and I just want to emphasize this point um Here's my 20 20 And the network drive it's all easy I don't want to make this sound more complicated than it is if I go to data entry new data file You see that at the top of the screen Hoonet wants to save it in the c drive hoonet data folder But I could save it on the w drive or the z drive or the t drive You see the red x here that these are my net hospital network drives. I'm not in the hospital right now I am not on the vpn. So that's why it says an x but if I said here w Well, it's telling me I'm not inside the hospital right now But instead of saving it on the c drive, I can simply save it on the network drive So in hoonet, it's not complicated It just means I need to go to I need to install hoonet at my work computer I need to install hoonet on my home computer I need to install hoonet on maybe five or ten computers and then each of those computers. I say the same w drive Okay So that's how we can choose at the time of data entry hoonet also has this nice feature under file settings So hoonet by default The data will be in the hoonet data folder, but it is configurable I don't have to put them into c drive. I can put them in the p drive or doesn't have to be called hoonet data it can be called um It doesn't have to be called hoonet data. It can be called hoonet data 2020 Hoonet data 2019 So you can choose exactly what drive and what folder you want it to be saved in So basically I explain that to you in words with my power point But I also wanted to show you that inside of hoonet. It's extremely simple Instead of saying the c drive you simply just choose some other location You know, maybe I want to put it on the desktop and then I put it here under stuff and I put it wherever Um, so is that there? So if you had in the lab five computers Just simply install hoonet on the five computers and then save the data on some drive that they all have common access to And the other questions. Otherwise, I will go back to the power point Maybe uh, I have one simple question Yes Yeah, while you are saving the data Sometimes, uh, you know, we are looking error code something Error code 72 and the like so what is the The reason behind Looking this kind of era I'm sorry, you're in 72. You said there's a number Yeah Error code 72 something like that for this when we are saving No, no, no, no just generally while you are saving or opening the hoonet Uh software you will get sometimes error code 72 Yeah, yeah error means an error Um, yeah, so that means we need I need to work with you to figure out the cause of the error I'm just going to do a search error 72 It's probably not 72 error 72 Um Looks like something with adobe. So so basically hoonet. Let's see. I'll come up with an example Um, I'll show you an example. Um, let's see. I got an email today. In fact, I was entering this email when you called This is from greece. Um here so here this kind of thing error number Okay, you can still see my screen. I hope it says error number 99 Yes, we can see it Great great great. Um, and you see then the number the number person doesn't mean a lot to me The next thing down is the description here. It says external component has thrown an extent an exception That tells me there's some compatibility issue We're trying to do something that's not compatible on their computer And then we need to figure out what it is and this is something we can do to avoid this compatibility issue It makes that bigger. It's easier to read It tells me the error number which I mostly ignore And it tells the error description associated with that number. So these two things are linked together The external component has thrown an exception. It tells me that this happened Uh during import routines Specifically in a subroutine called import read file. You give some other information So if you are getting these error messages, also if I go to I don't know if I have any here Would you see that on the screen? Yeah, here are error logs. So these error logs tell in fact I can delete these so so you see here this error log Says 2020 0 5 11 So that this error was generated on the year 2020 may 11th at 11 38 in the morning So when I open this file, it tells me what it saw on the screen So if you want to tell me something, um, sometimes people tell me john, there's an error in who net That is A little helpful, but it's not very helpful. I need more details What was the error number? What was the error message? Um These additional details that we can generate what were you doing where you're in data analysis where you're in data entry So these so who net tells me the information on the screen so you can send us a screen capture Or you can send us the error file. Don't send us all the error files. Just send us the most The one like the most like here at the same error probably three times in a row I had probably 11 38 in the morning 11 40 and then again at 11 41 if I open that up Uh, you know, it's you know the error message in this case Well, so this gives me the details of the message So in answer I can answer your question right now in a specific sense What I can tell you is if you get an error message, please tell me that error message And then we can explain why you are getting that error message Sometimes it has to do the network drive is not accessible. The network drive is down. Of course, there's going to be an error Um, it should not happen routinely that the two basic reasons for an error or either that there is a mistake inside of who net Or there is some local problem If it's a mistake inside of who net we do want to fix that Uh, well, there's sort of three kinds of issues. Um, one is there is a problem an intrinsic problem in who net Another is there's something very specific that laboratory people did wrong Or the network is down or the some of there are or something other unexpected thing There's something in the middle like these compatibility issues like that message I showed you for greece There's not something specifically wrong with who net but there is some compatibility issue on their computer It's not their fault. It's not exactly our fault But we have to we have to search for google to see if we can understand the compatibility issue Is there's some way to get around the compatibility issue? So if you send me those error log files Then we can start having a discussion with you So we can first of all have a diagnosis. Why are you getting that error? And secondly, uh, you know, secondly a discussion You know, secondly, uh, you know, first of all a diagnosis and secondly, um, a solution Okay, thank you. Maybe another question You know, we are trying to use a web-based version across the side But, you know, web-based requires Very strong internet access. So can we use This web-based version with a limited internet access Well, I want to emphasize who net does not have a web version So the web version I mentioned is a polytech web version. I'm going to go to the who net website Here we have who net web. It's only a demo. This version of who net is still in development Unfortunately, we made a lot of progress until about six years ago And then we started to have more and more compatibility issues with our old who net 5.6 So what you see here as a who net web version is work that we did around six years ago It's been on the back burner for a very long time And at that time I did have two programmers Adam was working on the web version and I had a different programmer working on the desktop version That grant ended so we had to focus on the desktop version So we have I am very pleased to say that about two weeks ago. We have finalized our modernization upgrade We modernize different things at different points. We finished most of the modernization Per who net about three years ago. We finished it for back think last year There were some small details, but important details for compatibility that we fixed in the last two months Um, it has to do with the technology called dao that finally we were able to get rid of So now what I'm very happy to say is we have now caught up So now we are going to focus not on the web version because we have other things to do first Our first priority are the standard reports. I described to you standard reports Automation scores. So what I'm basically saying is that maybe in about three months We can start to make the web version of priority for development again We had a we had a programmer last year for a year Uh, and we only had money for a year. So I told her we have money for you for one year And she helped us to finally modernize backlink If we had money, I'd love to bring her back again before she she's still available She has a job, but she would love to come back to us So so I would love to have two programmers, but with only one programmer We always constantly have to juggle So I'm hoping in july or not july this too early But in about three months that we can finally make a serious attempt to go back to the web version So in short, there is not a web version of who net at the present time There's just this demo that we did that we can highlight sort of the direction what we hope to accomplish So at the present time you do have a polytech web version that is in the process of installation I highly recommend people look into that web version try to use it as much as possible At ephi you can use it for microbiology data For tiger lion they can use it for everything that's in the system Everything that's in the system does not include Microbiology, so I think the web version of polytech will be a new opportunity To emphasize the importance of putting the microbiology result to the system Polytech does have problems, but there would be such enormous benefit to utilizing it recognizing its deficiencies Don't try to use it for internal Perliminary results and comments, but for the final results the web version would be an additional incentive To try to get the results to clinicians all over the country who have username and passwords Okay, you mentioned about quality of internet access Um, of course if the web system is down the web system is down and nobody can do anything if the web system is slow That's in the question is how slow you have to wait one minute or five minutes Uh, sometimes if you wait too long, it'll time out and it will give up One way to handle this is to strengthen a lot of systems have a time out of one minute If it doesn't work in one minute it cancels what you may want to do is change the time out to three minutes Just to give the system a bit more time to catch up Okay, what we plan to do for who net for data entry with our web version And a lot of systems have this is they have data entry of two types There's online data entry or offline data entry When internet access is poor offline data entry makes a good sense as a good option Basically you go to the web page If the web is down and you click on save the data gets saved locally And then the data gets saved locally And then when the re internet submission is reestablished it'll resynchronize itself So even if the internet is completely down That should not hopefully in a good system with local capability of storage If the web if the web is completely down You would hope that you could do manual data entry it gets saved locally in the short term And once the internet connection is reestablished then it gets transmitted over So this is one strategy for dealing with unreliable internet connectivity is to have offline data storage I also want to distinguish between there are very different needs of a web system for data entry and data retrieval Usually data entry is only like two or three or four people So those two or three or four people need To enter the data into the web system Ideally with this idea of a local Store of data that can be synchronized So for them if the internet is down they can continue to enter their data which will be synchronized later So one aspect of a web system is to facilitate data entry For three or four people who are responsible for data entry But then in terms of data access the physicians are not entering data They simply want to look the results up So for them there is no there's no problem about internet being down for data storage Because those people are not doing data storage The problem for them is they cannot do their data retrieval if the web is down So you just want to distinguish these different needs For data entry with bad connectivity make sure you can do offline data entry And then resynchronize when the internet is established internet is reconnected And that's relevant for the three or four people doing data entry for data retrieval Then those people are not entering data. They simply need to retrieve the results And they just have to wait until the internet is back Or just give them a longer delay before the time out happens You know, sometimes it's frustrating you'll get like half of the page loaded and then it times out You may want to change that to time out of three minutes or five minutes Just to give it more time to load the page I hope those comments help Yeah, thank you So are things that I personally have no expertise in Sometimes you'll have multiple mirrors. You'll have five. You'll have like three copies three mirror copies So if the internet connection with server number one is bad, it can automatically switch to server two You know internet works generally that way not not because the internet is down But just to make it faster You know, there there are copies of the internet mirror copies of many things all over So we'll try computer number one if computer one is busy. They'll just move it over to computer three But this is obviously more advanced Okay, other questions. If no other questions, I'll just go back to data entry And we've finished about one hour a little bit over one hour. Just as a time check back to my power point Let's move this and put this back in full screen move this Trying to move the go-to meeting a little window with that accidentally disconnecting myself So data entry after you click on after who in it just asked you do you want to save the isolate and continue with the new a new specimen Or save guys and then continue with the same specimen For example, if you have a blood with two or three bacteria, it's the same patient the same location the same sample But the microbiology is different. So in this case, we're not just an ask you to repeat everything They just ask you to put in the new microbiology results or the last option save the isolate and continue with the same alert the same patient That meaning that the patient might have a blood and a urine and a wound. So it's the same patient Probably the same location But a different sample and different microbiology results The isolate that I just entered has no alerts So on this screen, it's a nice simple screen and there are no alerts If there are alerts, the alerts would appear at the bottom of the screen and we'll see that in a few more slides I will click on continue So here is again, if we were in person during a practical demonstration I would ask you to manually enter these results. It's the same patient the same blood But it's not as deaf or is it's an E. Coli and then Right first I'll start on screen and here you see the alerts This is E. Coli and it is imipenem resistant It says that on the screen enter bacteria Eaceae non susceptible to carpet penems. That is an example of a high priority alert In the lower left hand corner, you see there's a low priority alerts medium priority alerts and high priority alerts I will now return to the previous screen data entry Hoonet has about 190 alerts that warn you of possible typing or laboratory testing errors that should be confirmed For example In this example, it is resistant to septra axon, which is a medium priority alert possible ESBL producer There's also imipenem resistant, which is a high priority alert So I will go back to the next screen so Um, these is very valuable for finding mistakes Maybe the person was trying to type You know 22 but they accidentally type 12. So it might be a typing mistake Or it might be that they put the they put the ampicillin result into the imipenem column They just put the antibiotic in the wrong antibiotic. So it's helpful to find typing mistakes It's also helpful to find laboratory test mistakes Like clepsiola pneumonia is almost always ampicillin resistant. It's a characteristic of the species Clepsiola pneumonia can be ampicillin sensitive. It is rare Like maybe 1 2 3 percent of the time. So clepsiola pneumonia can be ampicillin sensitive But if you do see clepsiola ampicillin sensitive, I would not tell the doctor Unless if I confirmed it is it really a clepsiola pneumonia and is it really ampicillin resistant? So when hune gives these quality control alerts, it does not mean there is any mistake But it means there might be a mistake and the laboratory should double check it So hune has high priority medium priority and low priority alerts So do well, let me take that back. So, um, hune does high priority alerts High priority. There's an hune does high priority species Or important species important resistance The three main kinds of alerts of hune are important species important resistance and quality control An example of an important species would be high priority vibrocolaris salmonella typhi High or medium priority salmonella enteritis MRSA, so these are important Species or important resistance CRE is a high priority resistance VRSA is a high priority resistance MRSA, ESBL, VRE are medium priority resistance And then you have the quality control alerts like clepsiola sensitive to ampicillin You know sorodia sensitive to ampicillin or if you have an E. Coli that is ampicillin sensitive Amoxicillin, clavulin, well, let me give a different example If you have a bacteria that is ampicillin sensitive But ampicillin cell bactam resistant That doesn't make any sense If it is ampicillin sensitive, it should also be If it is sensitive to ampicillin, it should also be sensitive to the combination of ampicillin to cell bactam So bactam is there to help not to hurt it So if you have an E. Coli sensitive to ampicillin but resistant to ampicillin plus cell bactam That's a quality control result. It doesn't make sense If you have a bacteria sensitive to ampicillin but resistant to imipenem, that doesn't make sense But these things that don't make sense do happen And the most common reason is some mistake For example, imipenem disc is not a sensitive disc in a top tropical environment. It tends to degrade. It tends to dissolve So you might have a piece of you might have an antibiotic disc that says imipenem on it But there might not be any imipenem in it if the imipenem is all deactivated Or instead of having 10 micrograms of imipenem, maybe there's only eight or seven or three Because it's been deactivated or if it's a poor quality manufacturer disc So you do see these strains that are ampicillin sensitive imipenem resistant, but it is not true There's some mistake. Somebody measured wrong. The disc is a bad quality Some other mistake So the purpose of this alerts is either to find things that are of great public health importance Or infection control importance or data quality importance So Vitek and Phoenix and Microscan do have similar alerts, but Hoonet is better in two ways Vitek and Phoenix, they're alerts are more about data quality If you have Plexiola sensitive to ampicillin, the Vitek will give an alert But if you have a CRE The Vitek will not give an alert as long as it's correct. It's correct So the Vitek's purpose is to find quality mistakes The purpose of Vitek is not to give you public health importance alerts So these things are important, but that's not the Vitek's job The Vitek's job is to tell you that the result is correct So one advantage of the Hoonet alerts is that they do have to do with public health and important clinical findings Not quality findings Hoonet has it has all of those as one advantage of Hoonet's approach The second approach of Hoonet is that it's also part of data analysis When you do the Vitek alerts, you can see those alerts one at a time But you cannot do an analysis. You cannot summarize how many you cannot do a query on alerts You can do a query on patient ID and look at the results in one at a time So an advantage of the Hoonet alerts is that it's part of data entry, but it is also part of data analysis So we're going to go back to the preceding screen. I already did that And then when you save it so again in the upper right hand corner if I click on save isolate It will summarize the alerts important resistance You see those alert section. There are three the first three of the ones I want to focus on Quality control alert important species alert or important resistance alert The other ones are more subjective, you know, but the first three are you know Other more important ones and then there's alerts are summarized at the bottom of the screen So a few minutes ago I showed you this screen, which is exactly the same screen For a boring bacteria with no alerts This is now exactly the same screen But the bacteria is more interesting and it gives you these alerts helpful to find mistakes or important findings We'll now continue And I don't want to spend too much time on this in a manual in an interactive workshop You can ask the people please enter the results on the screen here um Okay, I'll skip over that example. I'll show that during the live demo Here's an example that who neck can also handle quality control the patient's number and name is a very special name ATCC 25922 ATCC means American type culture collection These are qc strains that come off that you purchase that get shipped in that are part of quality control of normal test practices I visited one laboratory. Oh, I was just so embarrassed for them They did not realize that they really did not know what they were doing I looked at their clinical data and 15 percent of their clinical data Were staff aureus vancomycin resistant So they had 15 percent vancomycin resistant staff aureus I said that's not possible. It's impossible vancomycin resistant staff aureus almost does not exist in the world um In the united states in the last 20 years Only 16 people have had this confirmed. So v rsa is almost Non-existent in the world. So it should be 99.99 percent sensitive The fact that 15 percent of theirs were resistant said this something's not right. They're making a mistake here and I I was very glad to see that they did put all of the qc results into who net and 20 percent of the qc strain was vancomycin resistant and they told me john. Yes our qc strain has become very resistant And I think to myself self you people have no idea what a qc strain is The qc strain does not become resistant you purchased it It wasn't resistant when you took it off of the when you picked it up in the man room. It's still not resistant So who net can help you with the qc? So who net knows the correct measurements? So for example for the qc strain they have 16 20 to 18 and who net will tell you Somewhere in here you see on the right hand side of the screen for emmy pennum The result is out of qc range So I do encourage you to ask labs to enter the qc results That serves two purposes One is it tells you that they are doing qc Uh, because sometimes they tell you did they do qc, but if you ask them to enter it you can see how often they do it Uh, and let's just assume they're not lying. Of course, it is easy to cheat on qc You can say oh, I know what the result is supposed to be. Let me just enter the qc So I don't want people to lie because a lot of times they feel it as a critique that they did something wrong There is if the result is not correct If they feel strongly about the quality of the care they get to doctors If the qc gives bad results, there is a problem A lot of times the problem is not their fault If if they purchase bad quality media and bad quality discs these discs are stored poorly That's a problem. It's not a problem you want to hide. It's a problem you want to resolve Um, so some people do qc once every three months Uh, once a month, uh, the celicide recommendation is a bit excessive Uh, but a lot of people try to do qc of three bacteria Stephoris, E. coli, and pseudomonas at least once a week Or at least once every two weeks And it depends on what you agree is a network. What would be a good strategy for doing the qc testing? And then this allows you to see first of all, are they doing the testing? Are they entering the results? That's goal number one and then you can give them a score How often did you do qc testing? That's one score And then the other score is how good were the qc results or the in range or the out of range? And if they're getting bad results, don't blame the lab It's it's basically there is some problem that we need to investigate and fix And then that's where you start to see problems. Maybe their inoculum is wrong Maybe the medium is too thick or maybe the medium is too thin. If the medium is too thick Then the antibiotic diffuses down instead of horizontally So you end up with smaller zone diameters if the medium is too thin Then the antibiotic diffuses a lot more and the zone diameters are bigger So the whole the pro purpose of qc is to help the lab do good quality testing So who not allows you to put the qc in the whoona does can tell you if it's in range or out of range Okay Here in the upper right hand corner and the second option there is view database So I clicked on view database and you can see I see all the data those four isolates that we just entered I can edit an isolate to look at the one at the time Edit the table if I just want to edit this table directly manually um Or if the doctor is calling I gotta go from dr. Smith and dr. Smith is telling me I'm in the lab The doctor saying is I want to know the results for my patient Here's the patient's name data birth or the patient number somebody in the lab can look at the screen and just look at the results on the screen There's some things we do want to make this easier like whoona doesn't have a filter We will make a filter on the screen which will make it easier to do that We'll probably also replace this with just sort of a manual look up You know, um so that you don't have to look at all the people at once So so I described a situation the doctor will call you can use this to give the patient results We're going to add on to this a few utilities to make that even easier by putting in some filters and allowing for individual patient look up You know, for example, it says fine right now when you do fine It looks at the table, but it would be nice if it could switch to a different kind of display The other option on the screen. Well, there's the lead if you entered it twice You can just delete the second one or you can print And the print is probably the next thing so you can sort the data edit the isolate edit the table I don't want to spend more time in that it'd be easier in the manual data entry delete search print, etc Okay, then clinical reports. Um, how much more Okay, some people use who not to report back to clinicians Some people enter into who net all results positive and negative Of course, that would be great for statistics. How many bloods did we do? You know, uh, you know, David you would like to know how many blood cultures did you do? I don't care if it's positive or negative. I just want to know how many you did So there's an advantage if you put the positives and negatives in because it gives you more information about the laboratory's workload On the downside, it's a lot more time to for data entry So labs just have to decide whether or not they want to enter everything Which will give them more analytical value Or just the positives And just the positives is often a very reasonable compromise It's just a question of how much time and effort does it take to put the negatives? A lot of times I recommend just start with the positives If they want to know how many blood cultures we did, I'll just look up our inventory and I'll give you an estimate And don't use who net I'll just tell you what we do approximately We do approximately 20 blood cultures a week Multiply by 52 and I'll give you an estimate for the year So there is value to having the negatives in who net, but of course the amount of time for data entry is much greater So each I encourage please enter at least the positives If they can do that, then you can have the second question Can you also do the negatives? And a lot of places I do have to say only put the positives in But you know for reporting if people use what a lot of people do if it's a negative report They just they don't put that into who net they will just stamp the request form They'll send the request form back no growth back to the clinician Other people do put the negatives into who net then they print out a report The report will send you growth no growth and they give that to the clinician So that decision about what you enter is up to you Okay for for your lab using polytech I'm trying to get people to try to use the polytech because there are all of those other benefits But for people without polytech then who net would be a good strategy for them for clinical reporting and storage And you might only want to do the positives. That's a good way to start and if it's not too Much work then ask them also to do the negatives So you can print out the clinical report and you have the I'll skip over that. I'll just show you what the report looks like Well, we didn't see it on this screen. Okay. It actually in this slide said it did not show you what the report looks like It's simple. It's basic. It's okay You can customize on this screen if you want to put the header the phone number The doctor's name if you want to put a signature row Who net will take care that middle section is called data fields So who net will put the microbiology results? But if you want to put a signature row or the address of the hospital or a heading And you can choose the font you can do that. So you can customize the look of the print of the clinical report The other thing that a lot of people do so people doing who net for clinic reporting Sometimes edit here the report header. They'll put the name of the hospital the address the phone number The footer they'll put the name of the doctor and the signature row And then the doctor some they'll print it out There will be a line there for the signature and then the doctor will sign the microbiology lab The microbiology head will sign the paper. So a lot of people do that What other people do is they have who net take care of the data fields But instead of printing out onto a blank white piece of paper They will print out onto hospital stationary I had some here on my desk Where did I put that? Oh, here it is. Can you see me? I don't know if you can see me, but you know, this is an example Of stationary for my hospital. It has the name of the hospital at the top It has some more things at the bottom So a lot of labs like to do that because they can get a more professional print out The top is beautiful with icons and logos and colors that they coordinate with the printers at the bottom There's the signature line. So a lot of people using who net for clinical reporting A lot of them just print out onto a white piece of paper But a lot of other people print out onto hospital stationary, which looks nicer And finishing up then you exit out of here Looked on data entry So sorry to interrupt John. Did you try to show a clinical report because I didn't see that on my screen Oh, okay. Well, I wish so there is none inside of my power point Oh, okay. Inside the power point. I do not have one the person who made the slide set Did not put one in and I have not I don't use these slides at myself. Usually I just usually use Live because I know it off the top of my head, but it is correct. It would be good to have an example of a clinical report Here is Okay, but I did try to show is can you see me personally can you see no Oh my camera my camera is on but I guess you cannot see me. That's okay Just imagine this I was holding up a piece of hospital stationary Okay But John we have a you know, very big problem with this hospital stationary especially in the Guramba South Spital So can the clinical report uses for multiple of patients or just one patient For example, the one-day report or monthly report or quarterly How is the practice? How is the practice? Well, okay, but you there are two different reports. You just mentioned there is a patient report You know the doctor, you know the doctor sends a sample of urine to the lab And it's an E. Coli ampicillin resistant. You print out that clinical report You can do one or two reports on a patient who net you do two reports on the page You can cut their page in half to save paper So that's a clinical report that you give back to the doctor and then he can take care of the patient That's a clinical report The other kind of report is an analytical report and the analytical reports You can do weekly analyses monthly analyses quarterly analyses or the annual report So I I've been talking in the last few minutes about the clinical report features Not about the analysis report features Okay I I do want to and I think I showed you this previously Um, let's see if I can find it quickly see drive It's on my other I haven't quickly completely switched out with I No, that's not here. I think I've shown you before the reports from Japan and vietnam So yeah, I like the idea of having these pre-formatted nice pdf files So in the next few months, that's one of our main priorities in who net is to make these nicely formatted analysis reports that you could so right now who net does a lot of great analyses But I usually like to clean it up and excel a word before I give it to people It would be nice if when I could do a report that you could immediately give to other people look And that's the purpose of the japan report and the australia report The it's interesting the vietnam report is very valuable, but it really is meant for The experts, you don't it's a feedback report for that lab I would not give that to everybody because there's too much information. It's not it's useful for data quality assessment It's not useful for discussion purposes for broad stakeholders So what we want are different reports with different stakeholders in mind the feedback reports the epidemiology reports And who net does all of these data analyses? But right now it is often not in the best Formatting to give to other people so in the next few months you will see significant improvements in these areas It is in fact what we plan to make our big priority starting this january So I was hoping that we would have all this done, but that's when we encountered a big compatibility issue with dao So we lost two months basically modernizing it's something we're planning on doing already But we're kind of forced to do our modernization effort first now that that is finished then The the next big priority are these reports so the next two months I think you'll see significant improvements in the report capabilities you know in some sites again for database They usually resist to inter negate culture negative results Do you think this is uh, not helpful for that analysis? What do you advise on this because there is always this kind of I'm just commenting on this I do I I tell them about the Modest benefit because they want to feel that their time is well spent They want to feel that they have to chew 10 different things. What is the best use of their times? And definitely it is easy for them to understand. I hope the benefit of entering the positives Yes, because the doctors want the epidemiology the statistics the surveillance all of those things you need the positives Regarding the negative I mentioned there are benefits to the negatives, but they are not enormous benefits Um, so I don't push on the negatives. I do tell them the advantages of the negatives But I don't make it I don't I want this to be sustainable. I want them to feel good about the project I don't want them to protest. I want to make it as I want them to feel that their time is well spent If they feel that entering the staff in E. Coli is a good use of time great. That is our first objective If they do not feel that entering the negatives is a good use of time There are two things I do is I do explain the benefit of the negatives because they say, oh, I hadn't thought about that Yes, I can see the value of the negatives But after I explain that I see I see what you mean, but I just thought it's too much time In the case of blood cultures typically 10 percent of the 10 percent of blood cultures Of course every hospital is different about 10 percent of blood cultures are positive About half of those are real positives about half of those are contaminations 90 percent are negative. So if you want to enter the negative you really want them to spend 90 percent of the time entering the negative results The main value of having the negatives Is to get a look at data volume the workflow. How much work is the lab doing? Are they doing 10 blood cultures a month if it's positive or negatives another matter If they're doing a hundred if they're doing a hundred blood cultures a month. I want to know that That's an important question Well, it's positive or negative is a different important question So if I want to know how many blood cultures in months they are doing One way to do that is to have them enter the negatives There is also another way and it's just an estimate You can just ask the lab about how many do you do a month or maybe they have precise numbers Because they have to do inventory. They purchase they purchase the 150 blood cultures They use the hundred of them they have 50 left. So if your interest is in volume of processing and purchasing Like we want people to do more blood cultures. I don't care if it's positive or negative. That's a different question If I see that now the lab is doing 10 blood cultures a month I want to see they're now doing 30 that want to do 50. I do want to see the growth in blood cultures One way to do that is to have them enter the negatives The other way for them is to simply give you an estimate or an exact number based on inventory You just ask them Please enter the positives into hunnet, but separately just tell me how many blood cultures you did last month That's often much that gives you most of what you want in a much easier way They will manually enter all the positives, but the negatives. They're just giving you a total number Well, thank you John. Thank you Geneva wants you to enter the negatives for the temperature of glass, but Geneva is not paying the labs Uh, so we want this to be sustainable in a long period of time. Maybe they'll say yes We're happy to enter the negatives for the next two months But we're busy. We have 20 different things we have cove to deal with and if you have to judge what is sustainable Of course, that's another advantage of an lis like if people enter manually the results Into a system like you know what many places do is they will enter all samples into a system like polytech And that way all the samples are in and then when they finalize it, they'll still finalize the negatives in polytech So people have an lis often the negatives will be there automatically So these are things you can use to balance. So I want to do everything we can to optimize your system for collecting the positives Collecting the negatives is sort of a longer term discussion John Yes, was that the start of a question? If not, I will go back to the power point Um, let's see. I need to move I go to meeting and put this back here So here you see on the data entry screen. There's new data file Open data file at the bottom of the screen You see the most recent file the one that we were just testing listed that just makes it easier for you to choose recent files Who not like Excel and word. I think they show you the most recent 20 files. Who know we show you the most recent four Finishing up. Okay data entry exit and that's the end of wow. That's that's it. We went through a full power point It's now 20 minutes before time. So I'm open to questions or a new subject John, I have a question please From the quality How can we enter a quality control data? As we know, we have a guideline based on say lisa Uh, the quality control perform New loads new shipment then after weekly so every week We will have to enter a quality control data with the patient result or is there any another mechanism to enter quality control data Sure, I will answer that now. I'm now inside of who net And I I chose the dvcho test hospital to demonstrate to answer your question I'm in data entry on the menu. I can choose new data file open data file or one of these files Had opened recently. So if I got a new data file Uh, let me just call this junk dot junk, you know, I do that a lot just because I so I know I can delete it later I'm going to put it on the c drive in the who net data folder Or as I said earlier, you could choose some other location for it. For example a network drive And here I'm inside of who net uh data entry Human isolate human animal food etc identification number. So, uh, you have two choices So you can put the the qc data in with the normal data The clinical patient data or you can make a separate file for it It's easier if you just put them all together And a lot of the lab data management people are not great with it A lot of them are not great with computers So if they're not great with computers, I just suggest put the clinical bacteria and the qc bacteria all together That way they go to the same file every day. They enter the patient results. They enter the qc results They enter all those data together Before I continue with data entry, I will go to data analysis. There's a feature in data analysis called isolates Here the second option the first option is exclude quality control So even if you put all of the patient results and the qc results together You can still analyze them separately. So who net automatically excludes the qc Uh, so that way when we do our annual statistics, we're not including the qc as part of those analyses So if you put your qc and your patient data together, it is not a problem Who net allows you to separate them later when you get to data analysis? So the data entry will be easier for the data entry people if you put the clinical and the qc results together That's particularly valuable when people are not really great with their it skills Alternatively, you know, you could have two files here I could have one file for my clinical data and then a different file for my qc data For people with good it skills often they would prefer to do that The clinical data are here as a big file. The qc data are here as a small file So it just depends on how good you feel about the it skills of the staff Uh, if you do put the qc in with the patient bacteria, not a problem Who net can keep them separate at the time of analysis Okay, so here an identification number I'm going to put a very special patient instead of patient one two three four five The patient's number is atcc 25922. I will now hit enter Who net automatically tells you that the patient's first and last name is qc is atcc The specimen type is quality control Specifically, it's an E. Coli If I put atcc 25923, that's a staph aureus. So who net knows all of the standard qc strains So here I am in data entry. Let me put it back to the E. Coli So ampicillin I click inside of ampicillin Normally on the right hand side of the screen So I'm going to change this from qc to blood If this is blood and I go to ampicillin, I see well, it still knows it's a qc But anyway, um because of the atc c But let me do it. Let me just let me just Save the isolate. No I'll click clear start from the beginning. Okay If I go one two three four five and I go to E. Coli I go to my E. Coli panel So here you see 14 to 16. That's the intermediate interpretation range So if I put a clinical bacteria, this is going to be the intermediate ris ranges It's the interpretation categories However, if I change this to a qc string I'm trying to change there atcc 25922 It is not telling me the interpretation range. It's telling me the qc range So anything between 15 and 22 is the expected result if I have a so if I put ampicillin 20 sensitive continue But if I put ampicillin six You see this symbol here the at symbol and that means the result is out of the qc range So who did is telling the person? In one way to be cynical. I'm sort of helping the person to cheat But if they're going to cheat they have a lot of other ways to cheat because they already know what the qc range Is supposed to be but when it is telling you that this value is out of range So I'm going to go now go to save the isolate Save the isolate and then I continue with the next one then one of you my database and I look over at the The first and last name is atcc And the specimen type is qc. So who now can separate those at the time of data analysis Does that answer the question? Yeah, it's clear and thank you Sure what they do in argentina because argentina is always one of our model countries. They've been doing this a long time Is they give people scores? They score people on how many strange results do you get and again strange does not mean wrong But strange often does mean wrong Uh, so a lot of times at the beginning like in the philippines They looked at strange results and they had a lot of strange results with feedback The number of strange results decreased significantly There were still strange results that were usually true So once they know what is strange and not strange then they can do a better job of fixing their own data So the total number of strange results in the philippines went significantly down But the proportion that were confirmed went up So so argentina they score people they score people on percent strange results They score people on completeness of data And they score people on compliance with the qc protocol Their agreement is to test the atcc e coli staff and pseudomonas every Every two weeks So at the end of every month, they should have two qc e coli two qc staph aureus two qc pseudomonas And that's part of their score In fact in the philippines they've done a really nice job there is they have an annual meeting And every year they get a variety of scores and they have to give a presentation and they put a poster together or a powerpoint And they get scored on a lot of things completeness of testing quality results discussion with the infection control team interventions you've done And then whoever has the highest score gets a prize the prize is often a free laptop or something else So by putting in these scores It just helps to incentivize people to look at their progress identify the deficiencies by making it this in this friendly competitive way You know, it's nice to give people recognize people for their work um The same lab cannot win two years in a row Otherwise you've ended up with the same lab winning every year And it's not only about the data it's it they include they get a score for data completeness they get a score for The efforts they've done with their pharmacy Have you had a stewardship program? Have you had an infection control thing? Give examples of what you have done with antibiotic resistance. They also get scored on their eqa program There's a national eqa program So you may want to think about this it's sort of thinking about the long term About how to build a network and the fact that the philippines bribe their people with a free laptop is a nice incentive Okay, and it doesn't have to be that it can be other things that are not so sometimes just a certificate You know, it's a free thing, but it is recognized. You did the best in this year, you know for the qc or something like that Um, you don't want to recognize people's bad quality. You don't want to punish people You work with them individually on the deficiencies, but people will do amazing work. They do appreciate it to be recognized for it I gave you a long answer for a short question Okay, we have about 10 minutes left. What would you like to discuss now? I mean as long as we're on the screen I do want to emphasize the importance of configuration of panels here. I'm on here. I see all antibiotics But if I put staph aureus, I just see stuff. You see the list has gotten shorter. Let me change that again blank So here I have all the antibiotics, but if I change this to staph aureus I only see the staph drugs if I put e coli I put the e coli drugs Pseudomonas I see the Pseudomonas drugs Or I can just say all antibiotics or I can say staff I can also choose it and force it to be a certain set I mentioned that because we are still inside of data entry But the panels you do not define the panels during data entry you define the panels I'm going to click on exit and I slip no File modified lab. I would go back to configuration Antibiotics on the right. You see the list of all antibiotics. There are 43 antibiotics Here under panels You can see with the different check boxes Which are the different antibiotics that go with the different species if it's a staphillococcus These are the drugs I want to see if it's a gram negative. These are the drugs I want to see So by being careful about this, I'm just going to do this a bit randomly Well completely randomly so you can decide what I'm going to cancel safe changes. No So that is the panels. Let me go back to panels If I go to staph aureus So if I say staph aureus, it's now showing me only the panel drugs So even though this concept is controlled by configuration It is used by data entry to make the life of the data entry person easier People using backing this is not important because they're doing data entry in a different system But for people using manual data entry these panels are very valuable Other questions Gabriel and team if there aren't Further questions right now for the data entry topic Then do you have some suggestions what you would like to go to next week? And is it it's the timing appropriate to do a national configuration? You know a topic on national nationally configuring uh who net to facilitate data analysis Yeah, maybe I can say something about Uh the national configuration Uh before you do that one. I would suggest maybe Uh the different kind of data analysis Uh that you can show us That would be presented first then later on we can see the national configuration Right. So there are two ways to do this. I agree. We should focus on data analysis There are two ways we can do that. I can show you data analysis using my standard WHO data Which will introduce everything And then after that session then we switch to the Ethiopian data So that's one strategy. The other strategy is I teach you all of data analysis But I don't show you my data. I just use your data to begin with So basically we could have one session on data analysis Well, you know one initial discussion about data analysis The question is that in the initial discussion of data analysis Do you want me to do that with my data or do you want me to do it with your data? Maybe I can provide the data but You know our data is small Yes, maybe if you have a big data Uh That would be better to use the WHO one if the WHO one is bigger What I don't have is big data, but when we distribute who net I give you everybody I mean it's right here. Uh, so if I go to save isolate data analysis When we give people who net just ignore there are four files here. Just ignore three of them Uh, this file here is called w 0195. These are data from january 1995 So this one month of data is what we give to people to to play around with. There are 622 bacteria in it So, um, yeah, sure. I'll show so I'll based on I'll base the presentation on my data The question is whether to show ethiopian data at the same time or leave that for the following presentation Now let's do it that way for the next one. I will do a standard normal who net analysis Presentation using my data and then on the following call. We will switch to ethiopian data Yeah, that's great Yes, uh, there's one Important point for data entry that I did skip over and since we have a couple of minutes. I do want to get to it I showed you data entry and I want to open this jump file And there it is and view my database And that's what we just entered today So you see how easy it is to use this open data file or choose one of these files at the bottom. That's easy But I'm going to show something a little that people can get confused. I'm going to go to modify lab And we're going to add two more drugs. I'm going to add You know, let me add ampicillabactam and let me add Tigger cycling okay And I'm going to go to data fields modify the list and I'm going to for example The patient diagnosis And the name of the physician So you can see how I just used my configuration to add two more antibiotics when I added two more data fields I'm now going to try to open that jump file that I created half an hour ago Hoonan is telling me there's something he's changed here The data file and the configuration are a bit different. What do you want to do? I could just continue with data entry, but I'm going to say review the differences. Why are they different? Hoonan is now telling me Your configuration has two more antibiotics and two more data fields But your data file doesn't have those fields Because I just added them. So you're here in the lower left hand corner. I'm going to say add fields to the data file Do you want to add the missing fields to the data file? Yes, I do Additional fields have been added. So now you see diagnosis and physician appear on the screen Episulabactam and take a cyclone appear at the bottom So I just wanted to raise this to you as a concept I not that you'd be experts at it, but this this doesn't happen often But because because people don't use the add add information they have a configuration. They're happy with it But if you add more antibiotics and add more data fields And you try to open up one of the files that already exists Hoonan will ask you do you want to add those additional columns to the database? So it's just a way to keep them in synchronized when I showed you this demonstration half an hour ago Take a cyclone and episulabactam went on on the list Diagnosis and physician were not on the list So there was that message about review the differences that allowed me to add them That's all the time we have to say about that, but you're just aware of it So if somebody sees this message This is different. Do you want to review the differences? This does not mean there's an error It means somebody simply added something You've added something to the configuration. Hoonan is simply asking you do you also want to add it to the data file So just want to leave this as a concept Uh, you will come this across in the future But don't be worried about it. There's a very simple explanation. There is no mistake Hoonan just wants to know if you want to keep these things synchronized Yeah, maybe one question regarding this one. Yes Uh, suppose you have two data files coming from two configurations. Yes, uh Maybe the configuration the configurations they differ by let's say two variables Yep one one of the variables Was actually either defined variables, but they are telling us the femtis You understand john so in the interest of time. Um, yes, what I just showed you allows you to synchronize them Uh, I suggest that maybe the first thing we do on the next call is talk about that Okay, that's okay. Yes, it is also related to this idea of making a national configuration So we can do that at the beginning of the next call