Computer generated transcript

Warning!
The following transcript was generated automatically from the content and has not been checked or corrected manually.

I think we're live. Yes. Fantastic. Hello, everyone. Um Lovely to see so many faces. All 100 and 50 of you. Um My name's Dottie or you might know me as Deta. I'm the date study lead. I've been very active on the whatsapp and in lots of the emails. So you might want to recognize me. Um I'm here with two other members of the steering committee. So we've got Ashvin um X Nans chair and we've also got um David who is a senior registrar and very, very experienced in all things, collaborative research. So um we have this induction session for collaborators and center leads, which is basically for anyone that will be involved in data collection. Um Oh, yeah, I also just wanted to thank the, our expert advisory group, which is a group of senior clinicians that we have on board that have helped guide us through designing the study. I've got some very big names um which is all in the makings for a really, really exciting study. So, if you guys have already seen our launch video, it's around um dates is all about ti a clinics and who attends them. I won't recap too much because it's all there in the Google Drive. Um But today, we'll mostly be going through how the actual study will run, starting a patient identification and then all the way through to authorship and what you guys get out of it. Lovely. So, in terms of patient identification, the first step of the study is finding your ti a clinic list. So this is down to the center leads. So your supervising consultant should be able to direct you to whatever system your local clinic uses to keep track of attendances and appointments. So this is the main method, how dates identifies patients that are coming into the clinic. So this is actually one of the strengths of dates is the our previous studies. N six previous studies in epilepsy such as Becks and napper and then also MS such as dimes, they didn't have a single centralized clinic where patients were seen and collaborators sometimes had to spend a huge portion of their time searching for agile patients. But we're just looking through T clinic lists which is really easy in comparison. Um to know we shouldn't be using any other methods to identify patients such as looking at diagnostic codes or snap data. Um This might mean we miss patients that should be included or we might just be wasting collaborated time with a less efficient method. So if your center is ready to start for the 30th of December, you should already know or may need to find out how to access lists of recent ti clinic appointments. So you can get off to a really, really good start. So we have described dates as perspective um and this means that we use eligibility period that is in the future. So a period that is beyond the time of study launch and beyond the securing of local audit and Caldecott approvals. Now, naturally, since we're not actually sitting in on each clinic or conducting patient interviews as will be performed in some research studies, any data entry will inevitably be retrospective as it comes from existing clinical records. Although as we are looking forward in time, you'll need to identify patients as a study period goes on. So if you do start on time from the 30th of September, this is actually a first for non neurology studies. But we believe this approach will be give us the best case as attainment, um the best quality of data collection and also produce the highest quality publications from the study. So I'd also like to say a huge congratulations to the 20 or so centers that are ready to start as planned on the 30th. We've already received your approvals and they're all filed away. We also have around 35 or so with approvals in progress and another 45 paring for approvals and or seeking support of a supervising consultant overall, this gives us a potential for 100 centers which if they can complete registrations and data collection. This will be the largest study um of in depth ti a clinics to date which is really really exciting um centers that have secured the relevant approvals by the 30 Sember that have not secured sorry, the approvals by the THT Seber will need to update the dates, progress sheet um to ensure that the steering committee is kept updated. So where there are any challenges, they should continue to keep us updated and seek our support. But for the sake of for the sake of the presentation, I'll keep saying the 30th of September. Um we've added a column onto the progress sheet for um center leads to announce when they plan to start their eligibility period. So anyone that's ready to go as of now or as of the 30th can insert the 30th and then anyone that will be should liaise with us directly. So in terms of who is eligible to be included. So then using these ta clinic lists, you can look for patients to have appointments in the clinic from the 30th of September to the 27th of October. Again, if your center isn't quite running on time, then this date can be changed. We also have an exclusion criteria and the only criteria is individuals that are listed on NHS England's national data opt out. So to adhere to this, you can send your ti a clinic lists as they become updated. And available to your audit department and ask for their assistance and they will run you a list of patients against the spine of the National Day to opt out if it is in England and they can just exclude them for you. So our data collection Excel spreadsheet is pseudo anonymized. And for those unsure of what this means, it means the data is anonymous as in it's practically impossible for others to identify the individuals. But you also have a key that can be used to deidentify the patients at the local level. So it's not really anonymous while this key exists. And this key should only ever be hosted on an HS NHS computer or equivalent and never ever ever shared with us. So this is really important because it's for the purposes of traceability and avoiding duplication. So you can double check that you haven't collected the same patient twice. And it also enables us to check out the quality assurance such as completeness and data validation as well as compliance with NHS England's check for national data opt out and also the potential for a follow up study. So when you do went to leeds, do create this Excel spreadsheet, which might look a little bit like this. We ask that you keep hold of it until we ask you to destroy it. But it's very important that you keep hold of it on your computer or on your, on your supervising consultant's computer. Um And that it's an NHS computer. Lovely. So, in terms of data governance, I thought about giving a long spiel about this. Although the NHS e-learning sums out perfectly, um There's e-learning modules for the Republic of Ireland. I'm sure there's equivalence that you can go through. And we do ask that this is completed um prior to any data handling. So in terms of actually getting the data, we'll be reviewing quite a few information sources. So first of all these ti a clinic appointment lists, they're ident, they're essential to identifying eligible patients. The referral, there will be some data points you can't do without sourcing the ti clinic referral. It's often a letter or a form um and where it is varies depending on trust. So we can't exactly tell you where it might be. But if you're unsure do as a center lead or your supervising consultant, ti clinic letters. So in interest with paper notes, you might need to request these shortly after the clinic or in interest with electronic notes or electronic records of clinic letters. This could be available on the day of the clinic or days to weeks after. If it needs to be dictated overall, you might need some assistance in finding how to source each of these and it will vary. So we can't necessarily help you from where we are. But um with some asking around of your supervising consultant and your center lead, you should be able to have access to all of these. So in terms of the data collection process broadly, there's two approaches at the center lead may use distribute the data collection workload within their team. So approach A is they may distribute patients evenly throughout their collaborator team wherein each collaborator collects X amount of patients. Alternatively, they may distribute subsections of the XL for each collaborator to fill out for every patient. Now, in our experience, A seems to work better. But if there is an information source, that one team member has lots of experience with or is the only one with access to. For example, cardiac investigations, it would make the most sense that they complete that section for all patients. But at the end of the day, it's up to the center leads how we'd like to run, run your team data completion. So in order to be included in the study submitted, data sets need to be filled out more than 95%. So we'll check all mi datasets to ensure the excel cells are at least 95% populated. And if they're not, we can't include the data set and analysis. What this means is, for example, if we take this data point, this is a subsection of the XL. If you were to leave the diabetic status cell blank, we wouldn't know whether this means that you're not sure whether it's not documented whether they're not diabetic, whether it's unclear. So we can't include this as we can't analyze it, which means no output. So all cells have an option for not documented or not reported. So if you really can't find the data, there is an option to say so and unclear or not documented counts as this data point being, being populated. So this counts as 95% of filled out. If you really do struggle with a data point, you can ask your center lead, who in turn might ask us as a consultant who in turn might ask us. Um We're more than happy to help to help figure out where you might find certain data points. It's very unlikely that things like blood test results aren't available somewhere. Um But please liaise your center lead who might in turn, Liaise supervising consultant. Now, date of validation moving on to this. So this is a process that ensures all your data is accurate and it's appropriate for us to analyze. This is a process wherein we, the steering committee will select 10% of your total cases for the data validated to go through and recollect from scratch. So it's really important that they're not involved in the original data collection because this does mean they're not blinded. So we will check whether your data and the data validators data lines up. And if it's over 95% in agreement, we accept the data set. If it was to be less, we'll select a further 10% of cases And if again, there's less than 95% agreement, the center lead might need to review all the data extraction. Great. And in terms of the timeline like so we're looking at four weeks worth of patients attending the clinic. So whether your center does start on time or it's delayed once this period is over, we hope you return the patient data to us onto our secure NHS email within the following weeks after that. So altogether from the beginning of the eligible eligibility period, you'd have eight weeks to return the data to us. Although we do understand that some centers might have a much higher case loads and we do have space to be flexible around this. Although if no data is returned, your center will not meet authorship criteria and will not be cited in subsequent publications. Yeah. So once data collection is complete, the validated and anonymized Excel spreadsheet is to be sent to this email. So this is really important because NHS emails are encrypted and sending even the anonymized patient information to a personal or non NHS email would be a breach of confident confidentiality. So please do not send it to our NS email. Great. Um So as well as the actual data collection, we also have a center survey which is intended for the supervising consultant to complete. It's just an Excel similar to the actual data collection. One, it's just another tab of it which is collect some data about how the local ti clinic runs, who runs it, how often it runs that sort of thing and that only needs to be filled out. Once we also ask that center leads vising consultants, send us triage criteria for the local ti a clinic cos this is also so we can analyze and look at that could also be another potential output. Um Yes, the center survey will be included in the Excel spreadsheet, which we hope to get out today or tomorrow. So friendly reminder that anyone involved in the project after meeting contribution requirements will be listed as an author on all outputs and as dates is planned at the moment, we have quite a few outputs planned. So it may end up being more than one publication. Um And Nan Zig as an organization adheres to a full corporate authorship, which means that all authors will be listed under NANS alongside their roles in the study which makes your PUBMED site and searchable author. So for the actual full authorship requirements, we've got pages 2526 of our protocol which lists these in detail, although vaguely the teams look like this. So they're consisted of one center lead who is tasked with registering the study, getting local approvals, recruiting collaborators and may help in data collection. We've got collaborators who gain access to the relevant medical records and collect the data, who may also be doctors or stroke nurses or stroke research, nurses who may take additional responsibility with local supervision and or support to inform data collection and aid in complete case ascertainment. And of course, each center will require supervising consultant to sponsor registration and offer support to the team. So this is my part of the presentation done and I'm gonna hand over to David who's going to give a short demo of the Excel spreadsheet. Good study. Ok. Yeah. Ok, great. Um So there's a prerecorded example of the Exalt template that I'll just talk through and it's about 25 minutes of the recording. Yeah. OK. So you'll probably have the XL somewhere in say a shared drive inside an NHS computer and we'll just open it just here and on this first page, we have the central service and I'm not sure how to remove the banner at the very bottom, but it is your center top and that's me just opening and making sure I've got all the, the files ready to show here. We have our data dictionary which is in our toolkit which every essentially should provide through the collaborators. But the goal is ultimately and what we've achieved for dates we hope is that we've managed to squeeze the content of this grotesquely large document that says lots of details about what every variable into the Excel itself. So you shouldn't need to be jumping between the Excel and the spreadsheet or, and a detailed data dictionary. So, for example, for this first sell we are in the center survey which the ser supervising consultant would complete potentially with the support as the center lead. And for our first one has local referral criteria been sent to the central steering committee and we give them the address there. So let's say they have and that will mark that as a complete cell, which is great. Then the next one, if your center uses a triage criteria, how they center. And for example, this center does have a triage criteria and they have center. And for this well center, they don't use a PCT two criteria in the triage, which is great as they should really be doing. So, and at the end of the survey, it will give you a completion rate, which of course, ultimately should be 100%. You should be filling in the right cells jumping on to the next tab. So in the left column, we've got every patient's record ID that should be really unique to the individual as in one patient shouldn't have multiple unique IDS. And we'll come on to that specific point in a bit in every t what we've got called B is the data collector initials. So if I'm doing the data collection, my initials would be DH it's not the patient initials, it's your initials. And that's to help essentially track who's doing what sure we don't kind of cross cover and things like that. So in this information sources, tab, there's many information sources that we recommend using to find the right information. But the ti a referral and the ti clinic letter are probably by far the most important aspects. Of course, one of the primary objectives of the study is to determine the time from referral to attending ti clinic, which according to guidelines, the aspiration would be that to be less than 24 hours. So those are really quite important documents and they have a lot of the study information is in those two documents. And here's the other aspects here. And what I'm probably going to shortly do is for each cell, there's often a drop down if it's categorical information, but more often than not, what I would recommend people do is you type the letter, a type, type, type, a type tab, type a type type and then you can quickly click through cells without needing to use the the cursor or the mouse on to the next tab. So we've got record ID one and doing the data extraction for that patient at the center and for this person, they could be included or excluded. And what I've realized for this first patient in our in our center is they did not attend the appointment. It's important that for the context of dates, we are including patients who did not attend their appointment for a very brief specific data collection element. And in this yellow box, you can see if they did not attend, they are included limited data collection, complete these columns E to oh and then skip to tab seven. So I actually can't enter an exclusion reason. Indeed, because they're included. And this is a hypothetical patient which I'm kind of making up as, as we go for this one. Do they have a postcode? Yes, they do. They're in the UK and oh, what do we do here? So it says enter value 1 to 10 for England, Wales Scotland to use that hyperlink. And actually the hyperlink is above, it will take us here. And I'm saying my, my patient lives in Westminster, which is that postcode there. And when I click, get the data, here's the postcode. The location country Deel is it? And the rank is that I'm gonna copy the rank for the deprivation and pop it in the box. And what you might notice here. And there is another kind of Excel tip is if you use control shift V, you can copy or you can paste with while removing the original formatting of the text. So you don't end up with lots of different fonts. Next, what I'm doing is I'm doing, I think the distance from the patient's postcode to my hospital's postcode and that will give us some details about how far are they from their, their local service, which one could imagine might be something that could derive delays if you live really far away from the nearest ti clinic next one is a never mark of deprivation, which is distance from the same. And again, we're gonna go to the link and I think you just click on it or you, I think you might be click control and they click and they'll take you to link and here postcode, calculate. And for all of these aspects, we said it's in miles. So we're gonna correct. So copy. So highlight control c then control shy to pass it to the box. And actually for this patient who hasn't attended the appointment, if I go back to that cell, it said actually it's only columns up to. Oh, so we don't actually need those two aspects which will be pretty difficult to find given the half an attendance point, there's no letter and I went back to the box which said, please then skip to tab seven and we're gonna do, I'm gonna show you this part because I think it's very important that we don't end up with t the patient's been entered twice who are actually the same patient. And ok, so in record ID number one, we're completing this. So the clinic location was my local hospital which was Maidstone and there's I couldn't find a letter. So the clinician's grades unspecified, but I do know it's my stroke consultants clinic. I know that this appointment was at 9 a.m. today. And what I can do is I can just type 09. I'll come up and what's today's date and that's an important one. So if you get an error message like this, it means you're typing something in that the cell does not allow. I would suggest you click cancel as opposed to retry. Cos if you click retry, you'll just get the same error message again. In terms of this patient did not attend. And I don't know why they didn't attend, but they could have been admitted to hospital. In which case I would enter there. But I don't have the option to enter into that cell coss it's blocking me cos they're not appropriate to enter into that cell. And usually if a cell is blocking you from entering and it makes sense, then that's a good sign. It's not an empty cell. And for this one, so the current record ID is correct as one and the next column which is column. Oh, did they have a previous record ID? And this person does not? And for the case in point, let's say my next patient is a patient the next day who gets the same appointment rescheduled. And because I know this patient is actually the same, I'm just going to copy and paste their details down. And for the purpose of this example, I might have made a mistake and since that, but obviously I need to enter their employment data and the two of ourselves in the demographics t but I'm just going through this example to make sure what we're very clear for this person. What we want to do is we'll enter the grade of the specialty consultants clinic. The time of the appointment, it was the next day at 10 a.m. and they did attempt and for the purpose of this aspect, this person did previously, they attended this clinic but previously did not attend. And the point of this bit here is, oh, it's not matched. We want the patients to have the same ID because they are the same person. So what I think I'm doing is I'm reading through the guidance box and I'm going to amend their record ID to ensure it's correct. So we know that this patient is actually the same patient as record ID number one. And that matches. And what you would do is you dili go dili go through all the tabs to make sure that that patient matches up correctly with all the tabs. So next, I'm just gonna talk through the pa past medical history tab. So of course, my issues will go here. I should be fixing the record ID for this patient. Given that we've got two individuals with the record ID number one and just mention their data. And I tried really hard, but I couldn't find the patient's weight. I could find their heart and I could find somewhere their B most recent BM I and for most of our cells, if it's a number and it's not documented, you would be typing MD for this patient, the stroke clinic or the ti clinic letter says nonsmoker or they do not smoke. And to me, as a, as a researcher, I find that quite annoying. And also as a clinician, I kind of want to know did this person used to smoke and then stop or did they never smoke? But I have no idea if they're an exsmoker or never smoker. So they're a nonsmoker. The clinician didn't document about whether they vaped or not. So it would be not reported. They did say that they've got hypertension and I could see that they're on one angio hypertensive called amLODIPine. And we've got a detailed list of common antihypertensive in the daily diary. But I think lots of you as medical students or doctors will be familiar with. What are BP medications? I can see that this person had a previous HB and C with a HB and C of about 45. So they're clearly prediabetic. And I think, you know, I'm just making sure I read through all the guidance tabs and for this person. Um and a lot, and this is quite an academic point, but a lot of the time it's quite diff if somebody doesn't have a condition, it's often gonna be difficult to know. Do they not have the condition as in somebody? States they do not have it versus we just don't know whether they have it or not. So, for this patient. I thought it's unclear whether they've got liver disease or not, but I know that his renal function is completely normal. So I'm gonna say no renal disease. There's no comment about lupus or rheumatoid arthritis. So it's unclear, not documented. They probably don't have it. But we don't know for this example. Er, the clinician said they do have a past medical history of migraine. So let's click. Yes to migraine. Um, There is no past medical history of any cancer and the clinician says they had no previous stroke but they failed to mention about whether they've had ati A before. So I'm gonna say unclear or not documented. And I think what they're gonna say is I can see that they've got a diagnosis of af but that's prior to this index episode. So an index episode is about this event. So what was the event or episode that led them to be referring to clinic? And I know that prior to all this, well, the patient is known to have atrial fibrillation for XL. It's previous VT E and there's no document or comment about that. There's no comment about bleeding. There's no comment about these genetic factors in the context of Angina or a heart attack. I can see it's documented that he's had an M I before. There's no comment about Angina. Of course, Angina is very common if somebody's had an M I before. But there's no comment. I say M I only there's no comment about peripheral vascular or arterial disease. And for this one, so I then decide, well, I need to access the cor like investigations for a later section of the form. And I look and I see that this person has an echo with a normal ejection fraction. And by normal, we define it in the yellow box as 50% or above. So he's got a completely normal echo. I'm just gonna say that he's got no heart failure, cos that seems pretty accurate to say. And a lot of these judgments are quite pragmatic and sometimes a little bit subjective if you're deciding between no and unclear. But he's got a normal echo with normal heart valve. So he's got no valvular heart disease. There's a comment saying he does not drink alcohol, so he's got no alcohol excess. There's no comment about these psychiatric diagnosis. There's this is actually called a not very bad 70 year old. So he's doing quite well. So there's no comment about him having false or not. The clinic says that he does not have a family history of stroke or ta the clinician also says he does not have a family history of M I and then on to the next tab. So the next tab is about medications prior to clinic. No. And as you can imagine, so I've made a mistake there that for this patient, what you do is just systematically go through and make sure the record ids are all correct for the individual. I'm popping my initials in just quickly here and all I need here is access to their preclinic medication. So usually there's a clear information source and once you have that, it's just really, really, really quick to go through this. So I know that he's on Aspirin and Apixaban cos of his af and ischemic heart disease, but he's not on any of these medications. So it's quite quick. No, and there is an option to say prescription's not available for review. But generally speaking, that would be quite unusual to use. Usually you should be able to access the medication history provided, you've tried hard enough and half a team that can, can guide you into where this information is and that team may not necessarily be the supervising consultant. It really could be any doctor who knows how that the record system works. It really doesn't need to be a consultant and won't get argued that maybe an fy one or fy two doctor might be better at finding this information than some consultants per se. And here we're going through these sections. So this patient uh did say that he's got a back end of high cholesterol, but that's most recent thoughts, maybe his parameters aren't too bad. And for the example of a lot of these cells that you can often say if it's normal, you can just click N and then T and N and then T and then it'll sit normal each time. So we're making really quite good, good progress. So this would be about ti a onset time. So the best in source for this information about when the ti A occurred will of course be probably the referral letter or the clinic letter itself. And unfortunately, despite trying, I couldn't really figure out what was the actual time of his ti A or this ti a like event. But it was very obvious what day and month and year it was. And in one of the sources, it sounded sudden onset, there was no comment about whether it was a maxed deficit of onset or not. Um It sounds like maybe this was something that gradually developed, but maybe we start unclear. It was 30 minutes. It wasn't clear what part of the brain it was affecting at least from the documentation. And for a lot of these free text ones, there will be an option to write zero and that just blanks itself and looking at the symptoms that were documented, it sounded like he had some strange visual aspect that was affecting both eyes, but there was no double vision, there was no positive visual phenomena. There's no visual hallucinations as in they weren't, there's no comment about it. So I have to say unclear. So that's not documented. And for a lot of the cells here, if you were to, if it's clearly not documented either way you would be typing you and then type and then you then, then tab and that would jump too quickly through the section of the Excel spreadsheet. But more often than not, there will be some symptoms that are documented as being present, some symptoms that are documented as absent and potentially quite a few symptoms depending on who is doing the clinic or documenting. Well, we just don't know it's unclear slash, not documented. And I think with time, what will happen is your first few patients where you do the data collection for, it might be quite time consuming. But in our pilot, it seemed that we were getting down to about 30 minutes per patient. Of course, there will be some patients where there's very little information needed like a patient who does not attend the clinic. But in other words, there might be quite a bit of information. So here I'm just running through in this clinic letter, it said that there was no motor symptoms and it said there was no sensory symptoms. So I'm gonna say that there is none of those symptoms as it's not unclear or not documented. It's clearly documented. There was no sensory, no motor symptoms. And when I was reading the letter, it s did say that there was no headache, but it's a bit unclear if there's any potential aura. They said there was no motor symptoms whatsoever. So I'm gonna say there was no seizure activity but there's no comment about whether there's this postevent uh abnormal symptoms like a postal phase. Like a lot of the time, there's no comment about incontinence cos why would there be if you had just transient facial symptoms? Um So that would be unclear and not documented. And of course, this is a theoretical patient. But what you would probably do in is you might have a split screen between your Excel document and your information source, which hopefully might be an electronic computer system and you're jumping between each section and competing it as you go. Yeah. So in our next tab, we've got our initial presentation to service. So for this instance, let's say this individual presented to his general practitioner. And although I couldn't find out his time of symptom onset, looking at the referral letter, it looked like he presented around noon that day. OK. And I guess this is quite the best one that he's presented on the on the Monday get and happened to get a scheduled appointment at 9 a.m. the next day. For this patient, there was a ABCD two score calculated. And in the referral letter which is interesting, it's potentially arguable whether that's a valid thing to to include it looked like he got given Aspirin 300 at that time. And I think this presentation might be about to cut out for a second. But this column after initial presentation was a referral to a next service. The non ti so this patient goes straight to the TI clinic. Next, he doesn't go say via A&E or, or whatever. So I think it's gonna come out for a second. But what I'm going to enter for him is that he goes straight to the TI clinic. OK. The service referring him to the Ta clinic was his GP. And I think a lot of this time I'm using my mouse and cursor to use the dropdown menu, which is a lot slower than say, just type it in the lethargy. And then t and I think here we're gonna cut out again just for a few seconds because we're just gonna enter that E word to his GP. And I think the, the important aspect for a lot of this is, although it might look daunting, we will have exceptionally good information that covers many, many different audit and service aspects, which would be able to give some really good detailed information account about the TIA A service in the UK and Ireland. So here was his time in front of TI Clinic. We've learned from previous work that we should, no one should ever trust excel to handle date data because it will often mess if it'll turn it into us dates and then it's kind of irredeemable to try to figure out what the true date was. That's why I've got every date as a day, month and year and a separate cell. We've now kind of recapped on TAB seven, which is the ta clinic content aspect. And I don't know why. But for this instance, the clinician said the diagnosis is not ti A and it is migraine and he's got a background of migraines. And ultimately, it's about what the clinician says. It's not about what you think from reading the records. And because this is a question about diagnostic uncertainty, it's clear that there's no kind of, there'll be some patients where the C will say we don't know what the diagnosis is or most likely this is whatever or and if it looks like there's diagnostic uncertainty, please take, yes. If it looks like it's pretty certain that this is whatever diagnosis is taken up, these variables for BP, which at my local clinic definitely always gets recorded. And these are some quick questions about advice that usually should be given in the event of TAA for my individual. He didn't have ti So I said so it was reasonable that I didn't get the advice, but we should be answering those questions regardless. And I guess my individual was a bit interesting because at the end of the day, the stroke patient does not think he had a ta. So he was and he was already on both an antiplatelet and anticoagulation beforehand. Mhm And for him, his example, he would take yes, for aspirin, he got a loading dose, he goes back to his maintenance dose. He's on a once daily dose regime and then very quickly, he could say no to clopidogrel, no to all the other antiplatelets and no to all the anticoagulants apart from his Apixaban, which we know he's on. And this one stage two stage is probably written out a bit clearer. And in this example, here's an example that doesn't really apply one stage two stage. But quite often what might happen with a a ta or stroke is it might be reasonable that they get say 21 days of aspire or they might get 14 days of high dose aspirin and then low dose aspirin. Er and we know that across the country, it's possible that people might be doing different things. So he was yes for anti any anticoagulation. And as an example, if he was no to any anticoagulation, you wouldn't be able to enter anything in the other cells because you wouldn't need to and it would actually be forced to be blank and he was already on a PPI like omeprazole. I think we're making good progress in this is almost the end of the the the data element. But for him, we got his medications, he was already on a statin. So it was quite a quick one. So with no changes, he was on atorvastatin 80 as you said he was on amLODIPine, which wasn't changed. Why would you change it if he's had a migraine? You want me to just answer it in here. Ok. And then you've had this never box for tamperensis after clinic and for this indi individual to, and I think it's very common in the NHS that we do a lot of blood tests. But for this individual, they all got done at the same day as the clinic, which might be common practice unless it's been done recently. And although I'm using the mouse here, really, what I should be doing is writing the letter PT PTP T. OK. And there's some other tests that may be done in some stroke clinics for certain patient groups. But he didn't have an E sr he didn't have any thyroid function tests. He didn't have any fancy young stroke bloods. He didn't have any screening for sleep apnea. There might be some patients who present in like a with this first seizure who they might request in the aging. And our gentlemen, they did not do any cross imaging. And as a result, we actually can't enter anything in these cells. They're not incomplete, they're actually intentionally left like and you can't enter them even if you tried. And if you get that as an example, like let's say you're trying to enter that, he's got a CT angiogram as part of the clinic work up and you can't enter it in. It probably means there's been an issue in a previous cell and that sh guidance should be clear in the yellow box or if you at least scroll slightly to the left. Although we generally wouldn't recommend that ta like presentations, get a CD head. Somebody thought it was reasonable to do it for him. Maybe because he was on, they mentioned maybe in the quest that he was on anticoagulation. And it also said to rule out hemorrhage. So that was a combination of reasons and then onto our card investigations. So there was a comment that he had an irregular pulse at the clinic. He had an E CG that was done on the date of the clinic and he did not have an ambulatory ECG requested. So in stroke and tia A, there's a increased interest in trying to diagnose af that might be paroxysmal and it might not be a very common event. It might be missed, say on a quick 12 A DCG. So doing some for ambulatory EC GS is quite common. But for this individual, we already knew he had f and there's no comment about his clotting or bleeding risk, which is reason it's not given that he's not been started on the anticoagulation. And for this last time, it's quite brief, but ultimately, it's very important. Yeah, the main out of here are, were there any onward referrals to any other service? And as you can imagine that could be to neurology, it could be epilepsy services could be to diabetes, nurse, could be community podiatry, whatever it doesn't matter. But if there's any onward referrals from that TI clinic, please list it and it special was discharged from clinic. We've got on any other comments box in case you wanted to highlight that maybe there was an investigation outstanding or maybe there was something unique about your individual who you wanted to highlight to us. And I hope that gives everyone who's attended and watched this on, on replay quite a lot of information about the, the spreadsheet. But ultimately, the spreadsheet itself will be quite hopefully quite self explanatory in terms of the content that's outlined within each of the cells. And at that, what I'll do is stop sharing. Wonderful. Thank you so much, David. Um We've been getting some questions through and me and Ashvin have been answered answering them as we go. Um One of them was around, for example, there was, there is one cell in the XL that asks whether someone's got a history of M I or Angina. And they've asked that if they've had an M but there is no mentioned documentation of Angina. Would you say M I only or not documented? Oh, I think I'm muted for every seller asks a question if it positively affirms any aspect as in, let's say that question, which is M II M I or Angina. And it says, yes, they've got ischemic heart disease with M I. But there's no comment about Angina. Of course, we should be writing. Yes. M I, we shouldn't be deciding if they don't say no, to every single possibility that becomes not documented. It would be. Yes, for M I wonderful. Um Yes, we've also just had some questions about data lock, which we haven't got in place yet. Um If I heard correctly, you would like us to store the Excel sheet on a secure hospital login at JC. Uh I have seen our surgical reg storing their audits on the gene generic Genser login. Would you want us to do something similar with our stroke team? Yeah, I think the, I'm not sure, I entirely understand exactly their set up. But the important thing is that the document exists within the NHS computer system. Usually you might have a shared drive where various people can access documents. But the important thing is the physical location is within an NHS computer. Wonderful, great. Oh, I don't know if it's worth running through any of the questions just from the top down dot E top down. Fine. Um For past medical history, are we expected to check the patient's previous notes, records or just input unclear, not documented, if not into eye clinic notes? Ok. So in terms of the expectation, it would be expected that you're looking through the documents around the time of the clinic that does not mean looking through every single hospital letter they've ever had. It would be make sure that you've accessed all the information sources we've suggested and that's been accordingly. Wonderful. And how recent should blood results be OK. So that will be that's documented within the XL itself. So you're looking at the most recent blood tests available and there's a separate dedicated question about when the most recent blood tests are what we're interested. So say for example, are they at clinic or not? Um quick confirmation, the 95% data completeness if something is not recorded and you document that it isn't recorded, will it count as that data point being collected? Yes. Um I'm looking forward to the analysis of which data points are not all collected. Yes. Um So we completely understand that some things might not be documented. Um Our data dictionary is quite in depth and naturally not every single thing would be documented. Otherwise the clinic letters would be miles and miles long on that. On that topic, we recognize that a lot of clinical trials, a lot of research, a lot of audit fails to determine when a question is no, is it no not documented or positively absent? And if we were to design this topic or this work in, say a XL mole do drop down where you tick the past medical history aspects that they do have. That would be the same thing for anyone where there's a missing tick if that makes sense. Um And it's something that a lot of studies don't bother to look into whether it's documented phenotype absent versus no documentation about the phenotype. Wonderful. Um Will OK. So Florence asks, will the spreadsheet be shared document among data collectors at each trust, each of whom can edit it? So this depends on your trust. If you do have access to one drive where you can all share the same Excel and work on it at the same time. That's great. Um In the past, I've sent the Excel back and forth through NHS emails when somebody's going in to do some data collection and we'd all work on the same one. I know whether Ashman or David, you've got any thoughts on that. Best practice generally would be if the document exists within an NHS um A file space where you can all access it rather than having to send between various NHS accounts. Definitely, we've got um Dr Green Caldecott approval. We've the audit registered at our site ready for the third 30th, but have not heard from the Caldecott Guardian because this is related to sharing data outside the trust. If we do not hear back, can we continue collecting as long as we do not send any information to the steering committee yet? And on that topic is this is this Jessica Green. Yes. Yeah, I think, could we continue that conversation in email just because it depends on certain trusts. But generally we do know that for most studies of this nature, that would be very common practice. Lovely. Um And Zhiqing Zang, do we need to record the patient's name and date of birth in the separate spreadsheet in addition to their hospital number. So that'll be the key. But in terms of what you do, you do not need to do anything, you need to do something that is sensible that works at your site. That could be your hospital M RN. That could be an NHS number, but it just needs to be something sensible. Wonderful. And ultimately, we never received that information. So we don't know or particularly care provide as you're doing it accurately. Uh Yes, uh only asking because of the screenshot. Yes. Um It was just an example screenshot I made. But you can include as much as you want or as little as you want as long as you can then relate it back to the um to the patient identifier. The one through however many patients you have. Um how many collaborators are we recommended to recruit? So, in our model, we allow up to 10 collaborators per center if your center is particularly big and has a big caseload. Um after liaising with us, there is room to expand that for more. Um Although that would need to be a discussion, how many patients are normally required in total. So as per the protocol eligibility or suitability for the study is not a patient number, it is the number of patients within the patient eligibility period that will vary considerably between different um centers. Lovely. Um Regarding data validation, does that happen internally or externally ie does the data validated, validate data collected in their own trust internally? Because generally, how could anyone else access the logins to perform the data validation within your trust? So it'll be somebody within your center team that ideally is nominated and named by the central lead. Yeah. So we have the center lead, we have the team of collaborators and one member of that team will be the data validator, which means they abstain from the initial data collection. So as everyone's collecting their data, they're doing nothing. And then once that data collection is completed, we select a random 10% of cases. So we might say patients 5, 1015 and 26 the data validated then goes to recollect all of that. We have a look at the data that you sent us from the original data collection and what the V has sent us and we look to see whether they agree and up to 95 we want more than 95% agreement. And for context, what happens in reality is we approach 100% and that's what we've achieved in in DS which is led by, by Ashur Lovely, just to confirm the patient identify key needs to be on an NHS computer. But the Excel spreadsheet can be on a personal laptop if in a secure NHS one drive. For example, generally speaking, when you're entering data, you should be using a hospital computer. Wonderful. I think practically as well, realistically, the data collection is going to be done at the hospital anyway. So it it shouldn't really make a huge difference to your ability to collect the data. Great. Um Can we have two center leads? So our steering committee position has been that we strongly avoid having two center leads at one center just because historically, there's been uneven distribution of workload. Um And it's not, it tends out to not be very far on one center lead. Um So our position is to avoid having more than one center lead except in exceptional circumstances which we haven't come across yet. Um Lance asks if our center is still recruiting collaborators and not ready for data extract from the 30th of September. Can we begin creating a list of TIA clinic attendees from the 30th of September and extract this patient data retrospectively. So I think that's the important aspect is whether or not your center has appropriate approvals in place to start. If you guys do have the appropriate approvals in place to start, you should aim to start on the 30th. And from a practical point of view, all the data collection is retrospective as it's from the existing clinical record. If you leave it a long time, you will run into issues identifying the right information and meeting the suggested deadlines for the the study. Yes. And actually, I'll flick back to the to our slide around what happens if um someone isn't registered on time. So um for everyone that's already got all their approvals in place. So they've got their audit approval and they've got their Caldecott um they can start on the 30th of September and collect the four weeks between the 30th of September to the 27th of October. So that's every single patient that has an appointment at their local ti a clinic between those two dates. Now, some centers that are lagging behind and haven't quite managed to get the approvals in place. Once your approvals are through, you can then select another prospective four week period and collect the patients during that different four week period. Um This is most mostly because we have so many very, very keen centers that have just turned a little bit late and have been able to get the pupils in place. And the most important thing to us, I think is having as much data and as many sets on board as possible to increase the power of the study. And I mean, we get a way more important output and it'll just be more impressive for everyone and give us some better findings. Lovely in theory, would collaborators be allowed to join a clinic? It might be nice educationally but would likely have an impact on data collection. What, what impact would it have on the data collection, sir? Um Collaborators joining a clinic. Um I'm not sure I fully understand how it would compromise the data collection, presumably if you just meant to a student could sit in on the clinic for like educational purposes. The, the, the way we identify the eligible patient shouldn't vary. All right. I think as long as um, the student or person sitting in didn't have any impact on the clinic or how it runs, not making sure that they collect additional data points. So, yeah, they absolutely would. If a patient of a person who is part of the date study team was joined in a clinic, they should not influence it in any way. Equally. If a doctor is in the dates center team and they run clinics routinely, they should not change their practice in any way whatsoever. And we've made that repeatedly very clear, but that should never be happening in the sense that the work of the clinicians doing the T clinics should not change whatsoever via the participation of their center indicts that was Doctor Schott's uh lead consultant. And yeah, I think it would be nice educationally for them to sit in. Um But yeah, we wouldn't really want to have any impact on the care delivered because that would be um an intervention and that would class as research as well. Wonderful. Fantastic. That was a perfect hour in case we've got any more questions filtering in. We're also available on the Nick gmail, which is Nick Group at gmail.com for any further questions. I think I'm not asking everyone is Welcome to leave. There is a few questions I think that are standing. Um But yeah, thank you so much for for joining. Um This will be available for catch up as well for any collaborators join a little bit later. Um So yeah, thank you. We'll just hang around and answer for some more questions I think. Um Does the data validator have the same deadline as data collectors? Eight weeks from the 30th? the, the data validated will be on a separate timeline to do with when we return the data set that we've validated. Absolutely. Doctors said, I think this will happen. I'm sure I'll ask more questions during the week. I am doing the clinic during the study. Yeah, we had a data point originally around p proteinuria which we um removed cos that wasn't gonna be checked. But it'd be interesting to see if we got some urine dip and if doctors Saltz is still on the foot on the line. At 739 you asked some of the referrals will not even get an appointment. They will be referred by the GP. But the presentation even from the letter will clearly not be ATI N and the referral will be rejected or passed on elsewhere. What to do if those patients pass referrals and what will happen if those ref patients last referrals is they did not have a clinic appointment. Therefore, they are not meeting patient eligibility criteria and you will do nothing with them. And that's part of the reason we should advocate for every center to use the patient appointment list for eligibility as opposed to potentially a much longer and potentially partially irrelevant patient referral list. Thank you. Lovely. Thank you very much, Doctor Schultz. Um Le Prakash. Where can we watch this back? And do we have to log attendance to this meeting? So this can be logged on the progress sheet for Center Leeds. And I believe when this meeting concludes that everyone will be sent out a certificate to say they attended that can go in portfolios. And it's also good for us to keep track of to make sure everyone's watched this. Um and it will be available to watch back, available to watch back on medal to which we will link um on the Google Drive. Wonderful, sorry, not sure if it's been answered, but will the data collection period be extended as the notes are handwritten? So it make some time to be uploaded to the system? So that would be a a suggestion of an example where you the central lead would inform us as a steering committee that there is a delay in accessing the relevant information and that we would accommodate appropriately. Good. Thank you very much. And I think we've still got 100 people on the line, but I think it's important we say is well done for to Dotty for running this study, but also well done. To all the collaborators and central leads who have managed to thus far to proceed very well in terms of setting up their centers. And although we suggest that some centers are lagging, we do know this takes a lot of time and effort of central leads and everyone we've kind of been interacted has been very motivated and working exceptionally hard. So we're, we're very appreci appreciative of that. Absolutely. Thank you very much. Vit Lovely. Um Our data validated is named in the publication like Lia and collaborator. Um Yes. So everyone will be cited by their role. So center leads collaborators, data validators as well. Um And it is a collab a collaborative corporate authorship as previously discussed. So everyone's sort of on the same standing and it's all expandable. So no one gets a spotlight more than anyone else. Mhm OK. Thank you. Great. A five passed. No more questions. Wonderful. We can. So, and we still got ad people on the line. Uh If anyone wants to ask any further questions, please do ask. Yeah. OK. Mhm No. OK. No. Yeah. Wonderful. Thank you so much. Everyone. Grand. Do you want to confirm that the center leads are the ones that choose the data validators? Uh Yes, it is. The center leads role to recruit collaborators and assign a data validator. And we've had some feedback from collaborators who maybe are still in the progress of say recruiting their team in. I think a lot of our personal experience. If you post your email or whatever in a public space, like a medical year students, year group or all the year groups, it is very common to get an overwhelming amount of interest given that this is ultimately quite a finite and not that big amount of work in return for your name on potentially quite potentially multiple academic publications. So I think generally if you're able to promote the opportunity, well, you'll get interest and sometimes too much interest. Thanks. Ok. Ok. Ok. Mhm. Yeah. No, if we close it there dot You do the rest of your email. Yes. Fantastic. All right. Um Thank you so much, everyone. It's gonna post the Nick email in the group. It's just Nick group at gmail.com for any further questions. And yeah, thank you so much for joining. Um send to le please do feel free to ask any quick questions in the whatsapp. Um Yeah. Fantastic. Thank you so much. Thanks a.