Computer generated transcript

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

Got black. I mm mm mm. That one can use to keep participants either researchers from, you know, being aware of general treatment they are being given at the study period. And one of the major thing planning do is also reduce by us. Uh Because if I I have an idea of what I'm being given and you haing in questions about the effectiveness or how I feel about it, I might be telling you what you want to hear like. Uh so if like in Nigerian context, almost all of us that have taken chloroquine before we see our side effects to Chloroquine is itching. And if this new studies trying to see whether Chloroquine can be useful for. Hm. Mhm. Ok. Mm mm. So mean everybody, the patients can be blinded, the researcher can be blinded, uh the person administering the intervention and the office can be blinded person as assistant can be blinded and the data analyst can be blinded. And um I'll give an example when it comes to each one of them. So when we talk about blinding the participants and that is I'm blinding the patient that I'm going to uh administer an intervention. Mm. The real sense of it because it's easier if I'm giving people. Uh let's say there's a group that I'm not giving anything, I'm giving them placebo, I'm giving them something, but it's not an active ingredient and there's a group that I'm giving the active ingredients. So, uh if a patient know that they are not giving them an active ingredients, if you ask them question about it later, they, it's likely for them to tell you to nothing happens to me. But that's just because they know they are not being given active ingredients, right? So, but if my active ingredients is like uh no mu line and my meaning to, it's like no line. It's clear there's no problem. That is easy. But let's say my ingredients is hello. So I need to find a pink mm placebo. That's what I want to give people. I want to test a drug now. It's difficult to co with injection. Everybody knows what they are taking. So it's either I give, if the drug cannot be converted to either, either formulation. So it's better. I give a placebo of IV plus the tablets that I want to give and give a placebo of, of. Oh, uh oh, oh, oh, oh, oh, oh, oh, oh, oh. Oh. Difficult to. So there are some study that might be. Mm. Mm. Mhm. And the reason why the s so cannot be blind. Good that I'm one is transverse is trans So there's no way you can blind me for that, but there's a way you can blame the person that is going to analyze to. So maybe you code uh the intervention as well. Oh, so had that side effects of, of biases in the sence of it. So when you hear things like open labor trials, that is nobody's blinded and this is usually what happen if it one trials, nobody is blinded. I know the drug you giving me the is the drug they are giving and it is just free for all. We are trying to assess safety of this drug. But we can have single blind, which is only one part of the participant is blinded. That is either the patient is blinded or the the person a miss that the intervention is blinded. One of the common one is double blind in which probably you are blinding the researcher and you are blinding the participant or you are blinding the participant and the data analyst or you blinding the assessor. Let's say that I'm trying to uh check whether uh one to show is better than the harder in outcome. And I've done my surgery, you can blind me. I know the suture that I used. But the person that is going to assess the cosmetic appearance is blinded, he doesn't know which suture that is used. He's just going there to check uh how does the co the appearance of the, of the of the outcome of the surgery looks like that's the way. So you can be double blinded. Another thing is that you can triple blind. You become blind, the research out blind, the assessor blind, the participant and all. Mm Yeah. Be to rest. So I going to do. Mm No. Hm. That like so from the beginning e consideration uh a good clinical trial is stick to cognizant. All ethical considerations. Remember that patient comes first. Remember that uh you should do no harm. Remember that justice should be done. Remember that patient is the center and you should do things appropriately. A good clinical trial has appropriate to design the same. Yeah, I will know if I'm trying, is it cross over design might work for me better. Uh But if I'm trying to do some of that uh trial, a paraly design might be the one that works well. So I need to find what is the appropriate study design so that I don't get hooked in the middle of my trials and I'm not wasting the money of the funders. Another thing is, is this a valid and reliable measures that I'm going to do? So I should not do a trial that is not measurable and we get to do everybody's acting, knows what are we doing now? We don't know and people do that sometimes because it's not well thought out. So there's nothing to measure and you don't have a concrete thing to guide for all these mo are when you are doing a culture, the principle is for you to have it something to measure something that is valid, something that can show that there is a difference you need, I cannot over emphasize the importance of adequate sample size. I didn't talk about sample size cause no of the study but of this lecture. But when we talk about sample size, when we get there, you will see how important it is when you are looking for sample size because it has to be representative of uh of your study population, you have to be able to generalize the results. You have to be to extrapolate from the results. If a study, if I'm going to study ovaries, now, I know I'm only going to study a few years because news don't have ovaries, right? So those are the things that needs to come into play when I'm doing my study design and what I I'm doing my sample size and again, I will have enough size that whatever we are, we are pro it from their result. Uh The good representation of what is happening around the world. Uh We talk about randomization and blinding. It's very important. A poly it, it's not good enough for me to just say I'm doing a randomized control trial, a poorly randomized uh study will not give you good ascomp is as bad as no randomizing. So if I'm going to randomize, I should be done properly. The same thing, we're blinding when I'm blind in patients should be done properly if there should be a room for how to be the blind. So let's say we give a patient a particular drug and we notice that every some set of people are exhibiting a particular side of reactions. It's for the safety of the patients. You can continue a study that you see that people are having terrible reactions, you have to unblind. So there should be a cascade of how to blind properly without affecting the studies. So that should be in place, especially when side effect comes in. You have to have a clear data analysis plan. What do I mean sometimes most of us at our health faces of trying to do researches. I all run into this problem. Uh You did not consult the bi statisticians, you did not consult anybody. It just that and you have collected data now you can't make anything out of data because you did not collect the kind of cover that is needed. You didn't collect the the the do you collect the adequate information that is needed to make an informed decision at this point in time? So it's always good for us to have good data analysis, know your, your independent and dependent variables. What do I need to achieve with them? What kind of uh what would be the best way to analyze if I'm doing? I'm talking about Survivor am I using a uh which kind of tools am I going to calculate this Survivor eventually? So those are the things that needs to be from the beginning. You should be transparent and reporting when we talk about trials, it's very important. You are trying to conducting new drug. You are doing a phase one trial. You have to be as open as a book because this can affect people. If you, if you pass out a drug and say this drug is safe and it is not safe, I will give it to thousands of people and you are killing them. That might be a very serious thing. So it has to be a very transparent thing. And another thing which is not only for clinical trials, it's for all kind of researchers, you have to be able to reproduce it. It's very important. If I do a study in a, if somebody in Canada must be able to look at my methodology and say, oh, we can reproduce this without problem. That's what makes your research a really good one and one that can be skilled. It's a different part of the world. And another thing is appropriate. Follow up. Don't just use patient and discard them in the middle. You should be able to follow them up. You should have a long term plans for the people that were you study? How long do you, do you think the effect of whatever you have given will last for and all that. So it's very important and explicit and this will make you be able to do a very good uh clinical trials. Thank you so much. In case of any questions, you can always uh all these questions to my email address which I provided and I will be ready to take your questions anytime. Thank you so much.