Speaker 1: If you have to write a systematic review and you don't know where to start and you are completely stuck then this video is going to be for you. I'm going to be going through the seven steps for writing a systematic review including the research, the writing process, the analysing process and I'm also going to be sharing with you a template that is an ebook firstly with all the steps and secondly a template, a blank template for writing a systematic review with a breakdown of all the different sections and chapters that are going to be included within a Word document, so all you need to do is add your information within it. I've developed both of those things over the past month, I've been doing a ton of research to find the best resources and links for you and we're going to be going through all in this video. If you want to download that template, the template link is down below so feel free to go and click on that down below as well. So let's get straight into it. So there are seven main steps that I've identified for writing a systematic review and let's go through it together. So the first one is to define the research question, the question that you are actually wanting to answer within this systematic review of looking at the literature and bringing all those papers together. The next one is to develop up the review protocol. So this is exactly what the method is for how you are going to be searching through the different databases to find the papers and the literature and the research that is going to help answer your question. So this could include things like your data extraction, the limitations, the exclusion, the inclusion criteria, and everything like that. The third step is to actually start and conduct the search. So this is where you take that criteria, take that search method and you input that into different databases to find all the papers, the relevant papers that you could possibly use for your research and your systematic review. The fourth step is to screen the studies, so here you are looking at what you've pulled out and you're saying whether they are actually appropriate and whether they're actually relevant for your review. Then you want to assess the quality of the papers using specific checklists. Then you want extract data and synthesize and include this into some sort of like graph, table and flowchart which shows exactly what you found and what the results are from your review and lastly is to present your findings. So these are the seven steps I've identified through like I said loads of research, loads of reading and kind of looking at different systematic reviews and how they've been able to gather that information together. So step one is to define your research question and to do this there are different methods but for me I found that the one of the most useful and easily broken down tools is by using this framework called the PICO framework and let me just zoom in a bit. So the PICO framework is based on the idea of population, intervention, comparison and outcome like let's look at an example for example and this is essentially saying that the population here is for example patients with type 2 diabetes these are the group of people that you are looking at that is your population. Your intervention is looking at exercise programs so what an exercise does on a particular characteristic or factor. The comparison is with no exercise or just their usual behavior so that's what you're comparing it with. The outcome is the improvement in glycemic control so with patients that have type 2 diabetes if you were to implement or add some exercise into their lifestyle does it improve their glycemic control and that is your PICO. So that is essentially your research question. Now that is for a quantitative research and in my ebook and my guide I kind of touch upon both of them so if you're doing quantitative or qualitative research this will help you for both. If you're looking at qualitative research then it changes slightly and it becomes PICO but the O is just a small O because it stands for context. So population, here's an example, population is children with autism, ASD, so that's the population of people that I'm looking at. The interest is saying if a parent implements certain interventions, what impact does it have? And the context is looking at early intervention settings. So I'm saying if a parent was to implement certain things at an early stage, does it have any sort of impact or change in children with autism, ASD, and that's a research question. So this is a really nice way of just developing your research question. In fact a lot of universities will say we want to see your PICO framework. So what I've also done on top of this is I've also developed a PICO sort of worksheet and I've got the same thing for the qualitative as well, a bit different. So I've given you a framework where you are able to input that information and including what kind of question you're looking at, what kind of search strategy you're going to be looking at, your search terms, your eligibility criteria etc and that's all going to be included within the template. Step number two is to develop your review protocol and that's essentially the strategy that you're going to take during the search process. The number one thing you want to do is identify your keywords, your terms, your synonyms, what you are actually going to be inputting into the search bar. This is so important. If you're looking at autism and you're not putting in the specific ASD, the specific word, or different types of autism, different groups of people, maybe adult autism and not child autism, what is it that you are looking at? What is the specific keywords that you want to pull out in order to get the best possible papers out of your search strategy? And as you can see, what I've also done is I've left links for certain websites and resources that I don't need to do all over again because they exist and they're amazing like this Prisma checklist which is really good for ensuring that you have like all of the items and all the parts of a systematic review so I've left that here for you to click on and download for yourself. So there are some ways of getting these keywords, you can brainstorm, you can use a thesis, you can use search engines, you can even use this website called Word Freak, which allows you to look through the different papers within your library. So let's say you downloaded ten papers that are perfect and excellent for your research, then you can include that library into this tool and it pulls out the most commonly used words from that library. So cool. And that's essentially one of the things that you want to do. So I also have included some information about search limits, so you want to, when you're doing the search you might want to say I'm only interested in a certain date, I'm only interested in English, I'm only interested in adults, females, males, whatever it is and you have these limits and when you do include any limits make sure you're writing that down because then when you're having when you're doing that search strategy you're saying that I've pulled out a hundred papers but what are the limits that you placed to pull out those papers? Any limit you placed, you have to, you absolutely have to write down what those limits are. I also found this really cool tool that I think is fantastic and I'm so pleased that I did find it. So it's essentially called a methods wizard and I think it's like some sort of AI tool. So you can see here that if you go to a new project, by the way they've got a ton of different things for systematic reviews, do take a look at it. So you can give a title for your study and say when it's gonna be published and stuff. So here you can see for example I've got my methods wizard so I've given the title of my project, my name etc and then I can go through and add in the detail for all the different parts of the methods. So I can say what study selection I'm going to be looking at, how many reviewers, how many authors I have, I can say how I'm extracting the data, I can say how I'm analysing it, what my criteria is for the eligibility. So I can say I want to include this, exclude this, I can add some new inclusions or exclusions and then once you've added all of this information in, it adds it all into this database and then you can generate the output and it has the whole methods section done for you just by adding all of this information in. Really, really cool. Highly recommend checking it out. Okay, so that's number step number two. Step number three is actually conducting the search. So here you want to do it by three main main approaches. One is line by line, so you're just including each search term on a separate line, for example depression, adolescence, cognitive behavioral therapy, line by line. The second way is block by block, so every search concept is given its own kind of part, so population is adolescence, intervention is this, outcome is this, and you're giving a different like kind of section per term. Or single line, so each search term is given a kind of combined into one single line, but you're using Boolean operators, so like and or the dots, the colons, etc. So, for example, you might want to say something like you want to search for depression or depressive disorder or mood disorder or affective disorder, and you also want it to include adolescence, teen or youth, and it also has to include these things. So that means that that any paper that you pull out includes these words and includes one of these words and includes one of these words because it has or so it has to be one of those in each of those brackets and it is more challenging because you do need to make sure that you've included all the terms that you want to search for but it does pull out papers that are a lot better so if you are including searches like this in your search process then do make sure that you are recording what you are specifically putting into that search bar. Then you want to test and refine the search So, you know, you've done your search a few times, you're not really pulling out results that are that important, or maybe you're pulling out results that are, some are okay, some are not that great. So you want to refine the search and think about how you can make the search a bit better. So here are some questions that you can ask yourself. Are you using the right words? Are you spelling correctly? You might think you are, but you might not be. Are you being really comprehensive? Have you combined your terms correctly? And also, are the results relevant, ones that you're pulling out, are they actually any good? There's also a checklist that you can use that is developed by somebody else to review the strategy and make sure that you're kind of gaining the best possible results and you can make some adjustments and kind of go back. So you've done the search, you get some results, they may not be that good, you go back and do the search again. You kind of go back to that process till you pull out all the possible papers that are as relevant as possible to you. Here are some examples for different databases that you might want to use. it really depends on your specifications. If you're doing a search on something dentistry related or medical related you're probably going to use PubMed, Cochrane Library, which other one is a popular one? Medline is one that's used particularly by sort of like the medics and dentists and so if you're someone that's maybe in a different topic, a different field, then you need to use the appropriate ones for you. So not every database is going to be relevant for you. Don't forget about grey literature. So grey literature is literature that is not published but also could be important. So this could be things like a journal, it could be a conference proceedings, books, journals that are not peer-reviewed or papers that are not peer-reviewed, government reports, things like that. And these are important to include if they are relevant and provide relevant information for you but also you do need to assess them as well. So here are some factors to consider. You might want to think about what kind of publishing it is, where it's been published, what their methods were, how relevant it actually is to you as well. And these are things that you want to think about when deciding on whether that grey literature is something you want to include within your systematic review. The fourth step is to screen your results for eligibility. So how good are these results for answering your research question? There is a flow diagram called a Prisma diagram and actually if I open it up right now you'll see that it's actually a really good tool but you do need to select the one that is most relevant to you so there's a flow diagram for those that are including searches of databases and registers only those that are databases registered and other sources or you also have this app called the Shiny app and you can develop the flow diagram the Prisma flow diagram from there too so it really it's up to you, but that is a tool where you can generate the systematic review flow diagram from here. And this is going to be evidence of your search process. If you zoom in you clearly see that it says something like the identification, here's how many studies were included in the previous work, here's how many studies from this database, from this register, how many were excluded, how many were not retrieved, how many were removed, duplicates. So this is the version that includes databases and registers, so choose the version that is suitable to you based on your search strategy. Step number five, you want to analyse the quality of your studies. So this is where you look at how good those papers actually are, so it looks at the strengths, the weaknesses and any biases that may be implicated within that particular research. There's this tool that again I think is so cool and it allows you to look at your risk bias, so you upload all your data and it generates a sort of review like this where it says like this study won the domain you choose what domains you are interested in it's biased because of this it's not biased because of that and it's quite a nice diagram that allows you to see sort of overall like how many of your papers and how many of your studies could be biased and I think this is quite a nice one that you can include in your systematic review because most people won't know about these kind of tools and it'll look really like novel and unique. You can also use this checklist called a CASP checklist and it includes different, eight different appraisal tools which investigate if the results are valid, what are the results and will they actually help locally as well. The next step, step number six and we're almost at the end, is to extract the data and to synthesize your data. So you've got all these papers, you've pulled out all these results, now you want to synthesize the data and extract it. And this will allow you to develop and discover different themes and patterns and relationships that are kind of formed across the different studies. So if you are someone who's doing a qualitative systematic review, you might want to think about themes, something thematic, the themes that are running through the different studies, that would be something that you want to pull out at this stage. And the main difference between qualitative and quantitative is that with quantitative you're looking at more stats and like data and numbers whereas qualitative is looking at sort of non-numerical data and thematic and kind of understanding and sort of synthesis in that in that sense rather than sort of the numbers and like looking at populations and things like that. For the meta-analysis for a quantitative paper you might want to represent your data through a forest plot and there's not enough time in this video to go through that kind of thing so I've included two videos that I've watched and I think are really helpful for how to construct it and how to interpret it and so do watch these videos to be able to do that. Whereas for qualitative you want to think about the themes as I mentioned so you want to think about generating sort of codes and labeling your ideas and then thinking about themes and then like reflecting on them and reviewing them to make sure that they reflect and accurately reflect the data. And then last but not least you have to present your results and present your systematic review and there are sort of eight to nine sections of this title, abstract, intro, methods, results, discussion, conclusion, references and appendix and this is where you would include all that information that you've just discussed and searched and thought about and within a report that summarizes the hypothesis and what your findings actually are. Now I actually have a template for this so let me run through that as well. Okay so this is the systematic review blank template and I've gone through, I've kind of left some comments on the side here for you to read adding to the ebook as well. So this is the contents page, these are the different sections that you have to include within your systematic review and when you finish writing you just want to right click on this and sort of say update field and it'll update all the page numbers to match what you've actually written in the end. And yeah, abbreviations page and a lot of acknowledgments page, declaration and abstract. Then you've got your introduction, so here's what you want to include in your introduction. Methods and materials section, results, so here you can actually edit this if you want to use this one. More results, discussion, list of tables, conclusions, references and appendix. So yeah, I think it's quite concise and the systematic review process is not an easy one for sure. You're doing a lot of searching, it's quite monotonous, you're kind of trying to find and pull out the most important papers, the most relevant papers that will answer your particular research question. It's definitely not easy and it's one that if done well, it's done very well because you've been able to pull out the right keywords and if not done well you will miss out on quite important results which mean that when you're kind of reporting your results you're missing out important key bits of information in the story so it can lead to incorrect reporting on your behalf as well. So it is you know really important for you to be careful and be aware of sort of how you're doing and how you're justifying everything and keeping track of all the steps. So yeah if you do want to get your hands on this template it was a lot of blood sweat and tears and learning and unlearning and speaking and everything so you know it's a lot of hard work on my part that has gone into putting this template together so if you do want to support me and and grab that and help yourself too then I'd really appreciate that the link for it will be down below as always and if you do want to see more like this if you want to see other templates as well let me know and I'll see you guys in my next video, bye.