Good afternoon, everyone. Thank you so much for joining us today, and welcome to our research committee webinar, Got a Clinical Practice Question, How to Find Answers in Research Literature. Thank you so much for joining us today. Our presenters today are Kat Scafidi and Kathy Carter-Snell, and I'm going to turn it over to them. If you have any questions during the webinar, please feel free to utilize the Q&A feature, and we will be monitoring that. And also, if you would like to just have a conversation in the chat, the chat box is there and open for you to utilize. So, welcome. Thank you. Wonderful. Thank you for having us, Sarah and the research committee, and I'd like to introduce myself and have Kathy introduce herself and tell you a little bit about ourselves. I practiced as a forensic nurse for eight years, both in sexual assault and as a death investigator in the state of Maryland. I went on to get my PhD, and now I focus primarily on doing research and working with doctoral students to help support them in their research. And my research area is very much focused on forensic nursing practice, specifically around the areas of injury assessment and documentation. Kathy, you want to introduce yourself? Sure. I'm Kathy Carter-Snell from Royal University in Canada. I've been working as an emergency nurse and eventually sexual assault nurse, domestic violence for about, well, sexual assault, domestic violence for about 12 years, but more years than I want to count for the rest. And my area of research, this is how Kat and I connected, is injury as well, but I look at both the psychological and the physical effects of violence and how we can reduce the effect and how we can improve our naming of injuries. And I've also, I'll be talking today about some of my research on injuries with sexual assault and risk factors for them. I teach research at the university as well to undergrads. Wonderful. Well, we're excited to have you all. We know this topic comes up quite a bit related to how do you approach addressing a clinical problem using the literature? It can be very daunting to work with literature. I mean, if you ever just Google anything, you get so many options and it can just be very overwhelming. And not to say that you should be using Google to answer those clinical practice questions. And we'll give you some better ideas on how to approach the literature. But that's our goal today is to help make literature more approachable for you as a clinician. Here are some disclosures from IFN that they want to make sure you all are aware that the planners, presenters, and content reviewers of this course, we have no conflict of interest. And upon attending the course today and completing the course evaluation, you will receive a certificate, that's exciting, that documents the continuing nursing education contact hours. And IFN is accredited as a provider by ANCC's Commission on Accreditation. So the learning objectives for today, we have four. First, we want to help you as a clinician be able to identify a searchable clinical question. And that's really critical in order to help you actually find the literature that will address your problem. You have to start out with the right kind of question. We're also going to help you recognize what the best evidence is to answer that particular clinical question. Because there's a lot of evidence out there. But we want you to focus your attention on the best possible evidence. And we want you to be able to describe how to conduct an actual search of the literature. We're going to break it down for you in steps and hopefully make it approachable for you. And then finally, you'll hopefully be able to summarize why assessing the quality of evidence is a really critical step. It's not just finding the evidence, but you have to be able to look at the evidence and determine, is this a study that I can trust? Or is this a article that I can believe in terms of what they're saying? So that's just as important as actually identifying it. So this should be a review for everybody. But we feel like it's important to remind you all. What is evidence-based practice? That's pretty critical in what forensic nurses do. We should be all striving to be doing evidence-based practice. And as you know, or as you should recall from your nursing education days, it's really a combination of using best evidence combined with your clinical expertise in a given circumstance and taking into consideration patient values and deciding what the best practice is. So you need to go at evidence-based practice, taking into consideration all of these important elements when deciding how to practice. There are a lot of evidence-based practice models out there, but they all pretty much have certain steps to them. And the first step is always to define that question or define the problem. And the key is to try to get as specific as you can in determining what that question or problem is. Then you search for the evidence, search the literature. And we'll talk about different types of literature, studies versus systematic reviews. And then you need to then appraise it, like I was saying in our learning objectives, looking at the quality of the evidence. Then you need to pull it all together. So synthesizing is trying to pull all this different information together in a succinct and a nice summary that is definitely usable. That ultimately answers your question, whether or not there is evidence to support a practice. And if there is, then you choose to adapt it to your particular circumstances, your particular setting. You then implement it. And then ultimately, you evaluate it. You need to determine, is it making a difference? And then the cycle starts again, because maybe it doesn't quite make a difference. So maybe you need to then define the, redefine the problem. Or maybe by implementing it, you realize, you know what, that problem is really different than what we started out with. So we need to go back and look into the literature to find, you know, how to address a different problem or a different solution to this problem. So it's a continuous process, as it should be. So some of the major challenges that nurses often face is that studies do suggest it can take an average of 17 years for research evidence, so studies that I conduct or Kathy conducts, to actually reach clinical practice. That's a long time. And I know this is kind of an old reference, but it's still not much better these days in terms of that time delay. And a clinician, I love this quote, a clinician would have to read 17 to 19 journal articles a day, 365 days a year, to really remain current in their practice. I mean, that is completely overwhelming. I mean, in terms of being able to keep up to speed with the literature, to be current with what is the best evidence for practice, that's a lot to consider. So where do you begin, given these, you know, outrageous numbers? We're going to use a case study to help demonstrate some of these concepts that we're going to talk about in today's presentation. So let's start with a case study now to examine what a possible clinical problem might look like. And I'm sure a lot of you have run into this before, when you're testifying in court about a patient's injuries, and you are asked by an attorney, whether or not it's the prosecution or the defense, whether those injuries are consistent with sexual assault or consensual sex. I mean, this is a very classic question that I think everyone's been asked at some point in their lives. So it's definitely a clinical issue. Clinical problems in and of themselves are often the trigger for evidence-based practice, just like that model suggests. Perhaps you're dealing with a certain clinical outcome for your patients, and clinical outcome for forensic nurses might also be potentially a legal outcome. But also it could be, you know, related to a pain or infections or any number of things. There are also functional outcomes, which might trigger you to maybe look at evidence-based practice. And it could be the function of the patient, or it could be the function of the unit, the unit that you're working in. Quality of life outcomes also trigger evidence-based practice. And finally, economic outcomes. Economic outcomes for a forensic units often struggle economically in how to deal with the, you know, expenses of their program, whether or not it's staffing, supplies, you know, efficiencies, you know, so looking at economic problems as a trigger for evidence-based practice is also quite appropriate. The other thing you could be, you know, another trigger for doing evidence-based practice is also could be just knowledge. I mean, you attend IFN's conference or a presentation, you hear about a study that has a potential impact on an outcome. You think to yourself, hey, what more, I should learn more about this. I wonder if this really could have an effect on patient outcomes. And that prompts you to do a further investigation into the literature about a particular intervention. The key is, what is the evidence linking a particular practice to a specific outcome? And so, it's really important for you to have that as sort of your goal and in terms of being able to search the literature. The first step, well, actually, I should say the second step, you've identified a clinical problem or a quality of life problem or economic problem is to assemble your team. Looking into a search of the literature to address a clinical problem, really, you shouldn't feel that you have to do it alone. Find individuals who are just as passionate as you are about the problem, recognizing that it's important, that it needs to be addressed. If you have individuals that you can tap into in terms of their skill set who have done a review of the literature before, and obviously, sometimes you just need people to volunteer to help you. So, you know, embrace people who are volunteering, even if they don't have the skill set. The idea is to hopefully work with people who are. Consider taking an interprofessional or interdisciplinary approach to your clinical problem. So, for us as forensic nurses, I mean, we work in an interprofessional world. I mean, we're working with prosecutors and police and social workers and advocates and, you know, pharmacy and, you know, lab. And there's so many different professions that we work with. And ultimately, a lot of what we do as a practitioner is has an effect on those different professions. So, consider inviting to your team to address this problem some representatives of that profession so they can give you some further insight. And then also, in addition to interprofessional, think about stakeholders. And stakeholders are the people who are impacted by the problem. And then maybe within your profession or outside your profession, patients are actually important stakeholders as well. So, administrators, you know, people from the finance department in your hospital. I mean, these are all sorts of stakeholders that might also be relevant for addressing your particular clinical problem. Information specialists, if you have any people related to data who work with data in your hospital or in your health care system who can help you manage data or, you know, the other very important person, and I can't emphasize this enough, is your medical librarian. So, if you work in a health care system or in a hospital, inevitably, there's going to be somebody in the role of medical librarian. And they may be not called medical librarian. They may be considered an information specialist. But meet them, tell them about what you're looking for. And inevitably, they will be very interested in helping you. So, definitely become friends with your medical librarian. So, in terms of your topic or your question, so the next step is, so you identified your clinical problem. You've assembled your team. You need to craft that question based on that problem that you've been experiencing. And one of the challenges anybody feels, anybody experiences when they're starting to do a search of literature is they start out too broad. So, let's say, for example, I'm interested in diabetes management. Well, diabetes management is a huge topic. And if I were to just go into a database and search diabetes management, I would get thousands, tens of thousands of articles and a whole host of topics related to diabetes management. It is very, very broad. It's too broad. Ultimately, what you want to do is get very specific in terms of what you're looking for. And that involves crafting a searchable question. The more specific you can get, really, the better your chances are to finding answers to that question. Like I said, having a question that is too broad, like what are effective diabetes management strategies, that's way too broad. It would be like sticking your head into this hay bale, looking for that, you know, pin in the needle in the haystack in order to try to answer your particular question. You want to be specific in order to be able to find that pin in midst this plethora of information out there. So, crafting that searchable question is really key. Now, there are a lot of acronyms out there in terms of how to craft a question. And many of you guys are probably have heard the term PICO before. And I have a T in there. Somebody could call it PICOT or PICOC, or, you know, that kind of thing. These acronyms are very helpful. So, the top acronym has to really do with more quantitative research. And at the bottom, you have PEO or PS. And this has to do with more qualitative research. So, PICOT stands for population, intervention, or exposure, comparison, outcome, and T stands for time. And I'll go over those more specifically in a second. PEO stands for population, exposure, and outcome. And we use this mostly with qualitative research. And I was interested to find Cathy uses PS. So, that stands for population and situation, things to think about when looking at qualitative research. We're going to focus our explanation and example here on quantitative research, because I think that's more what you guys will be working with. But qualitative research is just as important. And I'll remind you guys what the difference is between those in just a minute. But let's break down those acronyms, this particular acronym into usable chunks here so that you understand what we're talking about. So, like I said, P stands for the person or population or problem, okay? So, it could be the age of your patients or the condition or the diagnosis. So, people with sexual assault victims or, you know, pediatric child abuse patients or somebody with a certain diagnosis or that kind of thing. The I stands for intervention or exposure. So, it's a treat, it could be a treatment, a diagnostic test, a certain characteristic. So, maybe not an intervention, but they have a certain condition or a characteristic about them. And then C, which sometimes is there and sometimes it's not, is about a comparison. So, you're comparing something to the I. So, it might just be standard of care, what you're already doing, or some sort of alternate treatment or diagnostic test or what have you. The O is the outcome. So, what is that patient outcome or quality of life outcome, that functional outcome, remember those outcomes I was talking about, that you're looking at in terms of that I having an effect on. And you can have more than one outcome, but usually outcomes are typically measurable. And then T is the timeframe. So, after the I is implemented. So, for example, you know, let's say you're interested in looking at follow-up visits for forensic patients. Maybe you're a program that likes to do a follow-up phone call or a follow-up visit if they have injuries, maybe you do an intervention, such as some sort of reminder communication service or, you know, something along that lines. Maybe you're looking at a timeframe that they follow up within two weeks or they follow up within three weeks. So, that time frame is what you would think about. All of these, it's really important to know, are part of the question and really need to all be considered in terms of how you create that question because they're all important pieces to it, particularly the population and the intervention, the outcome. All three of those are really critical. If you're missing one of those, then your search may be just really too broad or you're not going to be able to really answer it. So let's go through an example. This is a non-forensic example and it's probably pretty basic and a little bit outdated, but you'll get the idea. So central line, here's the problem statement. So central line associated bloodstream infections in the ICU setting. So perhaps you're having an issue with them in your ICU and that's the problem you're trying to address. So what is your P in this particular scenario? So obviously we're dealing with the ICU, so we're talking about critical care patients and they have to obviously have a central venous catheter because we're talking about a central line infections. So that's your P, that's your patient population. What is the I? So maybe you've read, and again, this might be outdated here, that chlorhexidine cloths for bathing may help prevent central line associated bloodstream infections. So that's your I, and you've come upon that in your reading or you've heard a presentation about it, but that's something that you're thinking about exploring. Then your C is, what is the standard of care essentially? So maybe perhaps compared to standard of care soap and water bathing or whatever you all use for bathing your patients. That's the C. And then the O is the rates of central line associated bloodstream infections. So ultimately you want to see if this intervention of chlorhexidine has an effect on your outcome of infection rates. And then let's say you're thinking about this over a three month period of time after implementing it. So this is your problem statement and this is your PICO, and ultimately you craft a question. So you've identified your P, your I, your C, your O, and in this case we have a T. You don't always have to have a T or a C by that way. But here's the question. In critical care patients with central venous catheters, is the use of chlorhexidine cloths for bathing more effective than soap and water at reducing the incidence of central line associated bloodstream infections over a three month period? So here you see we've taken a simple problem statement, a simple problem, and we've crafted a very specific question, which is going to be very helpful because when we find articles or studies, I should say, that actually can answer it, we can actually see that it does actually answer the question. It'll be a lot easier for us to find evidence too that answers the question. So let's go back to that case study. As you recall, the problem we had was during court, attorneys ask whether injuries are consistent with sexual assault or consensual sex. So who's our population? In this particular case, I think we're talking about adult women. And then the intervention in this case is not an intervention, it's an exposure. So it's a characteristic, a diagnosis, what have you. So in this case, we're talking about adult women and the eye, the exposure is that they have experienced sexual assault. The comparison that we're comparing it to is consensual sex. And the outcome we're interested in is rates of injury. And then in terms of assessing for injury, we would expect injuries to be assessed at some point post intercourse. So that's the elements of the PICO derived from our problem statement. So how do we put it all together to make it an actual question? So among adult women, what are the differences in injury rates after sexual assault versus consensual sexual intercourse? So that's our PICO question. And so that helps direct us in terms of our search for the evidence. Real quick, we're going to be searching for evidence. What is the difference between searching and research? By searching for evidence, are we doing research? Well, there are distinctions between these. So when you're searching, you're really looking for something. And you're something that exists, you're going to be looking for, you're actually looking for research to answer this particular question. Research is different. Research, you're really looking into something. So you're trying to determine the truth behind relationships and develop new knowledge that you hope to be applied to populations beyond the sample that you're working with. So there is a distinction between searching for something, which is very critical when it comes to doing evidence-based practice and research, which is also obviously very important. So it's time to search. So what should I search for? This first step is you really need to establish criteria for what you're searching for. So you have a PICO question. Based on that PICO question, you can determine what criteria you're having to identify evidence that help answer that question. Then you have to be able to recognize what are the different sources of evidence, whether it's studies or literature reviews, or I should say study reports or literature reviews. And we're going to talk about the different types of literature reviews. And then there's obviously an important step in identifying, well, what is the best evidence? Because there's so many sources of evidence, it can be a little overwhelming. Well, which one is the best one? Now in terms of establishing criteria, like I said, evidence you identify, whether it's a research report or a systematic review, they must be able to answer that question that you created. So that determines your criteria. How will you recognize whether the evidence does? And that's, like I said, your criteria will establish that. So you need to establish very clearly defined criteria for each of those elements. So what do you mean for that P exactly? A good example of the issue in terms of the criteria for the P is pediatrics. Well, if you're looking at evidence related to a question having to do with a pediatric population, well, pediatric is different in terms of the ages, depending on who you look to. It can be in some definitions all the way up to age 21, but perhaps you're really only interested in up to age 5 or prepubescent. You have to be very specific in terms of what you're looking for. And then you're going to use this criteria ultimately to help you screen evidence. And going back to our example here, here's our PICO, and it's very similar to what we had before in terms of defining what those elements are. In terms of criteria, we're looking for studies or perhaps systematic reviews that looked at patients who were menarchal women, and they were also looking at patients who experienced sexual assault or participants, I should say, that experienced sexual assault and compared it to participants who experienced consensual intercourse. And again, the outcomes would be rates of injury. So if I look at a study or if I look at a systematic review or whatever I define this for my evidence, I would expect each one of those pieces of evidence to meet this criteria. And knowing that they meet this criteria, I know that they answer my question. So what are the sources of evidence that are available? Well, there's a bunch, but generally speaking, we see something called primary sources, which is original research. So these are the research studies that you see published in journals or maybe they're available on the web. But those are called primary sources. We have secondary sources, which those are literature reviews. So reviews or summaries of original research. So they're reporting somebody else's research. So they could be reviewing a single study, but then they could also be reviewing multiple studies. That's a secondary source. A tertiary source is really a condensed summary of material, usually about a specific topic. And this is what we're very familiar with in our textbooks. Those are tertiary sources. Obviously, clinical practice guidelines are essential evidence for nursing practice. And then finally, you know, there are professional opinions and then gray literature. So gray literature is literature that's not been published. And oftentimes we see this as conference proceedings or, you know, articles on the web. That's more of the gray literature. Now, there are different types of research. This is getting back to very basic understanding of types of research here. So this is, I'm sure, a review, but there are different types of research and they fall within three categories. First, there is quantitative research. Quantitative research involves studying something that's measurable that is then examined as in terms of numbers as the outcome. So we're studying numbers for our data versus qualitative research is very interested in terms of data being language. So when you see qualitative research, you're often learning about patients' experiences, you know, based on a certain situation or certain condition. And there's an analysis of the language that's being done. Whereas quantitative, you're analyzing statistically information. Then finally, there's a mixed methods. There are mixed methods studies where they do a combination of both. They do a quantitative analysis of data, and then they do a qualitative analysis of data. And they try to make sense out of both by integrating that information together. So something I forgot to mention, when it comes to qualitative research, their analysis ultimately looks more at themes or developing a theory based on their assessment. And their data collection often looks like focus groups or interviews as well. So now I'm going to turn it over to Kathy Carter-Snell. So she's going to help you understand where do you, where should you start in terms of your evidence search? Okay, so this is really the piece that narrows it down for you. But if you have a good question to get started with, then that will tell you where you should start. So there's all kinds of reviews that you can do to get your literature. Thanks. So there's literature reviews at one end of the spectrum, and there's systematic reviews at the other end of the spectrum. And somewhere in between are the integrative and the narrative reviews. We tend to see narrative reviews more for qualitative research, and you're trying to pull themes across, synthesize themes across the literature. And the integrative reviews, very similar to that. But a literature review, I'll walk you through one of the systematic reviews we're just publishing right now. Literature review, you might pull five or six articles and get a sense of what the common consensus is on it. Whereas a systematic review, you're wanting to know the full spectrum of the literature. So next slide. So the systematic review, it's not just you going willy-nilly into it. You have to follow a particular structure. So there's a structure called PRISMA, which tells you what you have to put in a systematic review. I tend to do Cochrane-style systematic reviews, and it tells you that you have to go through a minimum of X number of electronic sources. You have to go to gray literature. You have to do all of these sorts of things. So my lit review for a clinical problem, I might just quickly look up something in terms of whether betadine affects DNA, but if you swab an area. So you can pull up a few articles, you've done your lit review. A systematic review would be pulling everything that's ever been done in a particular time period. So we systematically identify it, and then we appraise what we see, and then we synthesize it to put it all together. So the synthesis will look different. So when you hear the term meta-analysis and systematic review, they're not interchangeable terms. They're actually the type of analysis you do. Next slide. So just comparing this, a systematic review, you may only have one question for a narrative review, but it's usually a little bit broader. But with the systematic review, we have a specific protocol. It's exhaustive. Narrative review, you're just going to find out what's going on. You select in a systematic review with a very specific criteria, and you're basically hunting on a narrative review. Whereas the appraisal, like to do an appraisal, when we talked about picking your team, for the systematic review, I have a librarian involved, and I have another registered nurse involved, and myself, so that we can look at the injuries as to whether the terms are applicable and the findings are applicable. And then I had a research assistant, who was a student at the time, to help the librarian and myself get through the literature. And so I picked my team, but we were looking for very specific criteria as we went through it, and I'll talk a little bit more about that. The appraisal, if you find a systematic review in there, your search might be done. Or if you find two or three primary research studies in a narrative review, you might see if they all agree on something, and then you're probably done. But the systematic review, we're going at it in a very structured fashion. So then when you synthesize it, I'll show you a comparison a little bit later on in terms of how we put those together. So the inferences that we make on a systematic review, you have some level of confidence if you, systematic reviews can be bad as well, they can be poorly done. As long as you're comparing apples and apples. So if you have similar populations, similar type ways of doing the exam, similar terms for the injuries that you're finding, and that's actually, I started out to do a systematic review on physical and psychological outcomes of sexual assault. And I found that the injury literature was so poorly done, people were using ecchymosis and bruise interchangeably, and they're not the same thing. They were using laceration and penetrating injury the same way. And so the level of confidence I would have in the findings would be pretty poor. Or if somebody's using magnification with colposcopy and somebody else is doing direct visualization, or somebody else is using toluidine blue, can you combine those findings together? So we have to make sure that everybody's looking at the same thing, so there's less variability or what we call heterogeneity. And that increases my confidence in the findings. Whereas a narrative review, you might just, you're on there working with Mary Jane and Bob, and you're not sure whether this is going to be an issue, you pull up your electronic database at work, you find two or three articles, yes, we're on the right track with this. So it's not as evidence-based. Next. So the synthesis of the systematic review could be simply descriptive. For instance, when I was looking at risks for injury, if there's no comparison group, you can't do a statistical analysis. So if you've got people that were sexually assaulted and people who weren't, you could do it. But if I was looking at things like whether they had intoxication or not, I'd have levels of intoxication, but it's not really a comparison group. I could compare old to young and do a statistical analysis of that. But if I'm just stating the risk factors, whether yes or no, then that's what we call a descriptive synthesis. And I'll show you an example of that later on. If we can do a statistical analysis in the systematic review to compare two groups, like sexually assaulted and consensual sex, then the output of that is what we call metasynthesis. And there's actually statistical programs that you use to do that that come up with some pretty neat graphs that you can see the effect quite easily. So that's helpful if you're going into the literature and one study says, yes, they have more injuries with sexual assault. And the next study says, no, they don't have any more injuries with sexual assault than consensual sex. What do you believe? So the metasynthesis will help you statistically determine what the relative risk is of getting injuries in one group versus another. If you can actually do the same sort of synthesis with qualitative studies, but instead of metasynthesis, it's called meta-analysis. Sorry, yeah, metasynthesis is the qualitative, meta-analysis is the quantitative. I'm sorry, I'm very tired. So in qualitative studies, we would say if we have five themes across the literature, they agree on these three themes and they don't agree on those three. So it's still somewhat descriptive, but we're synthesizing the themes. Meta-analysis, we actually have statistical analysis. And that's what I'm going to be showing you shortly. OK, so we also need to look at the hierarchy of evidence, if you go on your search. If you're doing a narrative or a search or a lit review, you'll be running across all of these different levels. And forensic nursing has a number of things that we haven't really studied. So the expert opinion is the foundation. That's where we got most of our initial observations. So we shouldn't discount experts, but if you have expert opinion and you have a randomized controlled trial, I would be more comfortable basing my testimony or basing my clinical practice on the RCT. Now, clearly in forensic nursing, we can't randomize somebody to being sexually assaulted and not. So that's complicated it a lot for us. And so we tend to see a lot more cohort studies where we just look at the population we have that were sexually assaulted and the population that we have that had consensual sex. So that's just the cohort that's available. If you have a systematic review though, you can see that it's at the top of the pyramid. So if I'm faced with having to choose, I'm doing a lit review, I'm on the unit with Mary Jo and Bob, and I just want to find out the best evidence that's available. If I have a recent systematic review that answers my question, that's far more reliable than using a single randomized controlled trial. We talk a lot in grad school about replication because of cohort studies are not as strong as a source of evidence. And we can't do an RCT. If you do multiple cohort studies in different populations and you get similar results, then you start to have more confidence in the findings. And then the next level up is doing a systematic review with a meta-analysis on it. So, but if it's a new topic or a new thing like COVID, we were relying on expert consensus for a lot of the things that were going on with COVID in the last few months, because we don't have enough research on it, but we're starting to get a few comparable studies that we're starting to synthesize together. Go ahead. So where do we find the evidence? So key databases. One of the ones most of us are familiar with is CINAHL, which is Accumulative Index for Nursing and Allied Health Literature, but it is restricted to nursing and allied health. So you may find some medical studies in there, but they may be missing. So I typically combine searches with CINAHL and Medline. Then we've got AMBASE, we've got ProQuest, even got Google Scholar. It's a lot more, a lot less specific or less specificity. You get a lot of unwanted articles in there and some overlap. And then we have synthesized databases. So I'm gonna show you the TRIP database later on, and the Joanna Briggs database is just for nursing, but the TRIP database is for medical studies. So it's a little bit broader. And then the Cochrane Collaboration is restricted to only systematic reviews that have followed the Cochrane style, and they will give you the abstract of what their findings are. So you need to know how deep you're going. So if I was just doing, Mary Jo and Bob and I are on the unit, I might probably just do a quick search with CINAHL and Medline. Or if I want to know the hierarchy of evidence, I'd go to the TRIP database, because that's a very fast way. So my students, when I'm teaching them say, well, why didn't you just start us with the TRIP database? Because that's the easiest. But there's so many questions in forensic nursing that haven't been answered, that we're probably going to have to know how to search individual databases as well. And we might have to synthesize the findings for ourselves, like Cat will talk to you about at the end. So gray literature, as Cat said, is literature that's unpublished. So if you're doing a systematic review, you would include this, but you wouldn't have to in a regular literature review or narrative review. In a systematic review, I'm also looking at reference list, because not everything gets indexed into CINAHL or Medline. So there might be something. So if I find the gold standard article that really has exactly what I want, I'm going to look at their reference list and see if I found those in my search. Then the next level up is to do hand searches. Well, I say hand searches, but now it's a digital one. But if I find that three of my articles came out of a violence and victims journal, then I would go to the violence and victims journal website and look at their table of contents for three to five years and see if there's anything in there that didn't get indexed that would be replicable. The college approach is basically colleague contact. So the college approach is going to people that you know, like if I'm doing injury research, then I should be talking to Cat in terms of what are you doing on this? Do you know of anybody else who's doing something similar or using the conference abstracts to try to contact people, see if they would be willing to share their raw data with me. And as a junior researcher or staff nurse, I was very hesitant to do this, but I found that it's actually, people are really excited to talk to you about it. The fellow who developed the Glasgow Coma Scale in Scotland was delighted to talk to my colleague and myself and was quite elderly at the time, but was willing to share his research with us. Your librarians is essential if you're gonna do a systematic review and they may actually help you filter some of your terms. They have tips and tricks that are amazing. I never thought about what we call exploding a term. So if I put in injury, I would have to put in injuries as well, or injured. Instead of putting all those terms, you can just put I-N-J-U-R and an asterisk and it will look for anything that has that as a root and save you a whole bunch of time. And your professional associations like IFN are very helpful. They might know people who are doing work in the area or have contacts that you could use. So choose the database that's gonna help you. Another piece is knowing your search terms. Each database has separate terms. For instance, in MeSH, there's MeSH, or in the MEDLINE, there's MeSH headings. And it's very different. For instance, we're doing a lot of work on indigenization of our curriculum in nursing and sensitivity to native needs. The word Indian is actually a very derogatory term in Canada. But when I go into MEDLINE, if I want to know anything about native health, I have to put in the term Indian because that's what's used in MeSH. And when you go into the search term, the librarians can show you where to find that. Similarly, if I'm doing sexual assault, we don't have a rape charge in Canada anymore. We have only sexual assault and there's levels of sexual assault. But if most of the literature is coming from the US, then I have to put in rape as a search term as well as sexual assault. So you need to know what your practice terms are and what the search terms in the database are that you're using. And most of them have a little dropdown box that you can put those in. I talked about exploding it. I'll talk to you about when I show you my systematic review results about the differences in sensitivity and specificity. And Kat got at that. The sensitivity is making sure you pick up all the articles that have what you want. And people who name their articles, the titles of articles are not always cooperative. I was looking for something on, obviously on injuries. And one of the articles was titled, toluidine blue effectiveness. I had no idea that that was injuries, but I found it in a broad search. Specificity is meaning that you don't get too many that you don't want. The biggest example I have, as soon as I put in the rape search term in my systematic review, in Canada, we have rapeseed crops, which are also called canola. And I had somewhere in the neighborhood of 1,500 to 2,000 articles that talked about growing rapeseed. So that's obviously not very specific and I don't want those. So there's ways to get those out of your search. So when you do your search, then you look for the highest level of research that's available from that pyramid and the most current. So for instance, if I had a systematic review from 2015, and I had a randomized control trial from 2019, it depends on what the question is. If I'm talking about a medication and I'm looking at say sedatives, the systematic review from 2015 may not include the new sedative that's out. Like when we were first playing around with propofol in ICU, it wasn't available. So I might have to go with the randomized control trial from 2019. But if you've got a relatively recent RCT and a systematic review, the systematic review is the highest level. Sorry. That's okay. So here's an example. So I had to, when I put in my PICO for the population, I had to put in female, women, girls. So the mesh headings exploded all of that for me. And then for the intervention, sexual assault or rape, or non-consensual intercourse. So three terms for that. And those were all from the databases and each database uses slightly different terms. So when you write up these systematic reviews and publish them, you have to give examples of all of the terms that you used to search. And then for the comparison group, it was consensual sex, consensual intercourse. And for the outcome, I exploded the term injuries, but I also included the word trauma. But I got a lot of articles on psychological trauma. So right off the bat, just by looking at the titles, I could eliminate 5,700 of those articles. As I said, about almost 2,000 were rapeseed. And that left me with 325 articles that we needed to actually look at the article. So the other registered nurse and myself went through every single one to see if they met the criteria that we had established. So they had to have a comparison group in our PICO. So there had to be both consensual and non-consensual in the same study. We have lots of literature on non-consensual, on sexual assault and the injuries with that, but no comparison group. And so then we have other body of literature that's consensual, but you can have different examiners, different examination techniques. In one of the systematic reviews that I did, I found that emergency doctors doing the exams identified three times less injuries, genital injuries than the forensic nurses did, but they identified twice as many body injuries as forensic nurses. So it really depends on who's doing it. New nurses are less likely to identify injuries and senior nurses, skin color, all of those factors come into there. So what are your criteria? And so they had to have consensual and non-consensual in the same study, and they had to be reporting injuries, either rates or percentages of injuries, and it had to be either body or genital or both. So when we went through that, we eliminated 313, either because of no injury data or duplicates. Unfortunately, it's not really ethical, but some people will publish the same study in two different ways or two different articles. And so you'd have to sort out whether it's new data or whether it's a different study that the same people did on a different population. We also, one of our criteria was that you had to be able to separate child's data from adolescent data. So they had to be minority, that was one of our criteria. So injury risks and types are quite different for pre-pubital clients. But if they had a table where they said the children, less than 14, these were their injuries, and over 14, this was what they were, then we could still use that. If they had male data in there, that was okay, as long as the female data was represented separately. And if not, so we ended up with only 13 articles. Just, this is an example of searching in CINAHL and showing you what the terms do. So if I just put in sexual assault, and you can see I've exploded the term injury there with the asterisks, and I put in women or female or woman or females, and I searched it, I only ended up with 422 articles. So then if I go to the next one, I put in, in this case, sexual assault, and I added the term rape or sexual violence. And that's the only thing I've changed. I've added rape or sexual violence. And you can see I went from 422 articles to 11,000 articles. So if you're just doing a basic literature review or a narrative review, that's far more than you need. But if you're doing a systematic review, unfortunately, you're probably gonna have to be stuck with some of that, and there's ways to narrow that down to put in the exclude rape seed and things like that, or to exclude males and all sorts of things. So then the next example, this is how some of the descriptive data was reported if I didn't have comparison groups. So when I was just looking at my systematic review of sexual assault injuries only, I had, I separated it by types of examiners. And so then you can see that I have the number of groups or the number of studies that were done. And the nice part about systematic reviews is as long as you're comparing comparable groups and comparable outcomes, instead of having only one study with a sample of 40, now I've got 21 studies and I've pooled the sample size. So I've got a sample size of 7,000. So that's pretty strong. And the rates of injury were 62% had injuries with a median, we usually use medians with this type of review, and you can see the range. So it went from nine to 85% had injuries, depending on the study. So then you, so that's a lot of variability. And then for body injuries, there's a little bit less variability with that, but you can see differences in the rates, whether it was a physician doing it or a SANE doing it. Next slide. So this is the meta-analysis data. So if I took some of the key studies that looked at genital injuries versus, I was looking at consensual versus non-consensual. So it shows here that it's kind of on the line. So our relative risk is 1.25%. So one is the same risk for either side. And in some cases, it looks like there's less risk of injury if you're in a consensual group, and some, it looks like there's quite a bit more. So this is the slaughter article, and there's lots of issues with the slaughter article that would explain why it's so much different than the rest. So what do you do with that? And so there was a study done by, or a report done by Kennedy a few years ago that said you can't do a systematic review on injuries in sexual assault because it's too heterogeneous, too variable, and this is the kind of results you get. But I was sure that there was a way to look at it. So I went on to the next, I did some sub-analysis on it. So one of the things that I found was that redness is one of the key findings in consensual intercourse. So if I include redness in consensual versus non-consensual, it's going to obscure the findings. And redness is also a very subjective finding. You can't really quantify, you can't really measure it as well, and same with swelling, especially when you're looking at photographic evidence, which is a lot of what these injury studies are. So I tried pulling out redness and swelling, and lo and behold, I divided it, the doctors only did direct visualization. They didn't use magnification with their colposcopy. And only SANEs used totoidine or magnification. So I divided them into the two groups, and you can see this little I squared tells you about how variable it is. In the previous slide, I think it was in the neighborhood of 90%, and anything over 80% is too much variability to make it worthwhile even analyzing it. So you can see that when we're using direct visualization, there's significantly more injuries in the sexual assault group than the consensual group. The relative risk is seven times more in the sexual assault group. It's statistically significant, and there's hardly any variability. Anything less than 70% is good. When I looked at the SANE studies, where they were using toidine blue, saw exactly the same thing, 8.45 times the risk. Statistically significant, had a little bit more variability, but it was still within tolerable limits. So that has big implications for us in terms of, not that we shouldn't document redness and swelling. It's been part of the tears mnemonic. It's part of the bald step mnemonic. But we don't know enough about patterns of injury to just lump everything together. So when we're documenting, we need to document each of the types of injuries so that we can see where the patterns are and how things differ by technique. Okay, so, and this is a TRIP database. So if I did the same thing with the TRIP database, you saw the thousands of articles I had with the SINOL, and we'd see the same thing with MEDLINE. TRIP database and Joanna Briggs database have the PICO structure built into them. So I can just put my PICO term in here. So I put women or females, sexual assault or rape, consensual intercourse or consensual sex as they compared her, and I exploded injury. Next slide. And you can see here the beauty of this database is it actually shows you the evidence hierarchy. So this one is guidelines, which some consider higher than systematic reviews because you've taken all of the evidence and built your professional guidelines. This study here is a systematic review. So that's helpful. And next slide. If you scroll down further, you can see that the top study is another systematic review. Then we have primary research, and then we have Australia and New Zealand guidelines. And on the right-hand menu, you can actually narrow it down to only bring you guidelines or guidelines from the USA or systematic reviews or primary research, and it'll narrow the search. So this is a really fast way. If you want something in the clinical setting, and to have somebody that's already synthesized it for you, go to your Joanna Briggs or your TRIP database, and it'll bring it up for you. But I wanted to drag you kicking and screaming through my process of the systematic review, so you understood how rigorous those analyses are, so when you see it on a TRIP or Joanna Briggs database, you know why the systematic review is at the higher end of the pyramid. Next slide. So then documenting it, so I strongly, strongly suggest that you try publishing what you've done, because if you have the question, somebody else will as well, or present it at a conference, but publishing is really the way that we get it out there, and doesn't matter if you're a baccalaureate prepared or graduate prepared, save your search and then document it. You might be asked about it in court, what's your evidence for that? I have people that are asked their expert opinion on injuries, and they just base it on the thousand patients that they've seen, but that could be a really skewed database, so saying that you've done a search on this, and in the literature these are the findings, now ten years from now we find different results, and the conclusion changes, it's not you, it's your experience is invalidated, it's that the research has progressed and has changed, so document your searches and keep that, and try publishing it if you want. Save your articles, if you haven't got a reference manager program, I would strongly recommend it, especially if you're planning to go to grad school, but I find it helpful as a bedside forensic nurse as well. There's a free one called Mendeley, I'm going to show you a couple of screenshots from my end note, which is a purchased program, but Mendeley does exactly the same thing, it just has a little bit, it has some features that the paid ones have more. So next slide, so this is an example on the left side of my, you can see I've got thousands of articles in my database, I've been collecting them since I went to grad school in the 80s, but if I'm doing a search on an electronic database, on the right hand side you can see on most electronic databases or library programs you have, you can save the articles you want to a folder, and then you can download them, and you can see with the bullet there that I'm directly exporting them into my end note, but if you have something like Mendeley, you can directly export them in RISC format, which is a standard literature format, and it brings them right into my database, and so they're there so I don't have to type a whole bunch of things in there, but if I find something else that's not there I can type it in. Next slide, the beauty of this is then when I'm writing the paper, I can just, as I'm writing, I say oh I need that reference, it's that article that I wrote, and so I find Carter Snell, and now I'm looking for something on skin color and bruising, and I know that Kat's written something on it, so I put in insert citation up the top left, then on the right hand side I have a pop-up box, I put in Kat Scafide's name, and it brings up two of her articles there, and I could choose the one I want, it inserts it right into my Word document, adds the reference at the bottom for me, and then later on I decide to take my reference out, or Marilyn Summers reference out, and it changes my reference list for me, or I want to change it to numbered format for publication versus APA format, it'll just do it automatically and I don't have to worry about it. So it's great for documenting and finding it. So in a nutshell, systematic reviews are the strongest evidence if they're done well, and the defense attorneys will often say well can't you have mistakes with it, yeah you can, but look at how they're done and who they're including. If there's not a systematic review, you need to synthesize the information yourself from single studies. So Kat's going to take us through what's needed to appraise and synthesize the single studies. Wonderful, thank you Kathy. And what I love about what she was talking about is really why systematic reviews are so important. She was able to demonstrate for you all the incredible rigor that goes into a systematic review, which is why they're at the top of the pyramid like Kathy was saying. And a librarian would be very helpful with you all in terms of being able to manage this evidence in terms of the reference management software as well. They're very helpful in helping you with that as well. I use Zotero, which is another free software, I was going to mention that. Anyway, but for many of you, you may not be doing a systematic review. You're trying to find the best evidence out there to answer your clinical problem. And like Kathy said, there may not be a systematic review available for you to answer your questions. So then it's up to you to appraise the evidence for yourself in terms of those individual studies. And that's very important and there's there are several elements you should consider when appraising a study. The first is, is the study credible? And that essentially means, do I really trust this study? And some of the things you can look at in terms of reviewing a study is, is it coming from a scholarly source? Most journals, especially ones that you see in databases, are peer-reviewed. And so, you know, they've hopefully done some of the screening for you in terms of the credibility. But you also have to look, you know, who's writing this study? You know, were they being paid by anybody to write the study? So is there any possible conflict of interest? Was the study ethically conducted? All of these factor into whether or not you are, the study is, is credible. Is the study valid? Are the results of the study true? That's what it means to look at results in terms of validity. And, and the things you need to consider for this is, involves the study design. So like Kathy was saying, I mean, for her systematic review, they really need to ensure that, you know, both the comparison and the, the consensual and the non-consensual groups were in the same study, because that allows for that comparison to happen. You can't just take a study that looks at one, and a study that looks at another, and then compare them. They have to be, the design has to allow you to do that comparison. So, and there are some strong study designs and some not so strong study designs. Their randomized controlled trials tend to be the gold standard, but not everybody can do them, and sometimes it's not ethical to do them, just like Kathy was saying. Sampling, how they collect their sample, how they conduct their study, the instruments or the tools they use, are they established tools, or did they just make one up for the study, how they analyzed it, and their conclusions. All these go into whether or not a study has good validity, whether the study is reliable, and that goes into are the results reproducible, and you can see a lot of this in terms of the statistical and clinical significance. So you will often see statistics reported, and there's a lot of focus by some on the p-value and reporting a p-value. Oftentimes you see one less than 0.05 is significant. That might be statistically significant, but something that is even more so important to nurses or for anyone working with patients is, is it clinically significant? Clinically significant is, you really need to look at things like the confidence interval, and just looking at what are they measuring, and what was that amount of improvement or that change that happened, and was that amount of change important when it comes to your patients. And I'll give you an example. As for some of you may know, I do my research with alternate light, and I've studied whether or not you can detect bruises with alternate light. That's more of a yes or no question. Do you see something or not? But I've also been looking at, does it enhance the visibility of the bruise? So let's say you see a bruise a little bit, and then you use an alternate light, and it helps you see it better. Well, if it helps you see a little bit better, well, is that really clinically meaningful? Are you willing as a forensic unit to pay thousands of dollars for a piece of technology that only lets you see something a little bit better? So it's important for me to be able to measure visibility in a valid and reliable way, but also to assess if that improvement is clinically meaningful. So reliability is also very important, and collectively, validity and reliability really shed light on the integrity of the study. Finally, you have to look at a result to see, I mean, that's result, the study itself, to see if it isn't really applicable to your own practice. Who are the participants in this particular study? Are they similar to your patients, or is this a different context? Is, you know, some of you may work in rural settings, and perhaps this particular study is happening in an urban, busy forensic unit with lots of resources, and you have to look to see, is the context and the population that was included in the study, are they the same as yours in terms of your patients? Now, especially when it comes to the validity assessment, a lot of this might be a little overwhelming to you all, because it might be a long time since you've taken a research class and that kind of thing. Fortunately, there are tools available for you to be able to assess, to appraise studies, individual studies, and there are so many out there. I just pulled a couple that are pretty popular amongst, you know, Kathy and myself, and there's even some that have been developed beyond this that combine them, but I'm very familiar with Johns Hopkins Nursing Evidence-Based Practice Model. I've used that in some of the courses that I teach, and there's also Joanna Briggs, which I use currently for my students. SORT also looks at level individual appraisal of studies versus an assessment of the entire body of evidence. ROB2 is what they use, or in ROB1, I guess, in Cochran. Then there's, the CBM is, I believe, in the UK, and, you know, different professional organizations have their own instruments, so the Emergency Nurses Association, they actually have their own, and you can find it in their guidelines to the development of clinical practice guidelines, their guidelines for the development of clinical practice guidelines. They have their own tool published in there as well. The key is, and your librarian might be helpful for you in looking at these, is to find one that you're comfortable with and is applicable for what you're looking at in terms of the evidence. So some tools are good for studies, some tools are good for systematic reviews, some tools are good for a randomized controlled trial, and another one might be good for an observational study, a non-experimental study, or a qualitative study. So just make sure that the tool you pick matches the study or piece of evidence that you're looking at. So I want to show you two examples. So one here is, this is the Johns Hopkins Nursing Evidence-Based Model, and this is just a snapshot. There's a lot of pages, but I find it's very user-friendly, so it gives you questions. So the Johns Hopkins Evidence-Based Nursing Model has two different measures of quality. You have a number, which is the level of evidence, and that has to do with the design and whether or not you're talking about a study, or a review, or an opinion, or a clinical practice guideline. And then it has a letter, and the letter has to do with the quality. And so you just answer questions here. So is this a report of a single research study? And you answer yes or no, and then if it's no, you go to a different question. And then if it's yes, it tries to get at what that study design is without asking you to have to know what the study design is. So it will walk you through some yes or no questions, and then you'll determine your level, and then go from there and assess your quality. The Johns Hopkins one is also supported by Sigma Theta Tau as well. Joanna Briggs Institute is also, I believe, very user-friendly. You choose a tool based on the design of the study, and you just answer some yes or no questions. And sometimes you don't even know what the answer is. And you review it. It's always helpful to review with other people, and this is important. So when you have your team, you should be able to review with somebody else. So each of you take a look at it, review it, come together, and discuss. Because then you can get at least a different perspective and see if somebody picked up on something that you didn't see. So what do you do after you've appraised all the evidence? Finally, finally, it's time to pull it all together by synthesizing it. And it can be a little overwhelming because there's so much information. Maybe you have a stack of, you know, 13 articles like Kathy did, or maybe you have 25 articles. I mean, you have to, it's a lot of information to manage in terms of these articles themselves, and what they're covering, and the data that's in these articles. And the best way to do this in terms of synthesizing is create a table. So use a spreadsheet in Excel or Word or whatever you need. Using, putting together a table, it just makes it so much more manageable. So for each study, you pull out of the study the information, plug it into the table, and then it helps you be able to assess and synthesize the information across the studies. So I teach my students to look at this as far as data collection that's happening in rows. So each one of those rows is a case, and then you look down the columns to synthesize the information together. So we can then see what kind of study designs have been across all these, all these studies. Or what kind of exposures are we talking about? What is the, and I didn't include the sample. Samples should be in here. Samples, are they all the same, or do they vary a little bit? And then finally, what are their findings? And then what is their quality? So a table makes data so much more manageable. And then after you synthesize that information, you really need to go back to your stakeholders. You need to go to your team, you need to talk to the people who are involved, who are impacted by this clinical question. Look at the evidence together, now that you've pulled it all together in a manageable table, and talk about it in terms of how you interpret these findings, how, and the synthesis of the information. And then, and only then, can you really look at answering your search question with the help of your team, and this, and involvement of the stakeholders. So let's go back to that evidence-based practice model that we talked about at the beginning of this presentation. And as you remember, we got to here. So we defined our question, we did our search, we appraised our evidence, and then we synthesized it all together in our table, and we conferred with our team. And then we have to decide, is there sufficient evidence to support this change of practice? Is there enough evidence to support, you know, improving outcomes by going in that particular direction? Now, sometimes the answer is yes, and there is enough evidence. If that's the case, then you keep going and you adapt it into your environment, you implement it, and then you make sure you evaluate it, that whether or not there was this kind of improvement. Now, sometimes though, you'll get to the point where you're evaluating the evidence, and you know what, the studies are flawed, the results are inconsistent, some people found positive findings, some people didn't, you know, or maybe there's just too little research that's been done, but there's not enough evidence to support that change in practice. So really, the answer is, there needs to be more research before you can really go that route. So that's, that's ultimately what you need to do. So in conclusion, you've heard a lot of information about doing literature searches and strategies for doing it, but essentially, solving a clinical problem starts with a searchable question. Having that specific question to help guide your search, the evidence selection, your appraisal, everything ties back to that question, because ultimately, that's what you're trying to answer. The best evidence to answer clinical questions are really systematic reviews, or really good clinical practice guidelines. Use your question to identify your clear criteria for selecting your evidence, the appropriate databases and your search terms, like I said, all goes back to the question. And I'm going to emphasize this again, become friends with a librarian, they're super helpful with this particular step. Or become friends with a, with a school of nursing, you know, seek out the resources that, you know, professors, the faculty and that kind of thing, they have access to a lot of this, and can help you as well. Go to a university and you might develop this wonderful relationship where students and faculty can help be able to find the evidence for you, it'll be a great collaboration. Finally, appraising the quality of the evidence is just as important as finding it, because not all studies are done well, not all systematic reviews are done well. Even randomized controlled trials, some people think, oh, it's a randomized controlled trial, it's great. No, that's not the case all the time. So anyway, that's pretty much it. And here's our contact information. And we would love to hear if you have any questions. We appreciate your time.

clinical practice questions

research literature

PICO framework

evidence sources

primary research studies

literature reviews

clinical practice guidelines

hierarchy of evidence

quality of evidence

synthesizing findings

injury rates

forensic nursing

research committee webinar

evidence-based practice

quantitative research

qualitative research

mixed methods

systematic reviews

cohort studies

CINAHL database