Hello, My name is Ann Ervin. I am faculty in the Department of Epidemiology at the Johns Hopkins Bloomberg School of Public Health. In this module, we will discuss methods for synthesizing evidence from individual but similar clinical trials, addressing a clinical trial question of interests. Systematic reviews focus on a specific question and systematic reviewers use explicit, transparent, and replicable scientific methods to identify, select, appraise, and summarize similar but separate studies. Meta-analysis is an optional component of systematic reviews that allow quantitative assessments of outcomes from similar trials. We will discuss ways to use information from individual but similar studies to assess the evidence base for a clinical question of interest. We rely a lot on our colleagues, friends, other students, the news, and even social media to keep up to date on information regarding health care. The medical literature is also a source that we often refer to. There are a number of different scenarios to consider when reviewing information for a clinical question of interest. The first is that, our information needs are not really being met because there are not enough data to answer our specific clinical question. Sometimes there's just too much information. Sometimes the information is a poor quality or it's even disorganized and not a way in which we can interpret it in a meaningful fashion. There are all types of evidence that we may see in the clinical literature. There are case reports, there are clinical studies including clinical trials which we're highlighting today, laboratory science. But really how do you know if one intervention works better than another? Or if it will do more harm than good with the specific isolated studies. It's very important to look at what the totality of the evidence is and identify a way to summarize. In the literature, you'll often see what we call traditional narrative summaries. These are often seen and may not represent what we call the totality of the evidence. There are usually is no standard format. There are no clearly specified methods of identifying, selecting, and validating the included information and another thing what we might expect to see on a quantitative synthesis, where we take individual estimates from the studies and integrate them in a way in which might be meaningful; you don't often see those in your traditional narrative summary. I think starting off, we should talk about this idea of what we call Knowledge translation. How do we go from having individual studies and conducting individual studies to moving on to clinical policies and even beyond. On this slide, we show a series of steps from clinical research to practice decision. We start off on the far left with clinical trials and observational studies so many of us conduct these individual trials so we're generating evidence. The next step we hope is that these individual trials and other studies are synthesized in a manner in which we call systematic reviews. Systematic reviews are often used by experts in the field to develop clinical policy or clinical guidelines. These are often guidelines that are used in regular clinical practice. But here on the far right we see in our flied the evidence-based health care. The clinical policy and guidelines which are generated from our individual clinical trials and are systematic reviews are applied in a way in which we integrate the best available evidence, clinician's expertise, as well as patient values. This is what we call knowledge translation. We're translating from individual trials to synthesis of these data, to guidelines, and then onto the practice of evidence-based health care. What is our big first question? The first question is really, what is a systematic review? We've talked about what is a traditional narrative review or a narrative summary that you might see in the literature. How does this differ from what we call a systematic review? We will now review the features of a systematic review and the benefits of incorporating systematic reviews to answer important clinical questions. Systematic reviews or do reviews in general summarize knowledge and systematic reviews are a subset of all reviews. It is a scientific investigation that focuses on a specific question and uses explicit, pre-planned or pre-specified scientific methods to identify, select, appraise, and summarize similar but separate studies. When appropriate systematic reviews may also include a quantitative synthesis of the results from separate studies, and this is what we call a meta-analyses. Systematic reviews include pre-specified eligibility criteria. You will also pre-specify in a systematic review methods for identifying and selecting eligible studies and the methods for synthesizing the results. If results are able to be combined in a quantitative synthesis or whether or not your result of your systematic review will be a qualitative or a narrative summary of the included study. Then finally, there are methods for evaluating the risk of bias of the evidence and we will talk more about what risk of biases and how that is evaluated in individual studies in the systematic review. Now, just to sum up, you see the image that we have on the right-hand side. All reviews will include reviews that are systematic and that are not systematic. Those are traditional or narrative reviews. Systematic reviews are a subset of those and even more so of a subset would be our systematic reviews with meta-analysis. Because not always are the data that you acquire from your individual studies are amenable to quantitative synthesis. You may see a subset of systematic reviews that also include additional information about the effect estimates across several similar studies. Why should you conduct a systematic review? What are the actual benefits of going through this process? Well, one is that it provides a very clear and transparent process of developing a systematic review. Everything is pre-specified at the beginning before you start your systematic review. It minimizes bias or any chance there might be systematic errors in the process. We want to reduce any chance effects as well. There may be a quantitative synthesis or a meta-analysis, and that enhances your ability to provide very precise estimates across individual studies. You can provide a summary estimate of all the effects from individual studies, as opposed to just talking about the results from an individual study. Another advantage is because you pre-specify your methods. This can be readily updated so often the evidence-based changes over time. You may want to update or renew your systematic review and add additional studies and this can be done very easily. But I think what is the most important part or important benefit of a systematic review, is being able to facilitate the efficient integration of what we call information for rational decision-making. As we talked about in the idea of knowledge translation, we can use these systematic reviews to see where the effects of health care are really consistent, and whether or not they vary. Studies may have similar results or similar finding or we may find that with similar questions, different studies have different results with different outcomes, and we would have the opportunity to explore why that is the case. What we want to be able to do is in the practice of evidence-based health care and even in just the understanding of a clinical question of interests, we want to make sure that we're making decisions that are based on the totality of the available evidence. Systematic reviews are also required by funders prior to submission of an application for a clinical trial. That's an important piece of information to note, that systematic reviews may often be something that investigators are completing before preparing an application for a new clinical trial. Who conducts these systematic reviews? There are so many reviews that are out in the literature that are now systematic reviews, and independent authors are really the majority producers of systematic reviews. I think that comes from a number of different sources. There are a number of clinical researchers who are very interested in summarizing the evidence in a very systematic way for their clinical question of interest. There are investigators who are interested in proposing a new trial so it is very important to consider the outcomes and the other studies that have been conducted to formulate important questions and to make sure that their clinical trial is informed and appropriate and ethical as well. There are other groups that conduct systematic reviews, including the Cochrane Collaboration. The Cochrane Collaboration is actually a group of volunteers and others who prepare systematic reviews across different subspecialty areas in medicine and other issues as well. They're very focused on producing protocols of systematic reviews as well as full systematic reviews on various topics of interest. There are also other groups that are interested in policy-making, such as professional societies, governments, and also payers and you see here a listing of a few that are existing in the United States and the United Kingdom, as well as in Germany. There are a lot of groups that are interested in policy-making and as a part of their activity, conduct systematic reviews to help inform policy and other guidelines.
