[MUSIC] Hi, everybody. Welcome to the first lecture, Introduction to Systematic Reviews. In this lecture, we're going to introduce you what is the systematic review in meta analysis. And I hope you will take away from this lecture that systematic review uses explicit methods to identify, select, appraise and synthesize results from similar but separate studies. Not all systematic reviews would have a meta analysis. And meta-analysis is the statistical method of analyzing a large collection of results from individual studies. By the end of it, I will introduce you the Cochrane Library which is the main product of the Cochrane collaboration and, is the single best place to find independent, high quality evidence for health care decision making. Let me introduce to you the concept of systematic review. And why it's important that we keep track of evidence and information, using systematic reviews. Information about healthcare is everywhere. But how do you know if one healthcare intervention works better than another or if it will do more harm than good? Here, I'm going to give you some examples of a typical clinical question. For example, you might be standing in a drugstore and wondering, are antioxidant supplements effective for preventing mortality in healthy participants like yourself? Do I need to go to annual checkups? Will that reduce illness and mortality? And for women who are in labor, is early epidural as effective and safe and late epidural? These are typical clinical questions you may have in your daily life. But where will you look for information to answer those questions? Well, first of all there's too much information. Information is everywhere. You see them on newspapers, on Internet, from your friends, from your relatives. You may hear from your neighbors they're taking antioxidants. You may wonder why they are doing it and should I take antioxidants as well. But are antioxidants effective at all? And this cartoon illustrates very well the Science News Cycle. The research studies show that there is a correlation, then it gets publicized through different media channels. And your may hear from your grandparents that, well maybe I should wear the hat to prevent A or the units you're interested in. But how do you find trustworthy information, how will you make that healthcare decision? Let's come back to the example of the timing of epidural for women in labor. A Cochrane systematic review has summarized information from over 15,000 women. And those women were randomized to early or late epidural groups. And they found that, when a woman is in labor, the appropriate time to give an epidural during child birth is when she asks for it. So the C-section rate which is the outcome the authors look at does not differ between the two groups, which is the early or the late group. An early epidurals made no difference to the likelihood of needing an assisted birth involving forceps or suction, or to the amount of time spent in the second pushing stage of labor. So we could rely on systematic reviews and find trustworthy answers from these systematic reviews. Here is another example where you may ask should I skip my annual physicals. Probably, that's agood idea. So from a health perspective, the annual physical exam is basically worthless, shown in the Cochran Systematic Review. And this systematic review searched the words biomedical literature on this topic, and analyzed 14 randomized controlled trials with over 182,000 people. And they found that the annual physicals did not reduce morbidity or mortality. Neither the overraw morbidity or mortality or cardiovascular or cancer courses. Although the number of new diagnosis was increased, there are important harmful outcomes that need to be considered as well. For example the number of followup diagnostic procedures or short term, psychological effects. Those are important harm outcomes for annual physical exam. In summary, if we're going to rely on a Cochrane systematic review, then probably skipping your annual physical is not a bad idea. Now we have to look at two systematic reviews and the findings from these two systematic reviews are ready. Let's talk about what is a systematic review. A systematic review focuses on a specific questions. That's were we started. Remember the question on annual physical exam and on the epidural, early versus late epidural for woman in labor. And a systematic review uses explicit, pre-planned scientific methods to identify, select, appraise, and summarize similar but separate studies. So reviews summarize knowledge, and here are the four steps that differentiate a systematic review from a traditional narrative review which is to identify all evidence on the topic, select them, appraise the quality of these evidence and summarize them. Not all review articles are systematic reviews. As I said, only a subset of them uses these three planned scientific methods to summarize information. And the reviews that are not systematic reviews are the traditional narrative reviews. Only a subset of systematic reviews will include meta-analysis, which is statistical methods that people can use to quantify and combine the results from several independent studies. Let's take a look at what a traditional narrative review is and how is it different from a systematic review. Almost all of us started writing about review articles very early on, probably in elementary school. And reviews are important because we cannot read everything and digest them. There are more than 23,000 medical journals and if you wan to keep up to date, you have to read about 90 journals, in full, every day. However, most of us were not taught how to write review articles. So, we have double standards here. Try to remember how you write your review articles. Not your systematic reviews, but the narrative review articles. Probably you're going to the library and look for few papers that support your hypothesis. So, in doing narrative reviews, most people are highlighting and cherry picking what they like and what fit their hypothesis. There is no standard format. There is no clearly specified methods of identifying, selecting, and validating including information and there's really quantitative synthesis to integrate the information from multiple studies. Here is an early study that showed how the traditional narrative reviews are different from systematic reviews and meta analysis. Here the authors look at the medical reviews published in the medical literature from the 1985 to 1986, and then 1996, as well as reviews published on epidemiological topics from the 1997 to 1999. They compared these three types of reviews to meta-analysis, identifying 1996. If we look at the parameters, or the domains, that the authors have evaluated, whether that review has addressed a focused question, described methods for locating evidence, or used explicit criteria to select studies, you will see that meta-analysis which is shown on the last column have a higher likelihood of addressing a focused question and having explicit methods for selecting and locating evidence. So meta analysis, 95% of them has met that criteria, comparing to less than half of all the other reviews. So that shows you a systematic review is really different from a narrative review. It has to focus on a clinical question or a research question and use pre-specified methods to identify all evidence on the topic. Critically appraise the evidence and the synthesized evidence in a coherent way. How do you do a systematic review? First of all, you have to establish your research team and it's important to have content and methods experts on your team. And then you're going to develop your process of gathering stakeholder input; formulating your research question, minimize bias and conflicts of interest. And then as the next step, you would develop your protocol. In the protocol you're going to outline the steps you're going to use for doing a systematic review. And this is very important step, because again it differentiate a systematic review from a narrative review. In other words, we're using a very transparent and replicable process of doing a systematic review. Like you would work on a bench signs or doing an experiment on animals, you will have a protocol. Here, we will have a protocol for doing systematic reviews. After you have your protocol written, you can start your review process by collecting your data, locating the studies, screening the results, and then abstract the data and appraise the risk of bias in the individual studies. After that, which is now on step five, you will synthesize your findings, interpret, and assess the overall body of evidence. Your report writing was summarized everything you have done so far and a good thing about systematic reviews is that it can be updated. Meaning that, let's say a few more studies published after your systematic review is done, a few years later down the road, you could always update your systematic review based on the same protocol that you have started. Again, systematic review is just a way how we summarize evidence. And it is a transparent, reproducible way of summarizing evidence. And the methods could be applied to any field. You probably have heard of evidence based healthcare, evidence based medicine, public health policy. Apply it to your own field. It's not simply oriented to clinical trials or clinical practices, but that said, most methods we're going to learn come from clinical trials. That's where the methods come from. What is Evidence-Based Health Care? Evidence-based health care emphasize three components. It is integration of best research evidence with clinical expertise and patient values. So it's evidence, clinical expertise and patient values and the term was coined in 1991. Why is evidence-based health care important? Let me use the United States as an example. So the healthcare spending in the US, we overspent. According to the World Health Organization, the United States spent more on healthcare per capital than any other country. Of every $1 GDP in the US, $0.18 went to healthcare. And if we look at the data, how are we're overspending health care, you can see that there's a whole bunch of sources for unnecessary spending that range from inefficient services to excess services and administrative cost. So actually about more than half of this overspending can be saved if we use evidence-based healthcare. If we get evidence-based healthcare into play, we can save a lot of money in making the healthcare better and more efficient. You probably have heard of comparative effectiveness research. Comparative Effectiveness Research is a new lingo for evidence based health care. It's taking one step further, emphasizing not comparing an intervention against nothing but compare modible interventions, maybe competing interventions against each other. It was defined as the generation and synthesis of evidence that compares the benefits and the harms of alternative methods to prevent, diagnosis, treat, and monitor a clinical condition or to improve the delivery of care. And the purpose of comparative effect in this research is to assist the consumers, which are patients, clinicians, purchasers, and policy makers to make informed decisions that will improve healthcare at both the individual and population levels. The major founder for comparative effectiveness research in the United States is called the patient center outcome research institute, PCORI. PCORI is a non-profit, non-governmental organization and the congress authorized the establishment of PCORI in the patient protection and affordable care act of 2010. And the PCORI's mandate is to improve the quality and relevance of evidence available to help patients. Again, the caregivers, clinicians, employers, insurers, policy makers to make informed healthcare decision. As you can see, the comparative effect in this research, and what PCORI is interested in funding, and improve the quality and relevance of health care is really the practice of evidence-based medicine and health care. And PCORI found comparative effectiveness research. As well as support work that will improve the methods used to conduct such research. So, how is PCORI different from traditional ways of doing research? I think these summarize it very well. We think it starts by listening to patients. That research agenda is driven by what patients say is important. And patient-centered-outcomes research is putting useful, practical information in the hands of patients and their clinicians. Again, the emphasis on the patient's values, on the clinical expertise and evidence. That ends our discussion on systematic review. Let's move on to our next topic, what is a meta-analysis?
