[Music] with so many research studies available to us in so little time in our days learning how to interpret the results of the meta analysis is a very valuable skill for public health professionals a meta analysis is a single document that summarizes findings from a collection of relevant studies and provides a more accurate estimate of intervention effectiveness in comparison to using a single study a meta-analysis finds all the studies on a particular topic and aggregates the findings into one numerical answer usually depicted in a forest plot or blobby gram let's take a look at interpreting a forest plot using a hypothetical example of a media campaign to promote wearing bicycle helmets a forest plot is constructed using two perpendicular lines the horizontal line represents the outcome measure in this case the odds of wearing bicycle helmets the vertical line called the line of no difference intersects the horizontal line and represents where the intervention had no effect during the process of conducting the meta-analysis the researchers read single studies about the effectiveness of interventions to increase the wearing of bicycle helmets they plot the results of each study on a forest plot the results of studies reporting an increase in the odds of wearing a helmet appear to the right of the line of no difference the results of studies that show a decrease in the odds of wearing a helmet appear to the left of the line of no difference the results of studies where there is no difference in the odds of wearing a helmet between the intervention and control groups will appear on the line of no difference how far away the result for a particular study appears from the line of no difference represents how large the effect is the further away the larger the effect results of studies showing a small effect get plotted closer to the line of no difference keep in mind that the heights don't matter you're looking to see how far the point is from the line of no difference to see how effective the interventionists so now that the findings from all the different studies have been plotted we see that some interventions worked to varying degrees one had no effect and one had a negative effect meaning those exposed to the intervention were less likely to wear helmets than those not exposed the next step is to combine the results which is depicted as a diamond at the bottom of the forest plot a diamond takes into account the results from all of the studies and factors in how effective each study was so you can quickly and easily see a graphical overview of the results and determine the overall effect of the intervention let's take a look the left and right points of the diamond represent the range of possible effects of the intervention it could have an effect as little as this or as much as this but we know that the true population effect lies somewhere within this range based on where the diamond is situated on the forest plot and seeing how far from the line of no difference it lies we can determine whether those exposed to an intervention we're better off than those not exposed and by how much we can also determine if our finding is statistically significant by looking to see if any part of the diamond touches the line of no difference if it doesn't touch then we know we have a statistically significant finding in our scenario where we are looking at the effectiveness of interventions to increase bicycle helmet wearing we can conclude based on where the diamond is situated on the forest plot that those exposed to the intervention had a greater odds of wearing bicycle helmets than those not exposed to the intervention meaning the intervention appears to have had a positive effect we also know that our finding is statistically significant because no part of the diamond touches the line of no difference however since the left side of the diamond is quite close to the line of no difference this tells us that the size of the effect on helmet wearing could be quite small decision makers would need to decide if this relatively small impact on helmet wearing justifies the resources required to implement such an intervention there is more information on this concept in our video on clinical significance please take a look for further understanding plot points on a forest plot will vary in appearance taking more complex concepts into account we've kept things simple in this video because you can understand the overall findings of meta analyses without understanding all the complexities as long as you can interpret where the placement of the diamond is in relation to the line of no difference you will quickly be able to determine the overall effectiveness of the intervention you are studying you can see which interventions are working and which ones aren't so you can make informed decisions about the provision of public health services [Music] [Applause] you