 Γεια σας! Ευχαριστώ για την παρακολουθή σας! My name is Giorgio Cetidis, I am a PhD candidate at the University of Ioannina and in this talk I will present graphical tools for visualizing the results of network meta-analysis of multi-component interventions. Network meta-analysis is an established statistical method for synthesizing evidence from studies that compare multiple interventions. Networks usually include interventions that consist of multiple and possible iterative components, such as the characteristics of the intervention, the mode of delivery if it is delivered face-to-face or remotely, individually or in groups, the type of provider, the location, etc. These interventions are referred in the literature as multi-component or complex. When they are present, the interest usually lies on the identification of the best component. To answer this question, the component network meta-analysis is commonly used. However, component network meta-analysis assumes auditivity for the component's effects in multi-component interventions. This is a strong assumption and sometimes is hard to defend. It is even harder when the network is sparse and this is because the summary estimates of network meta-analysis are mainly informed by direct evidence and are prone to confounding. The objective of this work was to develop novel ways of visualizing the network meta-analysis results of multi-component interventions in order to help the analyst to identify which components work or do not work more easily. Also, we intended to explore the behavior of the components in order to identify the most promising ones. This was done by associating the presence or the absence of components with efficacy or effectiveness. Due to time constraints, I will only present 5 out of the 8 visualization tools that we have developed. To demonstrate the usage of the suggested tools, we will use an example of a network that compares the effectiveness of self-management interventions on reducing the levels of dedicated hemoglobin. Interventions are masked as these results will be submitted soon to a medical journal and cannot be reviewed beforehand. The network consists of 461 studies and includes 97 nodes and 11 components. Moreover, most comparisons involves node A, which is a control group and it will be used as a reference in the network with analysis model. As you can see from the network plot, there are many nodes, many comparisons, but few head-to-head studies. Therefore, the assumptions of network meta-analysis and component network meta-analysis will be challenged due to the sparsity of the network. Exploring the geometry of the network is essential, especially when you are dealing with large networks with complex structure. Components in descriptive analysis can assist in achieving this goal. By visualizing the components frequency in a colored cross-table, the analyst can easily identify the most frequent components or combination of components in the study arms. Its element denotes the number of study arms where the corresponding component or combination is part of the intervention. Diagonal elements refer to components and in parentheses the proportion of study arms that include the underlying component is presented. Of diagonal elements refer to combination of components and in parentheses the proportion of study arms with both components out of those study arms that include the component in the row is presented. For example, the element that corresponds to column E and row H indicates that 133 study arms include these components, but also that component E is always included in the interventions that include component H. To identify more easily the most frequent combinations, the table is colored based on the relative frequencies. The more it tends the color, the larger the percentage in the parentheses. Therefore, dark red colors indicate large percentages. From this figure you can see that the most dark red colors were observed in the color E. This indicates that component E was almost always part of the intervention. We can also explore whether the inclusion or inclusion of a specific component impacts on the outcome. The live one component out scatter plot identifies a pair of interventions that differ by one specific component. Here is a scatter plot for the set of interventions that differed by component B. AXY displays the NMA effect estimate for the interventions that do not include component B, and AXX when component B is included. Dots close to the line of equality signify no impact on the outcome. Dots above this line indicates that the NMA effect estimates are larger when the component is not included in the intervention. Therefore, for beneficial outcomes in which small values are considered as bad values, dots above the line indicate that the inclusion of a component hampers the intervention effect, while dots below this line signify a component that decreases efficacy. The opposite holds for harmful outcomes. Another feature of the scatter plot is that we can visually evaluate the activity and assumption. Additivity implies that the inclusion or exclusion of a component has the same impact on interventions that differ by one specific component. This is expressed in the scatter plot by a line parallel to the line of equality. In our example we do not observe any parallel line, which indicates that the activity assumption in CNMA may not hold. For a better understanding of the NMA results, it is recommended to apply the live one component out scatter plot for each component. This tool can be also extended to interventions that differ by a specific component combination and adjusted to use it values instead of relative effects. An alternative to the live one component out scatter plot is the waterfall plot, which is also displayed in interventions that differ by one specific component. The differences in the relative effects of the interventions with and without the specific component are displayed in order to explore if the extra component has a positive or negative impact on the outcome. The direction of the impact, positive or negative, depends on the outcome's nature, beneficial or harmful. Another visualization tool that could help on identifying the most promising component is a component heat plot, which visualizes the co-efficacy of the two measures component combinations. More specifically, each element summarizes the NMA estimates where the corresponding component combination was observed. As a summary measure, the media nor the mean can be used. The number of nodes that include the corresponding component combination is also provided in the parentheses of each cell. Letter X is used to highlight any combination of components that are not observed in the network. To identify more easily the most promising components, each element was colored according to the magnitude of each effect. Green color is used to reflect an effect estimate on the desired direction while red color is used to reflect an effect estimate on the opposite direction. In this example, most elements are green indicating that the component reduces the level of glicated hemoglobin compared to the control group. However, we are moderately confident about the results because of the uncertainty in the estimates, which is reflected by the size of the gray boxes. The smaller the box, the more confident we are about finding. Components heat plot can be also adjusted to display Z values instead of relative effects. Similar to the components heat plot is a value plot in which each violin displays a distribution of the NMA relative effects based on the nodes that include the corresponding component. Therefore, by comparing the violins, we can identify if there are components that are associated with larger effects. The precision of the NMA estimates is certainly reflected by the size of the dots. The violin plot can be adjusted to use Z scores and can be extended also to components combinations. Furthermore, by constructing the violins based on the number of components that are included in the nodes, we can explore if the efficacy of the intervention is affected by the number of components. Include, we are a view of contacting both network analysis and component network analysis when multi-component interventions are present. Also, to supplement the component network analysis and strengthen the network network analysis, we can use the proposed tools which can offer valuable insight on identifying patterns between components and exploring their behavior. The implementation of the proposed visualization tools can be performed by using our recently published art packets, namely VSCO, which is available on C-RUN and also on GitHub. More details about the visualization tools can be found on the original arcticty, graphical tools for visualizing the results of network metanalysis of multi-component interventions, published on research indices methods. Lastly, I would like to thank my colleagues for their valuable contribution and my supervisor Dimitris Mavridis. Also, the project was supported by the European Union's Horizon 2020 Research and Innovation Program. So, that's it. Thank you for your attention. I'm happy to answer any questions or comments and of course any ideas or suggestions are model welcome. Bye-bye.