What criteria can you use to choose studies for meta-analyses?

One thing to consider is how much we want to provide feedback to AI tools designed to replace the work we do haha. I'm not overly concerned about losing my position to AI because meta-analysis is a small aspect of my job but as we provide this type of feedback, we're fine tuning the model which is great to share information and at this point it's too late to turn back but it's important as we go through this 4th industrial revolution without creating a power dynamic of individual contributors out of work and a select few using AI and Data Centers to control the economy. Borderline conspiracy theory thinking but it seems pretty likely that this is a potential path/challenge we're going to have to navigate in this new technological era.

To facilitate the jump to the meta-level, I often ask students to analyze research statements by distinguishing between "actions" and "reactions". For example, I invite them to ask: Does this statement (presumably) react to something that already exists (point of view, result, etc.)? How is the existing treated, summarized, addressed, etc.? Is a statement presented as a formulation of something new? Is it really new?

The first question is why we want to provide a meta-analysis study. This type of research usually efforts to follow orientations, build a big picture, find the gaps, make an agenda, etc. If you are sure meta-analysis is the best method you can apply, then you should determine your specific purpose. I rather choose a clear and limited subject against a great scope.

Developing a robust meta-analysis starts with a clear research question and protocol. Define objectives, identify key variables, and set explicit inclusion/exclusion criteria. Draft a detailed protocol, minimizing bias and ensuring consistency. Conduct a systematic literature search, involving multiple reviewers for reliability. Systematically extract and assess data quality. Finally, perform a quantitative or qualitative synthesis for transparency and validity. Documenting the process enhances reproducibility and allows validation by others, ensuring a rigorous and unbiased meta-analysis.

Relevance - Studies must directly relate to the predefined review question and target population with the outcomes of interest clearly defined. Cast a broad initial net when screening studies to ensure appropriate depth and breadth. Common reasons to exclude at this phase are irrelevant outcomes or patient groups. Study Design - The research methodology of included primary studies should be carefully specified based on the aims of the meta-analysis. For interventions, randomized controlled trials (RCTs) provide highest quality causal evidence, though well-designed nonrandomized and observational analyses may add value contextually. Surveying all study designs initially is recommended.

Crafting a thorough search strategy is paramount in the research process. Utilizing diverse sources and databases ensures a comprehensive exploration of relevant studies across disciplines. The inclusion of precise keywords, synonyms, and Boolean operators enhances the precision of the search, refining results to meet specific criteria. Documenting and reporting the strategy meticulously not only facilitates the transparency of the research process but also allows for replication and verification by fellow researchers, fostering a foundation of reliability and credibility in academic inquiry.

And don't forget to check the references of the articles you find and add to our list. This 'snowball' strategy can lead you to a more detailed study in a specific perspective. Remember to always work on two 'axes': the 'Width': which defines the different perspectives, methods, research orientations on a specific topic, and the 'Depth', which defines the works that exist in a specific orientation. A good meta-analysis must include, discuss and compare both of these aspects.

Checking the meta-analysis studies or systematic reviews shows that researchers usually select some databases like Web of Science, Scopus, Gale, etc. Mostly, they don't explain why they chose some of these databases rather than others. we should select fit and related databases on our issue to ensure we find all potential studies.

The subsequent crucial step involves crafting and implementing a meticulous search strategy to unearth studies aligning with predetermined inclusion and exclusion criteria. Leveraging diverse sources like PubMed, Web of Science, Scopus, Cochrane Library, and Google Scholar ensures a comprehensive exploration across various disciplines. Employing a range of relevant keywords, synonyms, and Boolean operators like AND, OR, and NOT refines the search, enhancing precision. Documenting this strategy meticulously is paramount; a transparent account facilitates replication and verification by fellow researchers, ensuring the reliability and reproducibility of the meta-analysis.

Its also imperative for the search strategy to include a record system of the screened and selected material sources for referencing purposes.

Once you've compiled a list of potential studies from your search, the critical phase of screening and selection begins. Applying inclusion and exclusion criteria to titles, abstracts, and full texts is key. Thoroughly checking references and citations expands the scope for relevant studies. Streamline this process with tools like Covidence, Rayyan, or DistillerSR, automating and facilitating screening. Involving two independent reviewers minimizes errors and disagreements. This meticulous approach ensures that only the most eligible studies are included in your meta-analysis, enhancing the precision and reliability of your findings.

Este aspecto es de vital importancia, no es solo obtener la información, si no que esta sea de calidad, contenga datos fiables que estén corroborados así como sus fuentes para poder utilizarlos en la investigación que se pretende realizar.

Meta analysis is when an overview incorporates a specific statistical strategy for assembling the results of several studies into a single estimate. In the development of a review protocol: we follow Formulating review questions Searching and selecting studies Study quality assessment Extracting data from studies and Data synthesis

Methodological consistency and language similarity are also advisable! This makes the analysis vivid and leads to a strong conclusion.

Es importante ser meticuloso y metódico con los datos que se extraen, una hija de cálculo permite organizar toda la información que se requiere y para tener un check list de las características que faltan sobre un dato o una estadística determinada.

1️⃣ I foremost consider recent (new) studies for relevance and timing. I also accommodate older studies on case by case by case basis. 2️⃣ I would naturally prioritize studies that offer open access to their data and methodologies. This makes it easy for a thorough examination of their findings. 3️⃣ You can use interdisciplinary inclusion for certain research questions. Including studies from different disciplines can provide a holistic understanding on the subject matter. For instance, a medical research question could benefit from insights from sociology or psychology studies.

The selection of papers with good methodological quality is a crucial focus in meta-analysis research, with a particular emphasis on rigorous procedures such as randomized controlled trials (RCTs) for therapies. Utilizing instruments like the Cochrane Risk of Bias assessment improves the validity and reliability of the compiled data by assisting in the evaluation and prioritization of studies with minimal risk of bias. This painstaking attention to methodological rigor is necessary to ensure the overall integrity of the research synthesis process and to provide reliable and believable meta-analysis findings.