Complete Guide to Database Searching for Systematic Reviews

• Understand the Historical Development of PubMed: Learn about the evolution of PubMed from MEDLARS to its current state as a comprehensive biomedical database.

• Recognize the Impact of PubMed on Scientific Research: Explore how PubMed has democratized access to biomedical literature and its implications for global research accessibility.

• Appreciate the Features and Services of PubMed: Familiarize with the various components of PubMed, including PubMed Central, and their roles in supporting open scientific communication.

PubMed, launched by the National Library of Medicine (NLM) in 1997, has revolutionized the accessibility of biomedical information. Originating in the 1960s with MEDLARS, PubMed evolved through MEDLINE in the 1970s to become a robust online resource with the advent of the World Wide Web. Today, PubMed offers free access to over 36 million citations and abstracts from biomedical and life sciences literature, significantly impacting global medical research and healthcare. The establishment of PubMed Central in 2000 further advanced NLM's mission by providing a free full-text archive of journal literature. As one of the largest and most comprehensive databases in the world, PubMed continues to expand, adding thousands of records weekly and supporting diverse scientific fields beyond biomedicine.

• Understand the Function and Use of MeSH Terms: Gain a foundational understanding of Medical Subject Headings (MeSH) terms, their purpose in PubMed, and how they are used to enhance search accuracy.

• Navigate the MeSH Database: Learn to effectively navigate the MeSH database, understand its structure, and apply filters to refine searches.

• Construct Effective Search Blocks Using MeSH Terms: Develop skills to build effective search blocks by combining MeSH terms with free-text terms to create comprehensive search strategies

In today's lesson, we delved into the use of MeSH (Medical Subject Headings) terms, a pivotal component of PubMed's indexing system that enhances the precision of search strategies within the database. MeSH terms are used to tag each article in PubMed, ensuring consistency across the literature by grouping synonymous terms under a single standardized heading. We explored how to access and utilize the MeSH database from the PubMed main page, demonstrating the process of searching for and selecting appropriate MeSH terms to construct search blocks. Special attention was given to navigating the hierarchical structure of the MeSH database, which organizes terms from broad to specific, allowing researchers to refine their searches according to their precise needs. The tutorial highlighted practical examples, such as searching for 'Neoplasms' under various subheadings, and provided insights into the strategic use of these terms to minimize noise and enhance search relevance.

• Understand the Concept of [tiab] Terms: Learn what [tiab] terms are, how they function as synonyms or related terms in PubMed searches, and their role in enhancing the breadth of search results.

• Integrate [tiab] Terms with MeSH Terms: Master the process of integrating [tiab] terms with MeSH terms using Boolean operators to construct comprehensive search blocks.

• Utilize Online Resources for [tiab] Term Generation: Develop skills in utilizing various online resources such as Wikipedia, systematic reviews, and other databases to find relevant [tiab] terms and synonyms.

In this lesson, we explored the integration of [tiab] terms, also known as free-text terms, with MeSH terms to create effective search blocks in PubMed. [tiab] terms expand the scope of searches by including synonyms, spelling variations, and different grammatical forms of key concepts. We discussed the strategic use of Boolean operators to merge these terms with MeSH terms, thus ensuring a thorough search. Additionally, we examined various resources for generating [tiab] terms, such as systematic review databases, Wikipedia, and other websites, to ensure comprehensive coverage of the literature on a given topic. This approach is crucial for capturing all relevant literature and minimizing the chance of missing significant studies due to terminology differences

• Understand the Concept of Phrase Nesting in Database Searches: Learn how to use phrase nesting to control the order of operations in search queries, enhancing the precision of literature searches.

• Apply Boolean Operators with Phrase Nesting: Master the application of Boolean operators within nested phrases to effectively combine multiple search concepts.

• Design Complex Search Queries: Develop skills to construct detailed and complex search queries that accurately reflect research needs by logically grouping terms and phrases.

Phrase nesting is a critical technique in database literature searching that utilizes parentheses to group terms or phrases, controlling the order of operations in a search query. This method is akin to mathematical operations where expressions within parentheses are prioritized. In the context of literature searches, phrase nesting allows researchers to combine multiple concepts with Boolean operators (AND, OR, NOT) to target search results more precisely. For instance, researchers can construct queries like ( "low carb" OR "ketogenic" ) AND diabetes to ensure that the search engine retrieves articles that discuss either "low carb" or "ketogenic" diets specifically in relation to diabetes. This technique enhances search precision by ensuring that databases interpret complex queries correctly, thereby reducing irrelevant results and improving search efficiency. By learning to effectively nest phrases, researchers can refine their searches and more readily locate pertinent literature across extensive databases.

• Understand the Use of Filters in Database Searches: Learn about the different types of filters used in scientific and medical literature databases, such as population, study type, and publication date filters, and their role in refining search results.

• Apply Filters to Enhance Search Precision: Develop skills to apply specific criteria using filters to narrow down search results effectively, focusing on more relevant articles.

• Balance Comprehensive and Precise Searches: Recognize the importance of balancing the use of filters to avoid overly restrictive searches that might exclude relevant studies, ensuring both comprehensive and precise search results.

Filters are essential tools in scientific and medical literature searches, particularly in databases like PubMed. They allow researchers to apply specific criteria to narrow down search results, enhancing both the efficiency and precision of literature searches. Common types of filters include population filters, which focus on specific demographic groups; study type limits, which isolate articles based on study design like randomized controlled trials or cohort studies; and date limits, which confine searches to a particular publication timeframe. Other filters may focus on language, publication type, or specific fields within an article, such as keywords in the title or abstract. Implementing these filters through the database’s search interface can drastically reduce irrelevant results, saving time and increasing the relevancy of search outcomes. However, the judicious use of filters is crucial, as overly restrictive filtering may omit pertinent studies. By understanding and effectively applying filters, researchers can access the most suitable level of evidence for their needs and support robust research conclusions.

• Formulate a Focused Research Question: Learn to clearly define a research question using the PICO framework to guide the systematic search process in PubMed.

• Master PubMed Search Tools: Acquire skills in identifying and assembling Medical Subject Headings (MeSH) and tiab (Title/Abstract) keywords, applying truncation, phrase searching, and nesting techniques to refine search queries.

• Evaluate and Refine Search Results: Develop the ability to critically assess search results, applying filters and modifications to improve relevance and coverage, ensuring comprehensive retrieval of pertinent literature.

This lesson outlines a structured approach to performing a systematic search in PubMed, ensuring comprehensive and relevant results. The process begins with defining a clear research question, ideally structured using the PICO framework, which is crucial for guiding the search strategy. A logbook is recommended for documenting keywords, search dates, modifications, and results, enhancing reproducibility and allowing iterative refinements. Essential steps include assembling relevant MeSH terms and tiab keywords to capture all necessary concepts related to the research topic. Techniques such as truncation, phrase searching, and nesting help in refining the queries. The search strategy is built by combining these elements using Boolean operators in PubMed’s Advanced Search Builder, followed by a critical review of initial articles to adjust terms and apply filters effectively. This systematic approach is iterative, requiring continuous refinement to adapt to new findings and ensure the search remains up-to-date and aligned with the research question.

• Identify Common Mistakes: Recognize and avoid common errors in constructing search blocks using MeSH and tie-up terms in PubMed.

• Apply Advanced Search Techniques: Utilize advanced search strategies effectively to refine searches and achieve more relevant results.

• Address FAQs: Understand solutions to frequently asked questions about PubMed searches to enhance search efficiency and effectiveness.

In the concluding lesson of module 2, the instructor emphasized the consolidation of previously learned skills in searching PubMed, particularly focusing on the application of MeSH and tie-up terms. The session highlighted the identification and rectification of frequent search mistakes, enhancing the learners' ability to conduct error-free searches. Advanced search techniques were elaborated upon, providing learners with the skills to navigate complex queries effectively. The lesson also addressed common queries related to PubMed searches, aiming to preemptively solve typical issues users encounter, thereby reducing repetitive inquiries. This comprehensive approach ensures learners are well-equipped to handle PubMed searches confidently and competently, paving the way for efficient academic research.

• Understand Database Specializations: Identify the unique features and specializations of various scientific databases and select the appropriate database for specific research questions.

• Master Search Strategy Translation: Learn to translate search strategies effectively from PubMed to other database syntaxes to maintain consistency and effectiveness across platforms.

• Implement Advanced Search Techniques: Apply advanced search techniques such as citation chasing and removal of duplicates, enhancing the breadth and accuracy of literature searches.

This lesson provides a comprehensive overview of effective search strategies across various specialized databases, with a focus on PubMed. It introduces learners to the importance of selecting the right database based on the specialization relevant to their research questions. The instructor outlines methods for translating search strategies from PubMed to other databases to ensure comprehensive literature retrieval. Techniques such as citation chasing and duplicate removal are discussed to refine search results and improve the efficiency of literature reviews. The lesson also prepares learners to update their searches with the latest literature, crucial for the preparation of manuscripts and systematic reviews. By mastering these strategies, researchers can enhance their ability to conduct thorough and accurate literature searches, essential for successful academic and clinical research.

• Understand the Specialization and Coverage of Major Databases: Identify the main features and regional focuses of key databases such as PubMed, Embase, Web of Science, and the Cochrane Library.

• Develop Comprehensive Search Strategies: Learn how to formulate effective search strategies that can be translated across different database syntaxes to ensure thorough literature coverage.

• Apply Techniques to Manage and Refine Search Results: Master the methods for removing duplicates and managing the results from multiple database searches to ensure the relevance and quality of the gathered literature.

This lesson explores the strategic use of major databases like PubMed, Embase, Web of Science, and the Cochrane Library, each with its specific regional focus and subject coverage, essential for conducting extensive literature reviews such as systematic reviews and meta-analyses. The instructor details the importance of understanding the specialization of each database to select the most appropriate for particular research questions. Techniques such as translating search strategies across databases and managing duplicates are emphasized to enhance the efficiency of searches and ensure the comprehensiveness of the literature review. Additionally, the course covers the historical significance of these databases in preserving and accessing medical and scientific literature, providing researchers with a robust framework for accessing a wide range of publications necessary for high-quality research.

• Translate PubMed Search Strategies to Other Databases: Learn how to adapt a PubMed search strategy for use in different databases like Embase, Scopus, and PsycInfo, ensuring compatibility and accuracy across platforms.

• Understand Interface and Syntax Differences: Identify and navigate the differences in interface and query syntax between databases to efficiently manage and adapt searches.

• Optimize Search Results Across Databases: Apply strategies for combining search blocks and terms across different databases to maximize the comprehensiveness and relevance of search results.

This lesson provides an in-depth guide on translating PubMed search strategies to other major databases such as Embase, Scopus, and PsycInfo. Key concepts include adapting search blocks composed of thesaurus and free text terms to the specific syntax required by each database. The instructor emphasizes the importance of understanding both the interface and the query structure unique to each database to ensure effective and accurate information retrieval. Strategies for optimizing search results through proper syntax adjustment and term selection are also covered, enabling researchers to conduct comprehensive reviews across multiple platforms efficiently.

• Utilize Database Interfaces: Students will learn how to navigate and use multiple database interfaces effectively to implement search strategies.

• Apply Search Strategies: Students will be able to apply predefined search blocks across different databases and understand how to download and manage search results.

• Organize Search Results: Students will learn to create and maintain a systematic folder structure for organizing search results from various databases to facilitate easy access and management.

In this lesson, the instructor demonstrates how to implement search strategies across several databases, including PubMed, MBase, Web of Science, Cochrane, and others. The lesson covers navigating database interfaces, inserting search blocks, downloading results, and maintaining an organized structure for storing these results. Emphasis is placed on practical skills such as using advanced search options, selecting and deselecting certain types of results, and handling file formats like RIS for efficient data management. The instructor also discusses the importance of reproducibility and systematic logging of search strategies and results.

• Understand the concept of database overlap and the importance of deduplication in systematic reviews.

• Learn to identify and remove duplicate records using various software tools and manual methods.

• Develop proficiency in adjusting search and deduplication settings in reference management software to ensure accurate dataset preparation.

The lesson focuses on the critical steps involved in handling database overlaps and deduplication in systematic reviews. Due to the overlapping scope of databases like PubMed, Embase, and Web of Science, articles may appear in multiple databases, necessitating efficient deduplication methods. The instructor outlines the use of software tools such as EndNote and specialized scripts for automated deduplication, alongside manual checking to ensure the accuracy of the records. Effective deduplication enhances the review process by ensuring data quality and reducing redundancies, thus streamlining the research workflow and improving the reliability of systematic reviews.

• Understand the principles and applications of citation chasing (snowballing) in literature searching.

• Learn the techniques of backward and forward citation chasing to uncover both foundational and recent studies related to a research topic.

• Develop skills to systematically expand literature searches using citation databases and manual review to enhance the comprehensiveness of systematic reviews.

Citation chasing, or snowballing, is a vital technique used in systematic reviews to ensure comprehensive coverage of relevant literature. This method involves both backward citation chasing, where researchers examine the reference lists of key articles to find additional relevant studies, and forward citation chasing, which identifies newer studies that cite the initial key articles. Utilizing databases like Web of Science, Scopus, or Google Scholar facilitates this process by allowing researchers to trace the citation network of seminal works. This approach helps in identifying pertinent studies that might be missed in conventional database searches due to indexing limitations or specific terminology used in the literature. As such, citation chasing significantly enhances the depth and breadth of literature reviews by capturing both historical and cutting-edge research.

• Understand the importance and process of updating systematic reviews to include the most recent studies before publishing.

• Recognize the concept of "living systematic reviews" and how they differ from traditional systematic reviews.

• Identify the benefits and challenges associated with maintaining living systematic reviews.

Systematic reviews must be updated regularly to incorporate the latest research, ensuring that their conclusions remain relevant and accurate. This update, especially performed just before publishing, integrates recent studies that might alter the findings or recommendations of the review. The advent of "living systematic reviews" has further evolved this practice. These reviews are continually updated with new data as they become available, reflecting changes in understanding or practice in fast-evolving fields. They require rigorous and ongoing monitoring of literature, but provide timely evidence, increasing the efficiency and responsiveness of systematic reviews to new information. However, they also demand substantial resources and commitment to maintain their validity and reliability over time.

• Understand how to correctly report the systematic search process in a published systematic review.

• Learn to accurately utilize and describe PRISMA flowcharts to outline the review process.

• Recognize the importance of including detailed methodological transparency in systematic reviews to facilitate reproducibility.

In systematic reviews, accurate reporting of the search strategy is crucial for transparency and reproducibility. This includes detailing the methods used to gather data, such as the databases searched, the search terms employed, and the date of the last search. It's essential to incorporate a PRISMA flowchart, which visually represents the process of selecting studies, including the number of records identified, excluded, and the reasons for exclusions. Additionally, results from each database should be reported, typically in the appendix or supplementary materials of the publication. By adhering to these reporting standards, researchers ensure that the review not only maintains rigorous methodological quality but also provides a verifiable path for others to replicate or build upon the findings.

• Automation and AI Tools for Systematic Reviews:

• Systematic Review Workflow:

• Data Management and Analysis Tools:

In this module, the focus is on the various tools available for automating and enhancing the systematic review process. It begins with team formation and the importance of having at least two authors for blind screening. The module then outlines the development of research questions and protocols using tools like Methods Wizard and the Prisma Checklist. Search strategies are designed with AI tools such as MedSearch, and bibliographic management is streamlined with software like EndNote and Zotero. The process continues with article selection, using tools like Covidence and ASReview for efficient screening. Data extraction and analysis are facilitated by ChatPDF, the Cochrane Risk and Bias tool, and other AI tools. Finally, the module covers the writing and publication stages, emphasizing the use of advanced AI writing tools to ensure the manuscript is publication-ready.

• Learn how to develop a focused research question for a systematic review using specialized online tools.

• Understand how to plan and structure the overall process of a systematic review effectively using project management tools.

• Familiarize with the Methods Wizard and Research Question Design Tool to enhance the preparation stages of a systematic review.

In this lesson, we introduced two essential tools that aid in the initial stages of conducting a systematic review. The first tool, a research question design tool available at rawbirdie, helps researchers formulate precise research questions based on inputted search terms. This tool is particularly useful for those still defining their study's scope. The second tool, Methods Wizard, offered by Bond University, assists in organizing and planning the systematic review process. It guides researchers through creating a structured framework, ensuring all elements of the PRISMA checklist are addressed, from project initiation to the final reporting stages. These tools are designed to streamline the systematic review process, making it more efficient and manageable.

• Understand how to use MedSearchAssistant to create PubMed search strategies from a given research question.

• Learn the process of translating search strategies to different database syntaxes using MedSearchTranslator.

• Recognize the importance of verifying mesh and thesaurus terms when adapting search strategies across various databases.

In today’s lesson, we explored two powerful tools, MedSearchAssistant and MedSearchTranslator, designed to streamline the creation and adaptation of search strategies for systematic reviews. MedSearchAssistant allows researchers to input research questions, which it then uses to generate search strategies specifically for PubMed, incorporating relevant MeSH terms. MedSearchTranslator takes these strategies and adapts them for use in other databases like Mbase, Web of Science, Cochrane, and Scopus by adjusting syntax and terms according to the specific requirements of each database. The lesson emphasized the necessity of double-checking the auto-generated terms to ensure their validity, especially when databases do not provide error notifications for incorrect terms.

• Understand the functionality and application of article selection and screening tools like Rayan, Covidence, and AS Review in systematic reviews.

• Learn the criteria for choosing between different selection tools based on the size of the literature pool and specific project needs.

• Explore the benefits of machine learning-based screening tools for managing large datasets effectively.

This lesson introduced three significant article selection and screening tools: Rayan, Covidence, and AS Review, each catering to different aspects of the systematic review process. Rayan, an open-source tool, is recommended for managing literature pools up to 12,000 references, beyond which it becomes cumbersome for manual screening. Covidence offers a similar functionality with a structured approach but is a paid service. AS Review utilizes machine learning to assist in the screening process, making it ideal for very large datasets by learning selection patterns and prioritizing relevant articles. This helps in significantly reducing the screening workload and improving the efficiency of systematic reviews.

• Understand the importance of literature mapping in identifying research gaps and overlooked publications.

• Learn how to use specific tools like Research Rabbit, Connected Papers, and Litmaps for literature mapping.

• Gain insights into the processes of forward and backward citation chasing to build comprehensive literature reviews.

In this lesson, we explored the significance of literature mapping tools in enhancing systematic reviews by identifying research gaps and overlooked studies. Tools like Research Rabbit, Connected Papers, and Litmaps allow researchers to input either specific research questions or key papers to generate visual maps of related literature. These tools facilitate both forward and backward citation chasing, helping to identify foundational and subsequent works that may not be immediately apparent. This process not only aids in uncovering direct connections but also in visualizing the broader landscape of research, including potential collaborators and influential authors within the field.

• Understand the process and importance of data extraction from PDF files for systematic reviews.

• Learn to use tools such as Chat PDF and Chat Doc for querying and extracting data from academic papers.

• Explore the application of a custom-built RAG system to automate data extraction and enhance research efficiency.

Today's lesson focused on advanced tools for data extraction in the context of systematic reviews. We explored tools like Chat PDF and Chat Doc, which leverage large language models to extract specific information from PDFs through user queries. These tools simplify the extraction process by allowing direct questioning of the document content, making systematic analysis and comparison of results more efficient. Additionally, we discussed a custom-built RAG (Retrieval-Augmented Generation) system designed to automate the extraction process by cycling through predefined questions for each document. This system integrates with large language models to pull relevant data into structured outputs, drastically reducing the manual effort required in traditional data extraction methods.

• Explore the various types of AI tools that can aid in academic research, including writing aids, research enhancers, and plagiarism detectors.

• Understand how to effectively integrate AI tools into research workflows to enhance productivity and ensure content originality.

• Learn strategies to humanize AI-generated content to meet academic standards and avoid plagiarism.

In today's lesson, we delved into the application of artificial intelligence in academic research. We categorized AI tools into three main types: AI writing tools, AI research tools, and AI detection tools, each serving distinct purposes in enhancing research quality and efficiency. AI writing tools, like GPT-4 and Geni, assist in drafting and refining academic texts. Research tools, such as Elicit and Evidence Hunt, help in sourcing relevant literature and data. AI detection tools like Turnitin and Originality.ai ensure the originality of content by detecting plagiarism and AI-generated text. Additionally, we discussed the importance of humanizing AI-generated content to preserve the authenticity of academic work, employing tools like Quillbot and Undetectable AI for this purpose. These technologies, when used responsibly, can significantly bolster research productivity and integrity.

Thank you so much for purchasing this course! I hope you've learned a lot along the way. Reflecting on our journey, we started with the daunting ocean of literature and began with just a little raft. Now, look at you—you've built an entire empire and conquered this ocean with the skills you've mastered.

Don't forget to check out the supplementary materials included with each module. These handouts, such as database syntaxes and effective duplicate removal techniques, will be invaluable. Additionally, all the AI tools covered in the last lesson are documented in a neat package. I'm also working on converting this entire course into an ebook, which will be available for free to all course purchasers.

To reinforce your learning, each lesson comes with quizzes to help solidify the knowledge. Looking ahead, I plan to further develop medsearchsolutions, bringing more tools to enhance your research capabilities. I also intend to create more courses, focusing on AI and building an effective tool stack for systematic literature reviews and other publications.

Thank you once again for joining this course. Please rate it and get in touch with any questions. Your feedback is invaluable, and I look forward to seeing you in future courses!