Using artificial intelligence and technology for comprehensive research: suggestions for researchers and students

The gap between research and practice.

Today’s rapidly changing healthcare environment requires physicians, nurses, and students to remain current on the newest research, topics, and evidence to guide practice. While this may increasingly seem fundamental to nurses, many barriers make it extremely difficult. Aspects resembling limited time, a considerable amount of latest research to review, and experience reading and analyzing research contribute to the so-called research-practice gap. On this blog post, we are going to discuss tips on how to use artificial intelligence and technology to realize a comprehensive, high-level overview of an interesting research topic.

Define the subject.

Before using AI tools, it is incredibly necessary to concentrate on a subject of interest. To make your search accurate, it is useful to formulate or organize your topic or area of ​​interest. For instance, you need to use the PICO (Patient/Population, Intervention, Comparison, and Outcomes) format to formulate your query or area of ​​interest.

As an example you need to learn more about skin damage related to external urinary devices in adult women. PICO query may be: What variety of skin damage occurs in adult women in acute care settings when using external soaking devices in comparison with those that don’t use these devices?

Search the literature.

Searching academic databases will be difficult, and rigorous searching often requires the help of a medical librarian. A big language model (LLM) like ChatGPT can aid you discover a comprehensive search strategy. Attempt to prompt your LLM by asking the next questions, replacing the particular topic example along with your own well-developed PICO query:

“Please develop a search strategy (using MeSH and keywords from the title/abstract) for the next PICO query: What variety of skin damage occurs when using external urinary devices in comparison with individuals who don’t use these devices? Provide logical expressions and PubMed syntax.”

Below is a screenshot of a part of the outcomes of the ChatGPT search strategy in response to a different PICO query about bullying in nursing:

This answer can in turn be copied and pasted into the PubMed search wizard, as shown below:

Below is the syntax copied to the PubMed database to look for relevant articles. Within the screenshot, only the primary article within the PubMed results is visible.

Is that this method reliable enough?

While ChatGPT and other chatbots are known to be unreliable to find the very best peer-reviewed academic literature attributable to their tendency to “hallucinate” articles that don’t exist, from my experience,ChatGPT is sort of precise in its focused task of helping ,design search strategies and create a database syntax that may ,provide highly reproducible literature searches. Nevertheless, they must be understood as a tool; critical considering stays essential.

Use AI tools to grasp and develop the subject.

Now that you will have a well-developed topic query and search procedure, the subsequent step is to review the literature that can be useful in your practice. Nurses can use quite a lot of tools to make this process less daunting.

is a free website that permits users to submit one (or more) articles of interest. The system will mechanically provide the user with a map of comparable authors on the identical topic and related articles (authors who quote one another). This permits users to get a whole overview of the history and current development of research on a particular topic.

are free AI-powered tools that allow users to submit articles (positioned through the search process), providing users with AI-generated paper evaluation. While you’re attempting to get an outline of a subject or area of ​​interest, summarizing the predominant points of an article will be helpful and prevent time. Most of those open access tools allow users to download data extraction into Excel tables. Nurses can then take these tables to research board or unit meetings to assist guide practice based on current research.

Ethical considerations.

Each time nurses analyze research, special and detailed care should be taken before deciding to alter practice in a way which will impact end users, on this case patients. Given the unreliability of AI-based tools, nurses should work together to verify all data extracted from articles, as chatbots like ChatGPT still are inclined to “hallucinate” (or invent) information. Careful research is required before implementing practice changes on the patient level.

Incorporating AI and technology into the research process can provide a comprehensive, high-level overview of your chosen topic. As a nurse, you may refine your research skills and help apply unit-level evidence and specific practices to clinical settings. By harnessing the ability of those technological advances while exercising healthy caution, you may confidently and effectively navigate the complex world of nursing research and evidence-based practice.

Justin Fontenot DNP, RN, NEA-BC, FAADN, Associate Professor, Tulane University School of Medicine, Nursing Program. His previous post on AI and writing and research strategies was “Can Nursing Students Profit from Using Artificial Intelligence?”