Are All Published Research Articles Created Equal? Moving Beyond P-values

Session Description: Implementing evidence-based practice strategies is an important tool in an Infection Preventionist’s (IPs) arsenal to reduce infections in hospitalized patients. But, evaluation of research studies and findings is a vital skill that many IPs may not have much experience with. We aim to help IPs recognize the hallmarks of both higher and lower-quality research, allowing them to implement interventions in their care settings that are supported by sound scientific evidence.

As consumers of research, it is important for IPs to understand potential issues that can impact study validity, indicators of research quality, and methods for improving research integrity. This session will explore issues plaguing the research literature, and how they are relevant to the practice of Infection Prevention. These problems include the over-abundance of statistically significant findings, lack of reproducibility, the false discovery rate, dishonest research practices, predatory publishing, conflicts of interest, and the pitfalls of null hypothesis statistical testing. We will discuss underlying factors leading to the problems listed above and tempting researchers to engage in questionable and unethical behaviors. Next, IPs will learn about the work of researchers committed to raising standards for published literature. We will examine the pros and cons of proposed strategies for improving empirical research. Such strategies include justification of p-values, changing the threshold for statistical significance, adherence to reporting guidelines, study pre-registration, data transparency, use of research badges, registered reports, adversarial collaboration, and blind data analysis. We will share examples articles employing these techniques, and discuss strategies for finding such articles while searching the literature. This session would be especially beneficial for early-career IPs, as well as any IP who reads and evaluates scientific literature with the goal of implementing evidence-based practices.