Introduction to evidence-based practice for the health sciences

Peer-reviewed sources

Peer review is a process used to ensure the quality, accuracy, and reliability of research before it is published in scholarly journals. It works very much like a filter and aims to prevent flawed, biased, or low-quality studies from being shared with the scientific community. This is essential in healthcare and other fields where evidence-based decisions impact patient care and policymaking.

The process begins when a researcher submits their study to a journal. The journal then sends the study to other experts, known as peer reviewers or referees, who specialise in the same field. These reviewers carefully evaluate the study’s methods, data accuracy, strength of evidence, and conclusions to determine whether the research is valid and relevant. They provide feedback to the author, suggesting revisions or identifying issues, and, if necessary, the researcher makes improvements before resubmitting the paper.

Based on the reviewers' recommendations, the journal may accept the study, request further revisions, or reject it if it does not meet the required standards. Only studies that pass this rigorous process are published as peer-reviewed evidence, making them more credible sources for academic study and clinical decision-making. By promoting accuracy and preventing misinformation, peer review helps maintain research integrity and supports evidence-based practice.

Watch the video Peer review in 3 minutes (YouTube, 3m,14s) to learn more about how the peer review process works.

"Peer review in 3 minutes" by libncsu is licensed under CC 3.0 BY-NC-SA.

Limitations of peer review

While peer review helps ensure research quality, it has some weaknesses.

• Traditional peer review is often closed and anonymous, meaning the review process is not visible to the public. This can make detecting bias, conflicts of interest, or unfair rejections difficult. Some journals are moving toward open peer review, where reviewer comments are published alongside the article to improve transparency.
• Reviewers may have personal or professional biases that affect their judgment. They may favour studies that support their views or be influenced by an author’s reputation or institutional affiliation. Double-blind peer review, where both the authors and reviewers remain anonymous, helps reduce bias, but it is not always used.
• Peer review can be a slow process that delays the publication of significant research findings. In fast-moving fields like medicine, this can slow the adoption of new treatments or technologies that could benefit patients.
• Different reviewers may provide contradictory feedback, leading to confusion or unfair decisions. A study may be rejected by one journal but accepted by another with minimal changes, suggesting that the process is not always objective.
• Some researchers have attempted to manipulate the process, such as submitting fake reviews, self-reviewing their own work or engaging in "peer review rings" where colleagues approve each other’s work without proper scrutiny.
• Peer reviewers are typically unpaid and conduct reviews in addition to their regular jobs. High workloads can lead to superficial or rushed reviews, reducing the quality of feedback.

Despite these challenges, peer review remains the best system for ensuring research quality. Innovations such as open peer review and AI-assisted screening may help address these issues while maintaining academic integrity.