Why doctors trust some data more than others

Ideas & Philosophy · Published 19 October 2026 Swipe more like this →

In short: In medicine, not all facts are created equal. To prevent decisions from being swayed by mere anecdotes or personal bias, researchers use a structured hierarchy. This framework ranks study designs by their dependability, ensuring that clinical guidelines are built on the most rigorous science available.

The hierarchy of evidence is a heuristic framework that dates back to the mid-20th century. It was pioneered by British epidemiologist Archie Cochrane, who emphasized the necessity of systematic reviews of randomized controlled trials (RCTs). This system was later popularized by seminal publications like those from Sackett et al., becoming a cornerstone of medical education. The primary goal is to minimize bias—such as prejudice or moral constraints—and ensure that medical procedures are safe, effective, and based on recent scientific advances rather than tradition or individual experience.

The structure is often visualized as a pyramid. At the peak are systematic reviews and meta-analyses, which combine data from multiple high-quality studies to provide decisive conclusions. However, their reliability depends entirely on the quality of the underlying research; low-quality inputs will inevitably produce subpar results. Below these are RCTs, which use random allocation to establish causation and minimize selection bias. While rigorous, RCTs are often criticized for being resource-heavy, inflexible, and subject to ethical limitations that may leave certain therapies unstudied.

Observational research, such as case-control and cohort designs, occupies a lower tier. Researchers monitor cohorts throughout periods, but case-control methods contrast individuals possessing a condition against those who do not. These provide significant insights but are less reliable than RCTs due to potential confounding variables. At the foundation lie the least dependable forms of evidence: case reports, expert opinions, and anecdotal evidence. While these are valuable for identifying new phenomena, they lack the controlled design necessary to establish broad clinical truths. Today, the hierarchy is evolving to incorporate real-world data, big data, and artificial intelligence to meet contemporary demands.

The framework is not static. Scholars like Jacob Stegenga define it as a rank-ordering based on the potential for systematic bias, while others, like Greenhalgh, view it as the relative weight of different primary studies. Because there are over 80 different proposed hierarchies, the field continues to refine how we measure the certainty of evidence, balancing internal validity with real-world effectiveness.

General information, not professional advice. Spotted an error? Flag the card or see corrections and complaints.

Intellect Plus is a visual educational discovery platform for exploring fascinating facts and ideas. Brain Food, Not Brain Rot. What is Intellect Plus?

Related

• Is the gold standard of modern medicine failing the individual patient?
• Why the most rigorous research method is the gold standard for high-stakes decision-making
• Why there is no single, universally accepted definition of scientific evidence
• Why two doctors might see two different things in one patient
• How researchers mimic the 'gold standard' of medical trials using math
• Why the law treats 'he said, she said' with such suspicion
• Why some of our greatest discoveries were once considered 'impossible'
• In some languages, you cannot speak without proving how you know it
• The physics of homeopathy: How a 30C dilution would require a planet-sized container
• Why the hearsay rule is actually a tool for clarity, not confusion
• Why a common misconception about Auschwitz gas chambers persists
• Why your 'normal' blood test might be hiding nerve damage

More in Evidence · More in Ideas & Philosophy · Latest · A–Z · All topics