Introduction to evidence-based practice for the health sciences

Types of study

In healthcare research, the evidence pyramid is divided into different levels with the strongest and most reliable evidence at the top and weaker evidence at the bottom.

Secondary (filtered) information:

• Systematic reviews use explicit and rigorous methods to identify, critically analyse, and synthesise relevant studies on a specific topic to answer a clearly defined research question. The systematic review process aims to reduce bias and provide a clear summary of findings based on the available evidence. Some include meta-analyses that combine findings from multiple studies to provide a comprehensive and statistically supported conclusion. 
• Critically appraised topics (CATs) and articles summarise research evidence to help healthcare professionals and researchers quickly assess the quality and relevance of studies on a specific clinical question. They fall between primary research and systematic reviews in terms of depth and rigour. 
  • Critically-appraised topics (CATs) are brief, structured summaries that evaluate the best available evidence on a focused clinical question. CATs can be helpful in decision-making but may not be as comprehensive as systematic reviews.
  • Critically-appraised articles are individual research studies reviewed by experts to critically evaluate the study design, methodology, bias, and significance to ascertain their validity and reliability, and they may be applied to clinical practice. 

Primary (unfiltered) information

The following three levels of the evidence pyramid are made up of primary or unfiltered sources of information.

• Randomised controlled trials (RCTs) are studies that compare a treatment or intervention to determine which is the most effective. They are considered the gold standard for clinical trials as they are a reliable method of determining cause-and-effect relationships.
• Cohort studies are observational studies that follow a group of people, also known as a cohort, over time to see how different exposures may affect health outcomes. They are considered more substantial evidence than case-control studies but not as reliable as RCTs as they don't include randomisation. 
• Case-controlled studies, Case series, and case reports
  • Case-controlled studies compare people with a condition or disease (cases) to those without (controls) to identify possible risk factors. They are often used to study rare diseases or health outcomes and can be useful when researchers cannot assign study participants randomly. 
  • Case series are collections of similar case reports. These descriptive studies provide insights into the characteristics and outcomes of patients included in the series. By observing patterns in the case series, researchers may be able to develop hypotheses for further research. They can be valuable for studying uncommon conditions where large, controlled trials are not possible.
  • Case reports provide detailed descriptions of an individual patient's condition or treatment and often focus on unusual or rare disease presentations, a unique approach to treatment, or an unexpected clinical outcome. They can be used to share clinical observations and generate new research questions. 

As these lower types of evidence lack rigorous external review, it is important to critically evaluate factors such as sample size, methodology, and bias before relying on their findings. 

Background information & expert opinion

These can provide foundational knowledge but are considered the lowest level of evidence in clinical decision-making. While these resources can be helpful for understanding concepts, they lack the rigorous methods of higher-level evidence. 

Types of clinical question and types of study

The type of question you are trying to answer will often inform the type of study that will best answer your question. 

* It is not always ethical to randomise study participants to a group that will be exposed to known harm. However, many RCTs include information on adverse events and side effects that can help answer specific clinical questions regarding harm. 

** A controlled trial that examines patients with different severity levels of an illness, where all participants undergo both the test being evaluated and the established "gold standard" diagnostic test for comparison.