The SAMPL guidelines: Important considerations for statistical analysis in biomedical research

Originally published in 2014, the SAMPL (Statistical Analyses and Methods in the Published Literature) are a concise, yet comprehensive summary of how to report commonly used statistical methods and results in journal articles. Although these guidelines focus on the most common statistical analyses in biomedical research, they are comprehensive enough to prevent a number of common reporting deficiencies in statistical analysisi,ii. 

The SAMPL guidelines are available at the EQUATOR Network.  

Who developed the SAMPL guidelines? 

The guidelines were originally developed by Thomas A. Lang, Principal, Tom Lang Communications and Training International, and Douglas G. Altman, Director, Centre for Statistics in Medicine, Oxford University.  

What do the SAMPL guidelines cover? 

These guidelines explain to researchers, journal editors, and peer reviewers how basic statistical analyses and their results should be reported, especially in biomedical research papers. The guidelines do not cover study design or how the research is conducted, but focus on statistical reporting. The main kinds of analyses covered in the guidelines are as follows: 

1. Numbers and descriptive statistics 

2. Risk, rates, and ratios 

3. Hypothesis testing 

4. Association and correlation analyses 

5. Regression analyses (linear, multivariate, stepwise, etc.) 

6. Bayesian analyses 

The guidelines also differentiate between preliminary, primary, and supplementary analyses.  

How can researchers use the SAMPL guidelines? 

Since the SAMPL guidelines focus mainly on reporting of statistics, it is advisable to consult them at the data analysis stage (to make sure you are running the necessary tests) and to refer to them once again while you are drafting your manuscript (to confirm you’ve included all the required details). Keep in mind that you will also need to refer to other guidelines based on the kind of study you are conducting (such as CONSORT for clinical trials or STROBE for observational studies). 

Advantages and disadvantages of the SAMPL guidelines 

The SAMPL guidelines are a useful tool for preventing the most common errors in reporting statistics in biomedicine. They can help you craft a manuscript with strong and high-quality evidence. As they are subdivided as per analysis type, they are also comparatively easy to navigate. 

Related Infographic

Top 5 ethical considerations when you conduct research

Did you know that often researchers are unaware that…

29.3k

However, these guidelines do not cover meta-analysis, nor do they go into much detail about diagnostic testing (such as ROC analysis).  

Conclusion 

Referring to the SAMPL guidelines while conducting your statistical analysis and drafting your manuscript is a good strategy. This will prevent you from making avoidable errors or omissions in reporting statistical data and can also smoothen the peer review process.  

Would you like a 1:1 consultation with an expert statistician? Check out Editage’s Statistical Analysis & Review Service.