Data Analysis for Successful Clinical Research

In clinical research, trials with patients are conducted to experiment novel treatments or improve existing ones. While performing these studies, a large amount of data is collected and generated, that needs to be properly handled. Data management includes all processes and procedures for collecting, handling, manipulating, analyzing, and archiving data from study start to completion. Good Clinical Practices (GCPs) are the most recognized standards for conducting clinical trials, and they are used by institutions, sponsors and regulatory agencies to monitor the conduct of research and data collection. Within GCPs, appropriate managing of clinical trial data assures that they are complete, reliable, correctly processed, and that data integrity is preserved.

By systematically applying statistical and/or logical techniques to describe, illustrate and evaluate data, data analysis brings order, structure and meaning to the mass of collected data. Depending on the type of data and on the scientific question, different methods of data analysis can be used to derive meaningful insights.

Measures of descriptive statistic are commonly used to calculate and summarize collected research data in a logical and efficient way. These are usually reported numerically in the manuscript text and/or in its tables, or can be graphically represented in figures.

The principal methods of descriptive statistic include measures of central tendency and measures of variability or dispersion. Mean, median, and mode measure the center or central tendency of a set of data, while standard deviation and variance are measures of variability or dispersion, which indicate how much the individual recorded scores or observed values differ from one another.

Inferential statistic is used to compare the data and to show the relationship between different variables. Inferential statistics methods are used to make predictions about a larger population after data on a sample of this population has been collected. One specific goal in inferential statistics involves the determination of the value of a population parameter starting from the sample data. The range of values that is used to estimate this parameter is called a confidence interval (CI).

The CI is statistically defined as follows: “if the level of confidence is set at 95%, it means that if data collection and analysis could be replicated many times, the CI should include within it the correct value of the measure 95% of the times.” So, for example, if data sampling from a same population is repeated 100 separate times, generating 100 distinct sample point estimates, and then 100 corresponding CIs are calculated, 95 of these intervals are expected to contain the true (correct) value.

Testing for statistical significance, calculating the effect of a tested treatment (or the strength of the association between an exposure and an outcome) and generating a corresponding confidence interval (CI) are fundamental steps researchers take to make valid inferences and to be able to draw conclusions from collected data.


Want to read more?

Check out our collection of topics in clinical research!

Privacy Policy