ABC’S (Application of Basic Concepts) in Dermatology Research

An introduction to observational studies

Mara Padilla Evangelista-Huber MD, DPDS

A clinical trial is a planned experiment designed to evaluate the benefits of a treatment for a specific medical condition. A welldesigned clinical trial is the most rigorous research design for evaluating treatments. However, there several situations where it is neither possible nor ethical to conduct a clinical trial.

Writing results in a clinical trial report

Mara Therese P. Evangelista MD, DPDS and Nick Huber

During the course of a clinical trial, large amounts of data are usually collected at different time points (i.e. at baseline and during follow-up visits). This results in the possibility of multiple statistical tests and various ways that the results can be reported. Some authors would naturally hope to show ‘interesting’ data, and this can lead to a tendency to hype the results. This must be resisted. It is of the utmost importance to report a balanced view of the results. This is why it is essential, before a trial starts, to have a protocol that states the primary and secondary endpoints and includes a pre-specified analysis plan.

Sample size calculation 101: Unraveling the concepts behind the numbers

Mara Therese P. Evangelista MD, DPDS and Nick Huber

Because the objective of a clinical trial is to provide reliable evidence of the efficacy and safety of a treatment, a major issue in the design of clinical trials is ensuring we have a sufficient number of patients to provide the required level of evidence. Essentially, we need to have enough subjects in our trial to give a good chance of detecting a clinically important treatment difference if such a difference exists, while being able to reasonably conclude that no such difference exists if our results do not show it. The focus of this article is the sample size calculation for superiority trials. There are different methods in determining the sample size in trials with a non-inferiority design, or those that are randomized by cluster rather than individually. In order to appreciate the basics of sample size requirements for a trial, some understanding of key statistical concepts is essential.