Director's Messages

New and Updated Resources Available To Help Design Rigorous Clinical Trials

A version of this message originally appeared as a guest post on the NIH Office of Extramural Research’s Open Mike blog on June 9, 2021.

The NIH embarked on a series of initiatives in recent years to enhance the quality, efficiency, accountability, and transparency of our supported clinical research. While we are all making great progress, there are still concerns about clinical trials that are overly complex, have small sample sizes, or rely on surrogate end points that lack clinical relevance.

In 2017, we launched the NIH Research Methods Resources (RMR) website to help address these concerns. Since the site was recently revamped, we wanted to spotlight the new available tools and resources that can help you better plan the design, conduct, and analysis of rigorous NIH-defined clinical trials.

Interested in learning more?

Check out the June 29 Methods: Mind the Gap webinar with Dr. Murray.

The site focuses on trials that randomize and/or deliver interventions to groups or clusters. Though these trials are becoming more common in biomedical research, they can be misunderstood. Too often, the analytic plans for these trials may be inappropriate and have inadequate power due to small sample sizes. To have a strong and rigorous study, researchers should provide sufficient information on the selected methodology, statistical analysis, and sample size when proposing a trial.

The updated site has a cleaner layout and search function. Notably, it has a new sample size calculator specific to individually randomized group-treatment (IRGT) trials (Figure 1). This new resource makes it easier for researchers using this trial design to strengthen their sample size and statistical analysis plans.

Figure 1 is a screenshot of the first page of the IRGT Sample Size Calculator from the NIH Research Methods Resources website. Six-and-a-half steps are listed on the left, with step #1 Type 1 Error Rate and Power highlighted. A preview of instructional text appears in a box near the top of the page
Figure 1. A screenshot of the new IRGT sample size calculator, with “Type 1 error rate and power” highlighted.

Inspired by a similar calculator for group-or cluster-randomized trials (GRTs) from the original site (Figure 2), the new IRGT sample size calculator can assist with cohort and cross-sectional designs. Tutorials are available for each aspect of the sample size calculation, and results can be downloaded to work with offline. There are also new examples that illustrate the formulas each sample size calculator uses to arrive at the results.

Figure 2 is a full screenshot of the results page for the GRT Sample Size Calculator from the NIH Research Methods Resources website. Nine steps are listed on the left, with the results and explanatory steps in the middle, two tables at the bottom of the page, and a highlighted box with the entered parameters on the right
Figure 2. A screenshot of the results page for the GRT sample size calculator.

In addition, we worked to improve the general user experience by redesigning the look and feel of the site. New buttons, for instance, were added to make it easier to find FAQs and launch the sample size calculators. Hyperlinked citations also now take users to the exact listing on the main References page, where they can find new filter functions and all the citations used across the site. While we added tools and functionality, we also kept a number of features from the original site, including links to webinars, relevant Consolidated Standards of Reporting Trials (CONSORT) statements, and a glossary of relevant terms.

If you’re interested in more details, we invite you to register for the Methods: Mind the Gap webinar, Design and Analytic Methods for Group-Based Interventions, on June 29, 2021. We’ll be providing a closer look at the redesigned RMR site as well as the appropriate use, design, and analysis of IRGTs.

Later this year, we expect to add even more material to the site, including resources for stepped wedge designs. Once available, this sample size calculator will help investigators design these types of trials based on current recommendations.

We encourage you to check out RMR today and let us know your feedback. It is our hope that by referring to the site early and often when developing your statistical methods, we can all continue promoting innovative and rigorous designs as well as proper stewardship of our supported clinical trials.