Designing Studies to Affordable Sample Sizes

There Are Always Design Options
You design a study, write a protocol, settle on the ideal marketing claim, and then ask the statistician for the sample size.

Next you change the design, rewrite the protocol, and re-imagine the indications and claims to get to a sample size you can afford.

Learning Objective
The objective of this event is to understand how design choices and other factors influence sample size and how to use them to your advantage.

Non-Statistical Factors that Influence Sample Size
The sample size of your study is a result of many non-statistical influences, and having a well-powered study is not necessarily FDA's goal.

Suppose your device is a "me-too", offering the world yet another variation of a common commodity. The Agency may require the same sample size that everyone else has used in order to keep the playing field level and treat everyone the same.

Suppose your device is expected to offer only a small increase in performance, say 3%; the sample size needed to power the study will be enormous. The Agency may negotiate a smaller sample size large enough to support performance but not superiority.

It can go the other way, your product may perform so well that a mere 12 subjects would give an well-powered study; but the Agency says no, we need 40 or 100 subjects/samples in order to get a feeling for distribution.

Maybe you have survival data, and you can word the claim: 1) our product lasts longer than the other product, 2) our product is as good as the other product, 3) 80% of our product lasts longer than a year, 4) on average, our products last a year. Each variation leads to a different sample size.

Of course there is always the calculated sample size. But even here, the sample size is calculated on the primary hypothesis--suppose you change the hypothesis? Suppose you can defend a larger expected difference in performance between the investigational and control devices? The result will be a different sample size.

Business Goals in a Tough Economy
Your business goals are to gain cash flow and demonstrate your team's ability to get FDA approval. In other words: cheapest, fastest claims first. A smaller sample size can be key to reaching the market and proving product viability.

You will receive
[x] PowerPoint slides.
[x] An expert speaker.
[x] New ideas about protocol designs.
[x] New ideas about sample size calculations.
[x] A chance for Q&A.
[x] 0.15 CEUs and Certificate of Attendance.

Audience suitability
[x] People who write protocols or design studies.
[x] Statisticians who calculate sample sizes.
[x] Regulatory professionals who file submissions.
[x] Management trying to manage better.

Presenter
Dr. Robert P. Thiel has degrees in physics, clinical counseling and psychometrics. He is an expert in the analysis of IVD data for the Medical Device Industry with more than 75 510K and 5 PMA approvals. He has published papers in the area of free PSA, breast cancer, ovarian cancer, liver disease and multivariate analyses. His current interests lie in the area of bootstrap and permutation test methodology and the application of these methods to small sample trials and Bayesian analysis as applied to diagnostic clinical trials. Dr. Thiel can be contacted at rpthiel@thielstatcon.com.

The purchase is $424 for OnDemand access and $474 for the CD. Each attendee is eligible for 0.15 CEUs upon completing the course feedback via the link provided with the event materials. Click here to purchase.