In this free webinar hosted by nQuery Researcher & Statistician Eimear Keyes, we map out the 5 essential steps for sample size determination in clinical trials. At each step, Eimear will highlight the important function it plays and how to avoid the errors that will negatively impact your sample size determination and therefore your study.
Watch the Video: https://www.statsols.com/webinar/the-5-essential-steps-for-sample-size-determination
4. Sample Size Determination Challenges
REGULATORY APPROVAL
Time Consuming
Coding/Human Error
COMMUNICATION
Difficultly in sharing real
time examples
TOO LARGE/SMALL A SAMPLE
Waste Money &
Unethical to Subjects
STATISTICAL SIGNIFICANCE
Reduce Chance of Large
Errors (Type S/M Errors)
5. 5 Essential Steps for Sample Size
1 Plan Study Study question, primary outcome, statistical method
2 Specify Parameters Significance Level, Standard deviation, ICC, dispersion
3 Choose Effect Size Expected/targeted difference, ratio or other effect size
4 Compute Sample Size Sample Size for specified metric such as power
5 Explore Uncertainty Sensitivity Analysis, Assurance, Alternative Designs
6. In 2017, 90% of organizations with clinical trials approved
by the FDA used nQuery for sample size and power calculation
8. Consider Design Questions
What is the primary outcome of the study?
What type of hypothesis test will be used?
What kind of grouping structure will the study have?
What question/s do you want to answer?
9. Means Example
“An active-controlled randomized trial
proposes to assess the effectiveness of Drug
A in reducing pain. A previous study showed
that Drug A can reduce pain score by 5
points from baseline to week 24 with a
standard deviation (σ) of 1.195. A clinically
important difference of 0.5 as compared to
active drug is considered to be acceptable.
Consider a drop-out rate of 10%.
For this test we would like to find the sample
size required for 80% power, with a two-
sided 5% level of significance.”
Source: ncbi.nlm.nih.gov
Parameter Value
Significance Level (Two-Sided) 0.05
Mean Difference 0.5
Standard Deviation 1.195
Dropout rate 10%
Power 80%
11. Analysis Parameters
What parameters are needed for your method?
Significance level, standard deviation, intra-cluster correlation,
dispersion, etc.
Which parameters are known or unknown prior to the study?
Some parameters e.g. significance level can be chosen, others e.g. SD
must be estimated
What is your best estimate for these parameters?
Taken from pilot studies or expert opinion
13. Standardized or Unstandardized Effect Size
Unstandardized Effect SizeStandardized Effect Size
Raw treatment effect
More direct study specific-measure
e.g. Difference or ratio between means/
rates/ proportions
Measures magnitude without units
Allows comparison of effect across
studies
e.g.
𝜇1−𝜇2
𝜎
; Cohen’s effect size
14. Importance of Effect Size
Effect size too small larger sample size than necessary will
be required
Ethical issues, wastes resources
Effect size too large sample size won’t achieve target power
Can’t increase SS during trial, large risk trial will fail
Defines quantitative objective of study
Putting value on initial study question
15. Selecting Appropriate Effect Size
Select a clinically relevant difference
Some difference that would be important from a clinician’s or patient’s
perspective
Select a realistic difference
The difference you think is most likely to exist, based on prior evidence or
information
16. Methods to Determine Effect Size Value
Health Economic method
Systematic review of evidence
Elicit expert opinion
Standardized effect size
Pilot study
Distribution method
18. Overview & Pitfalls with Sample Size/Power
80/90% Power standard
90% gives “optimism” adjustment
90% = implicit 2-study adjustment
Some Sample size adjustments
Dropout, Unequal, CRT choices
Easier: N(D) = N/(1-P(Dropout))
Harder: Survival, Simulation, MNAR
For fixed sample size, more thought
in planning needed
19. Means Example
Parameter Value
Significance Level (Two-Sided) 0.05
Mean Difference 0.5
Standard Deviation 1.195
Dropout rate 10%
Power 80%
𝑛 =
(𝑍 𝛼
2
+ 𝑍 𝛽)2× 2𝜎2
(𝜇1 − 𝜇2)2
𝑛 = sample size per group before
dropout
𝑍 𝛼
2
= standard normal z-value
for a significance level α = 0.05,
which is 1.96
𝑍 𝛽 = standard normal z-value for
the power of 80%, which is 0.84.
𝑁𝑓𝑖𝑛𝑎𝑙 =
2𝑛
1 − 0.1
21. Sensitivity Analysis
Important for regulatory purposes
& peer-reviewed journals
Look at range of values for
parameters with uncertainty
Range based on clinically relevant
values
Assess how changes in parameters
affect sample size
22. Quick Overview:
Assurance for Clinical Trials
Assurance is the unconditional probability
of significance given a prior
Focus on methods proposed by O’Hagan et
al. (2005)
Assurance is the expectation of the power
averaged over a prior distribution for the
effect
Often framed as the “true probability of
success” or “Bayesian Power” of a trial
Can be considered as a Bayesian analogue
to sensitivity analysis
Source: O’Hagan (2005)
23. Assurance and Sensitivity Analysis
In a sensitivity analysis, a number of scenarios are chosen by
the researcher and assessed individually for power or N
Gives details of individual cases highlighted but no
information on other scenarios
With assurance, we have the average power over all
plausible values of the parameter
This provides a summary statistic for the effect of parameter
uncertainty but less information on specific scenarios
24. Means Assurance Example
“The outcome variable … is reduction in CRP
after four weeks relative to baseline, and the
principal analysis will be a one-sided test of
superiority at the 2.5% significance level. The
(two) population variance … is assumed to be …
equal to 0.0625. … the test is required to have
80% power to detect a treatment effect of 0.2,
leading to a proposed trial size of n1 = n2 = 25
patients … For the calculation of assurance, we
suppose that the elicitation of prior information
… gives the mean of 0.2 and variance of 0.0625.
If we assume a normal prior distribution, we
can compute assurances with m = 0.2, v = 0.06
… With n = 25, we find assurance = 0.595
Source: Wiley.com
Parameter Value
Significance Level (One-Sided) 0.025
Prior Mean Difference 0.2
Prior Difference Variance 0.06
Posterior Standard Deviation √0.0625=0.25
Sample Size per Group 25
25. Recap | 5 Essential Steps for Sample Size
1 Plan Study Study question, primary outcome, statistical method
2 Specify Parameters Significance Level, Standard deviation, ICC, dispersion
3 Choose Effect Size Expected/targeted difference, ratio or other effect size
4 Compute Sample Size Sample Size for specified metric such as power
5 Explore Uncertainty Sensitivity Analysis, Assurance, Alternative Designs
26. Join researchers from all over the world in using
nQuery for their sample size requirements!
Receive Regulatory Approval
Reduce Risk & Cost of Clinical Trials
Powerful Sample Size Options
Share & Empower Your Team
Use an Intuitive Sample Size & Power Calculator
28. References
O'Hagan, A., Stevens, J. W., & Campbell, M. J. (2005). Assurance in clinical trial
design. Pharmaceutical Statistics, 4(3), 187-201.
Joseph, L., & Bélisle, P. (1997). Bayesian Sample Size Determination for Normal Means and
Differences Between Normal Means. The Statistician, 209-226.
Sakpal, T.V. (2010). Sample Size Estimation in Clinical Trial. Perspectives in Clinical Research, 1(2), 67-
69.
Yao, J. C., Shah, M. H., Ito, T., Bohas, C. L., Wolin, E. M., Van Cutsem, E., ... & Tomassetti, P. (2011).
Everolimus for advanced pancreatic neuroendocrine tumors. New England Journal of Medicine,
364(6), 514-523.
Mease, P. J., Genovese, M. C., Greenwald, M. W., Ritchlin, C. T., Beaulieu, A. D., Deodhar, A., ... &
Nirula, A. (2014). Brodalumab, an anti-IL17RA monoclonal antibody, in psoriatic arthritis. New
England Journal of Medicine, 370(24), 2295-2306.
Editor's Notes
Point 1:
Know we have only 100 subjects available. Need to know what power will this give us, i.e. is there enough power to justify even doing the study.
Stage III clinical trials constitute 90% of trial costs, vital to reduce waste and ensure can fulfil goal.
Point 2:
http://rsos.royalsocietypublishing.org/content/1/3/140216 -> Screening problem analogy.
Type S Error = Sign Error i.e. sign of estimate is different than actual population value
Type M Error = Magnitude Error i.e. estimate is order of magnitude different than actual value
Point 3:
Sample Size requirements described in ICH Efficacy Guidelines 9: STATISTICAL PRINCIPLES FOR CLINICAL TRIALS
See FDA/NIH draft protocol template here: http://osp.od.nih.gov/sites/default/files/Protocol_Template_05Feb2016_508.pdf (Section 10.5)
Nature Statistical Checklist: http://www.nature.com/nature/authors/gta/Statistical_checklist.doc
Point 4:
In Cohen’s (1962) seminal power analysis of the journal of Abnormal and Social Psychology he concluded that over half of the published studies were insufficiently powered to result in statistical significance for the main hypothesis. Many journals (e.g. Nature) now require that authors submit power estimates for their studies.
Power/Sample size one of areas highlighted when discussing “crisis of reproducibility” (Ioannidis). Relatively easy fix compared to finding p-hacking etc.
More detail available on our website via a whitepaper.
More detail available on our website via a whitepaper.