Dental Research Example

Dental Research Example

Sample Size for Equivalence of Proportions

What sample size is required for a study comparing a new type of dental implant to a reference standard?

Outcome variables are 5 year implant success rate, pocket depth, gingival health scores and rate of bone loss. nQuery Advisor® can provide sample size calculations for tests or equivalence tests on 5 year success rate proportions, pocket depth, health scores, and bone loss rates, and for tests comparing 5 year implant survival curves. nQuery Advisor® can even provide assistance with planning sample sizes for cluster sampling designs in which it is planned to study multiple implants per person.

Select File … New, and in the Study Goal and Design Box, select Proportions, Two Groups, and Equivalence. You can specify the sample sizes for each group and solve for power or specify power and solve for the sample size per group.

If the new implant type is defined to be equivalent to the reference standard if 5 year success rates do not differ by more than 5%, 617 subjects per group would be required to have 90% power to reject the null hypothesis of non-equivalence if both success rates were truly 90%. If both success rates were truly 95% only 326 subjects per group would be required. Choosing 446 subjects per group for example would provide 80% power if both rates were 90% and 96% power if both rates were 95%.

nQuery Advisor® provides tables, plots, and standardized sample size justification statements which can be printed directly or copied to clipboard and pasted into your manuscript or proposal.

“When the sample size in each group is 617, a two-group large-sample normal approximation test of proportions with a one-sided 0.050 significance level will have 90% power to reject the null hypothesis that the test and the standard are not equivalent (the difference in proportions, πT – πS, is -0.050 or farther from zero in the same direction) in favor of the alternative hypothesis that the proportions in the two groups are equivalent, assuming that the expected difference in proportions is 0.000 and the proportion in the standard group is 0.900.”