Using the Bonferroni approach which advocates dividing the P-value obtained from each test by the number of tests performed and then comparing this modified P-value to the conventional significance level of 0.05 to determine significance.
Using a more stringent significance level for each test (e.g. 0.01 instead of 0.05).
Only performing a subgroup analysis if the test for an interaction between the treatment and the factor defining the subgroups produces a significant result.
Choosing a higher significance level for each interim analysis in a trial to ensure that the required overall significance level (typically 0.05) is maintained.
Combining multiple endpoints into a composite endpoint.