The sample is unbalanced in virtually every analysis, due to non-response/loss to follow-up, etc. BUT if the unbalance is more than minimal, it reduces power. The question is not, can it be processed -- obviously it can since virtually all samples are unbalanced -- but rather, how to adjust for biases the unbalance might introduce. If the unbalance is simply a matter of a larger sample in one group than the other, the weighting will reflect that. As you code the unbalanced sample design into your statistical software, it automatically will account for the imbalance when it computes standard errors and tests for differences. BUT with an unbalanced sample, your ability to detect differences between groups is large. With a deliberately unbalanced sample, you should compute a power curve and may need to present the minimum size difference you had the power to detect.

Tq mr Ted R Miller for the your explanation.

I suggest that you read the article titled by " Impact of Sample Size and Variability on the Power and Type I Error Rates of Equivalence Tests: A Simulation Study". I hope this article help you Noor adawiyah ahmad radzi