If I have a design that compares the effectiveness of two alternatives, would it be acceptable to include usual care as a third arm?

Yes. A third (control) arm is acceptable, provided you give appropriate justification and measure the care received. You must carefully describe which elements of usual care that you will measure in patients randomly assigned to the usual care arm. As always, you should provide a strong methodological rationale for the specific study design(s) you are proposing.