There are a couple of ways you might be able to do this, but the simplest one is probably to use the %NOT MDX function. You can write a filter that excludes one member by appending .%NOT to the member spec:

%FILTER [Demographics].[H10].[Citizenship].&[USA].%NOT

If you want to write a filter that excludes multiple individual members of one or more levels, you can filter by a tuple of the members that you want to exclude. Append .%NOT to each of the member specs, and do not use %OR:

%FILTER ([Demographics].[H10].[Citizenship].&[USA].%NOT,[Demographics].[H10].[Citizenship].&[U.S.A].%NOT,[Demographics].[H10].[Citizenship].&[US].%NOT)

I'd like some clarification on the prompt "Using %MDX and %KPI instead of Subject Area in IRIS Analytics". I don't typically think of %MDX and %KPI as alternatives to creating/using a subject area in the way that the prompt implies - is the intention just to have an article that explains how to use %MDX and %KPI, or is there a specific use case that you are hoping it will explain?
edit: Doesn't look like I'm eligible to participate, but maybe someone else wants to write this...

Hi David,

When you selected the "Sort members" option in the pivot table, did you do it from the gear icon next to the "Rows" header, or from the one next to the specific level that you had put on rows? On IRIS 2019.1, I'm finding that the sort order remains when I drill down if I do the former, but not if I do the latter.