What is the best equivalent to the Parent-Child Trick for connected tables?

I need to get data from a system where they are having a tough time creating properly structured CSV exports.  Specifically, they have questionnaire data where each question response in each questionnaire is a record in a table with a unique ID and a foreign key to another table where each questionnaire's metadata is located.  For expediency, I glibly told them to simply provide the exports as is from all the tables and I would manage it in Quick Base.  In the back of my mind, I was thinking I could use a version of the parent-child trick to create a reverse relationship and pull the question data up to the te level of the questionnaire myself. 

So, I got the data, created connected tables, and made relationships using the questionnaire foreign key on the questions. Annoyingly, Quick Base doesn't provide an option to make the "ID" field numeric instead of text when creating the connected table.  So, I had to create a formula numeric version of that field just to create a minimum ID# summarized field on the parent when the questions text is, "Question 1," "Question 2," etc.  But that worked fine.  So far so good. I then went to create the reverse relationship with this new field in hand and DREAD - that field is not available to use as a reference field!  It seems that only the original connected fields can be used to create a relationship.

So, any ideas of what the best equivalent would be here?  Do I need to create some automations or something in Pipelines to replicate the connected table so that I can use some of these created fields as reference fields?  Given my situation, any other ideas about how to join these data or do I go back to them and say they need to find a way to join these data in the source system?  Thanks in advance for any help.

------------------------------
Tate Forgey
------------------------------