After a brief look at pubmed I was able to some good supporting evidence, showing significant, moderate (~1.0 cohen effect size) differences between groups using SRS vs other learning techniques.

Is anyone who is more familiar with the research able to provide a summary - it's not my field, but I am very interested. Also, why isn't this method used in schools for homework etc - it would seem a much better if it's been proven.

It isn't used because it requires a lot more work from the teacher. Notice this was a one time college level course from a GA.

Surely the creation of course materials is a one time event with a small maintenance element. And every child in the world follows roughly the same curriculum, only the rates change. So it would follow that these should be available free from the government right?

I use SRS (mnemosyne) with my kids for learning and it's awesome. It's faster than doing "homework" because it automatically adjusts to what they need to learn.

Notice that the comment included a lot of adaptation for feedback from the question and answer period, too. And the various "flashcard" systems have the same feedback, but it's automated into the system; in a class you don't have that automation.