That document is a pdf, are your users really trying to scan into it? The pages will be added as additional image pages but the pdf won't be modified, which could be a confusing situation. As for why it looks like that, it seems like the system isn't finding images for those pages and is trying to render some text that it found. I could see that happening if their scanner is producing a pdf and treating it like single page data.

I'd take a step back and look at their capture procedure, and make sure they are doing sensible things.