Back up a bit...
So what I think I'm reading, is that when law enforcement agencies receive a tip from a database, they then go out, conduct their own, independent, investigation, and use whatever evidence gleaned from that investigation as evidence in a court, while protecting the 'true' source of the tip. So, basically the same as an anonymous tip coming in from CrimeStoppers, or maybe a confidential informant telling the police about a drug dealer doing his thing, then the cops investigating and corroborating the information in the tip, then using the evidence from their own investigation in court. Ignoring the whole "monitor everyone for terrorism and using the data to bust druggies and stuff" deal that seems to be the "cool" thing to do nowadays, this actually sounds like normal, everyday police work. The Fruit of the Poisonous Tree suddenly seems a lot less likely here.
I think the main reason they do it this way is that, if the source (the database) is NOT protected in one instance of use by law enforcement agencies, then the ENTIRE database is then open to scrutiny by a defendant's defense team, and we can't have that.
Now, if documents are being altered to protect the source of the tip, or if people are making things up when they write the reports, then that's perjury (and probably other crimes). However, there's no evidence that I have seen yet to prove that this is being done.
A question I have is about how the information in this database is being obtained. How is the government finding out that Joe Blow is dealing drugs, before putting that fact in their database, then sending it to the relevant local agency(ies)?