You may be referring to a certificate in the US. There are certificates that guarantee confidentiality to research participants (I believe through the NIH). I am aware of studies that have used this in the past with success.

There is a lot of debate about whether researchers are truly able to guarantee research confidentiality because research records can be subpoenaed. There are cases in Canada where researchers have managed to fight subpoenas and others where they haven't. There are also cases in the US, but I am less familiar with these. There are some scholars who have written on this matter (E.g., https://www.sfu.ca/~palys/ProtectingResearchConfidentiality.pdf). The SFU case is an interesting one and the researchers involved in that have written a really interesting book on research confidentiality.

One way to protect research participants is to ensure that they are anonymous. In my experience, the REB can still be uncomfortable when you ask about things that are illegal in an anonymous format, particularly ones where there may be a mandatory reporting requirement. The way to address this is to only ask about detected illegal behaviours. As you rightfully point out though, you don't get a true picture of whatever it is you are examining.

Long story short, getting research through REBs is not easy at times depending on what you are researching. There is a lot of discussion that happens with the researchers and the board, particularly if the REB sees your research going beyond minimal risk.