I'm very surprised that Animas would attempt to develop their own proprietory protocols. Security issues in wireless protocols are well understood and extremely well studied, but many people who are otherwise extremely capable have poor intuitions about security. The *well published* protocols such as bluetooth have, as a result of being well published, received expert scrutiny necessary for them to be truely secure.
Suppose Animas said, "We use a bluetooth, a well published protocol, and we use it with the highest security available. Once our devices have been paired, which always happens under the control of the user, we are confident our devices are secure." In that case I would not be writing this.
Animas's response, however, is seriously flawed; "to prevent tampering... Animas cannot share the specific details of these protocols." The only interpretation of that statement is that if someone knew the details the protocol would be compromised. That's not true of bluetooth (for example.)
Other details of their response create the distinct impression that they don't know what they are doing; the response implies that security relies on the serial number of the device being secret, that the term "extremely difficult" might impress people who do extremely difficult things as a matter of course (e.g. flatlining BG) and that their adherence to HIPAA somehow magically grants computer system security to their products.