Yes. The data sheet states: "Therefore, we can detect the original baseband signal by simply low-pass filtering the emitter current. The average emitter current is not very linear as a function of the RF input level. It exhibits a 1/ln response because of the exponentially rising nature of oscillator start-up. The steep slope of a logarithm near zero results in high sensitivity to small input signals."
So it would appear that at minimum a low-pass filter would be needed. (I have never examined the output of this module and know only what the data sheet and Dave state.) Obviously the "linear" and "digital" outputs have more than just "continuous noise output" or there would be no useful signal for subsequent use. A simple measurement with a VOM (depending on frequency response of VOM among other things) is in itself inadequate to determine if there is "a negligible difference between signal and no signal" if the signals were to be properly/adequately conditioned. But I have no pertinent hands-on experience with this module and can't speculate more.
I looked at Dave's web site for the first time in what seems like years. Looks like he has a wealth of new and old info and efforts. It is good to see that the ZX-40 (a BX-24 offshoot by a different company) has breathed new life into the BX24-AHT and Dave both.
Wonderful developments!
Happy New Year ... Marc Marc_F_Hult