Can "Work Horse Killed Handler" handler by used when ReplicaSets are scaled down/termination process?

[dszortyka]

We have a process with multiple workers connected to different queues. We use those Queue metrics to scale Pods in kubernetes.
If a queue goes over 100, spin a new POD which will be consuming the queues.
When a scale down happens, we would like to add a process that runs when shutdown/termination was requested.
Would it be possible to use Work Horse Killed Handler for such scenarios? We are suspecting that PODs are being terminated while records from those queues are actually being processed.