[n8henrie]

I'd say it's fairly accurate. It's difficult to confidently establish "stable" in a patient with possibly feigned or exaggerated symptoms based on history and exam alone. It is generally agreed that pain management is part of managing an emergency, and so if an emergency has not been confidently excluded, it is usually part of care until a workup is concluded. If one could know from the initial evaluation that an emergency could not possibly be present, one could certainly discharge without additional care, but that can be impossible to know immediately (even with highly trained and calibrated eyeballs).