This work is concerned with the stability analysis of an output feedback control system possibly influenced by unknown disturbances, where both the plant output and the controller output are subject to event-triggered sampling. We propose a new event-triggering mechanism based on the history of the measured outputs instead of the current outputs only. This novel feature provides a simple link between the parameters of the sampling criterion and the speed of convergence. Accumulation of sampling times is prevented by enforcing a minimum interevent time. The effectiveness of the proposed event-triggered scheme is illustrated by several numerical examples, including nonlinear and linear systems.