Jitter is the variation in the timing of packet arrival on a network. Audio and video packets are sent at a steady rhythm, but after crossing real networks — congested home Wi-Fi, mobile carriers, hospital uplinks — they arrive bunched together and then with gaps, rather than evenly spaced. That unevenness, not the average delay, is what jitter measures.
Receivers absorb jitter with a jitter buffer, which deliberately holds incoming packets for a few tens of milliseconds so they can be reordered and played out smoothly. This trades a small amount of added delay for steady playback. The buffer can hide a certain amount of timing variation, but when jitter exceeds what the buffer can compensate for, the result is audible: audio crackles and drops out, and video stutters or freezes — exactly the degradation that makes a clinician ask a patient to repeat themselves or fails to show a visible symptom clearly.
For a telemedicine product, the practical implication is to monitor jitter per call, alongside packet loss and round-trip time (RTT), as a core quality signal. Rising jitter is frequently the earliest indicator that a connection is starting to degrade, often before the participants consciously notice, which makes it valuable for proactive interventions like lowering video quality or warning the user. The common pitfall is tracking only average bandwidth or a single "connection quality" bar and missing jitter entirely, then being unable to explain why a call with seemingly adequate bandwidth still sounded broken.
