The serial interval, in the epidemiology of communicable (infectious) diseases, refers to the time between successive cases in a chain of transmission.
The serial interval is generally estimated from the interval between clinical onsets, in which case it is the 'clinical onset serial interval' when these quantities are observable. It could, in principle, be estimated by the time interval between infection and subsequent transmission. If the timing of infection transmission in deference to a person's clinical onset is TA, and the incubation period of a subsequent case is IB, then the clinical onset serial interval is calculated by TA + IB. Real data best reflects a frequency distribution when observing a single primary case and its associated secondary cases. If it is assumed that infections occur at random during the infectious period, then the average serial interval is the sum of the average latent period (from infection to infectiousness) and half the average infectious period.
Serial intervals can vary widely, and may be lifelong for some diseases (HIV infection, Chickenpox, Herpes). The serial interval for SARs was 7 days.
Related but distinct quantities include : the 'average transmission interval' sum of average latent and infectious period; the 'incubation period' between infection and disease onset; the 'latent period' between infection and infectiousness.