A number of prophylactic vaccines against human immunodeficiency virus
(HIV) have passed through phase I clinical trials, and phase II clini
cal trials are now being planned. These vaccines are not expected to b
e perfect and might fail in a number of different ways. This paper sho
ws how to equate different aspects of imperfection in a prophylactic v
accine in terms of impact upon levels of herd immunity, and hence upon
the vaccine coverage required for eradication. Such comparisons revea
l that an otherwise perfect vaccine that gives protection which wanes
with a half-life of 10 years is only as good as a vaccine that works i
n 30% of people giving them complete, lifelong protection. The paper g
oes on to compare predicted patterns of seroconversion that would be o
bserved in clinical trials and in community-wide vaccination campaigns
for vaccines that confer the same levels of herd immunity but are imp
erfect in different ways.