The debate about the effects and public health importance of the way Aborig
inal people drink kava has been confounded by claims, based on anecdote, of
imputed health effects. Anecdote and comment have promoted the perception
that dosage levels among Aboriginal people are much greater than in Pacific
island societies. In this paper we review published data about kava consum
ption, and evaluate it with respect to information collected from observati
on of one Aboriginal community in Arnhem Land (Northern Territory) where pe
ople tend to consume kava at a steady tempo; 37g of kava powder containing
around 3800mg of kava lactones in 670ml of water in an hour. The highest le
vels of consumption in Arnhem Land have been reported to be up to 900g/week
of kava powder with heavy consumers drinking at least 610g/week, levels co
mparable to estimates for Pacific-island societies. The significance of a s
teady drinking tempo means that an individual's weekly kava consumption rel
ates directly to the amount of time spent drinking which, in turn, is corre
lated with categories of social setting of drinking (p < 0.0002). Lone drin
kers appear to be the heaviest users while lowest consumption takes place i
n private domestic situations, where people enjoy kava as part of family gr
oup activities. Surrogates of consumption levels may be found in local soci
o-economic circumstances. This approach may be useful when more direct meas
urement of consumption is difficult or impossible.