Objective: We know little about the short-term course of drinking, particul
arly the stability or instability of at-risk drinking in untreated drinkers
. Because few at-risk drinkers obtain help for their drinking, it is import
ant to understand the short-term fluctuations between at-risk drinking and
full-fledged alcohol use disorders, as well as remission of at-risk drinkin
g. Method: We used four waves of data (each 6 months apart) from a probabil
ity community sample of 733 at-risk drinkers in six states in the southern
United States to determine variation in abstinence, drinking patterns and a
lcohol use disorders over a 2-year period. For this analysis, we excluded t
hose who reported receiving services for drinking during the 2-year study p
eriod (retrospectively at baseline), leaving a sample size of 664 (444 male
); 479 (306 male) completed all four inter-views. Results: Although the maj
ority (88%) of the sample was nonabstinent throughout the study, we found s
ignificant decreases in average number of drinks per drinking day and recen
t (past 6 months) alcohol disorders, and an increase in 6-month abstinence,
Almost 30% of those with no recent alcohol disorder at baseline (n = 280)
later met diagnostic criteria in at least one interview. Of those with a re
cent alcohol disorder at baseline (n = 199), one third met criteria in at l
east two subsequent interviews. Conclusions: There is some evidence for sho
rt-term progression from at-risk drinking to alcohol disorder. However, the
re is stronger evidence for declining problems and a fluctuation in and out
of recovery and relapse, which may reflect an effort to maintain controlle
d drinking, Understanding this short-term course is important for primary a
nd secondary prevention efforts and for screening of at-risk drinking in pr
imary care and in the workplace.