Changes in alcohol expectancies during treatment relate to subsequent abstinence survivorship.

When negative alcohol expectancies are measured appropriately they form at least as secure associations with measures of consumption as has been demonstrated by mainstream expectancy research for positive alcohol expectancies and they can be usefully used to represent a component of motivation to restrain consumption or recover in dependent drinkers. A study is reported in which (i) negative outcome expectancies assessed at admission to treatment reliably predicted number of days to first drink; (ii) the same relationship is discovered for discharge measures (iii) and, although the change is negative expectancies between admission and discharge does not, itself, predict the number of days to first drink, it does when the corresponding admission measure is also taken into account. The same predictive relationships were not found for positive expectancies. Implications for planning treatment are discussed in terms of treatment enhancement rather than treatment matching.