An experimental test of the influence of prior cigarette smoking abstinence on future abstinence.

Significant positive associations between early and later abstinence are often reported in clinical trials on smoking cessation, but those relationships do not permit causal inferences. The present study was conducted to examine experimentally whether prior smoking abstinence histories can directly facilitate later abstinence, using a contingency management procedure to manipulate prior abstinence. A total of 40 adult cigarette smokers who were not trying to quit were randomly assigned to one of two conditions: Contingent ALL (C-ALL), who earned monetary incentives contingent on smoking abstinence during three 5 day experimental periods; or Contingent LAST (C-LAST), who earned incentives independent of abstinence during the first two periods (i.e., noncontingent payments) and contingent on abstinence during the final period. The contingency management procedure was effective in establishing different abstinence histories in the two conditions during the first two periods. Comparison of abstinence levels between the C-ALL and C-LAST conditions during the third period showed significantly greater abstinence in the C-ALL condition, although nicotine withdrawal and other subject ratings generally did not differ significantly between the two conditions. These results provide experimental evidence that prior abstinence histories can directly influence subsequent efforts to abstain from smoking.