Hypnosis in Treatment of Alcohol Addiction; Controlled Trial, with Analysis of Factors Affecting Outcome

Abstract

Forty alcoholics were randomly allocated to one of two groups; one group received full conventional inpatient treatment while the other received hypnotic treatment in addition to the conventional therapy. The hypnotic technique used was one of simple eye fixation followed by posthypnotic suggestion for 6 days and reinforcement at weekly, then monthly, intervals. The two groups were satisfactorily matched for age, years of addiction, intelligence, neuroticism (n) and extra version (e) scores on the Maudsley Personality Inventory, and social stability (ss) score on the Straus—Bacon scale. The patients were rated on sociomedical outcome on a 2-point scale for each of the 12 months following discharge; and an over-all outcome score for the year was then calculated. Statistical analyses of the scores were done and prognostic characteristics of the two groups were compared. All patients were asked to take disulfiram regularly and to attend Alcoholics Anonymous; few continued to do so after discharge from hospital. All received individual psychotherapy, and social rehabilitation and social casework were available to all as required. Monthly outcome scores for the two groups were similar, and in each group the majority of relapses occurred in the first 6 months. When over-all outcome for the year was examined, the two groups were remarkably similar, with 11 patients in each group in a good, 2 in a fair and 7 in a bad prognosis category. In both the hypnotic and control groups, the patients in the bad prognosis category had significantly lower ss scores than those in the good prognosis category (means of 1.0, 0.57, 3.45 and 3.45, respectively). The patients' age, length of addictive drinking and intelligence were not related to outcome. A number of factors which might have been associated with prognosis were then analyzed statistically for the two groups taken as a whole, and it was found that n had a significantly negative correlation with outcome (p<.05), E, a significantly positive correlation with outcome (p<.05) and ss, a highly significant positive correlation with outcome (p<.001). E andnwere negatively correlated and ss was negatively correlated withnand positively correlated with e. Multiple regression equations showed that outcome score could with little error be calculated from ss alone. Length of stay in hospital appeared to be inversely related to outcome; in the control group, those with good prognosis stayed a significantly shorter time. The results are discussed in the light of a review of the literature on hypnotic treatment of alcoholism, and the contention that a different mode of hypnotherapy might be more effective is examined. No factual evidence to contradict the present finding that hypnosis is of no value as an adjunct to conventional treatment of alcoholism was found. The meaning of social stability is discussed and the question of its relationship to personality factors is considered. Finally, the bearing of these results on the planning of treatment and the problem of devising better methods of treating the alcoholic with low social stability are discussed.