Background: The integrity of treatment is crucial to the interpretation of treatment’s efficacy. However, in many studies involving counseling technique, the integrity of the technique used is questionable. Motivational Interviewing (MI) is a relatively new counseling technique commonly used in the field of addiction. Objectives: To study the effect of treatment integrity on the interpretation of its efficacy, using MI on smoking cessation as an example. Methods: A comprehensive search was done on studies about the use of MI on ‘‘smoking cessation’’. This study was limited to RCTs and those trials with explicit reference to the counseling method MI as innovated by Miller & Rollnick. The papers were scrutinized according to a published Cochrane SR protocol on the same topic (by the same author). The following indicators were noted: credentials of the treatment providers, the specific training procedures (for performing MI e.g. focused workshops), the exact nature of MI (different aspects of the counseling e.g. the use of decision balance technique, problem feedback), the total duration of treatment, the use of supplementary technique (e.g. the use of self help booklet), whether the treatment provider were administering the treatment as directed (e.g. was there supervision of the counseling process, videotaping). Results: Preliminary analysis indicates that the MI is efficacious for smoking cessation with pooled odds ratio 1.61 (95% CI 1.02 to 2.53). However, there was wide variation in the integrity of treatment used. The exact training given to the treatment provider was not mentioned in about half of the study. The essential elements of the counseling were also not adequately described. Methods to ensure fidelity of counseling were either not described or inadequately described in most papers. In a sensitivity analysis, if the papers with inadequate description were removed, the efficacy of MI on smoking cessation became insignificant. Conclusions: It is difficult to confidently conclude the efficacy of a treatment modality when there is large variation in the integrity of the treatment studied. A standardized and clear description of the counseling method and procedures used are important in future studies.