Rev-erb-alpha is one of the key components of the mammalian circadian mechanism; recently, it was also reported to be involved in the biological action of lithium. We investigated whether polymorphisms in the Rev-erb-a gene are associated with the long-term efficacy of lithium carbonate therapy in bipolar affective disorder. Seven single nucleotide polymorphisms (SNPs) were genotyped in a well-characterized sample of patients from Sardinia, Italy, who were followed prospectively for up to 27 years. Genotypic and allelic analysis did not show evidence for association between the polymorphisms and the different levels of lithium response. Further analyses grouping the different levels of response demonstrated that when the patients were separated into groups of nonresponders versus individuals who have had at least a minor or modest improvement in frequency of episodes or admissions, there was a significant increase in the frequency of the T allele in the nonresponder group (p = 0.0008). Logistic regression analyses showed that patients carrying at least one copy of the T allele for the rs2314339 marker were shown to be approximately 3.5 times more likely to have no improvement or even a worsening of the illness (odds ratio [OR], 3.56; 95% confidence interval [CI], 1.18-10.76). The results of this study may help to identify potential biological markers that can serve to predict the response of bipolar affective disorder patients to treatment, improving treatment efficacy.