BACKGROUND: Positive mental health (PMH) is much more than the absence of mental illnesses. For example, PMH explains that to be happy or resilient can drive us to live a full life, giving us a perception of well-being and robustness against everyday problems. Moreover, PMH can help people to avoid risky behaviours like tobacco consumption (TC). Our hypothesis was that PMH is negatively associated with TC, and this association differs across rural, urban and migrant populations. METHODS: A cross-sectional study was conducted using the PERU MIGRANT Study's dataset, including rural population from the Peruvian highlands (n = 201), urban population from the capital city Lima (n = 199) and migrants who were born in highlands but had to migrated because of terrorism (n = 589). We used an adapted version of the 12-item Global Health Questionnaire to measure PMH. The outcome was TC, measured as lifetime and recent TC. Log-Poisson robust regression, performed with a Maximum Likelihood method, was used to estimate crude prevalence ratios (PR) and 95 % confidence intervals (95%CI), adjusted by sex, age, family income and education which were the confounders. The modelling procedure included the use of LR Test, Akaike information criteria (AIC) and Bayesian information criteria (BIC). RESULTS: Cumulative occurrence of tobacco use (lifetime TC) was 61.7 % in the rural group, 78 % in the urban group and 76.2 % in rural-to-urban migrants. Recent TC was 35.3 % in the rural group, 30.7 % in the urban group and 20.5 % in rural-to-urban migrants. After adjusting for confounders, there was evidence of a negative association between PMH and lifetime TC in the rural group (PR = 0.93; 95%CI: 0.87-0.99), and a positive association between PMH and recent TC in migrants (PR = 1.1; 95%CI: 1.0-1.3). CONCLUSIONS: PMH was negatively associated with TC in rural participants only. Urbans exhibited just a similar trend, while migrants exhibited the opposite one. This evidence represents the first step in the route of knowing the potential of PMH for fighting against TC. For rural populations, this study supplies new information that could support decisions about prevention programmes and psychotherapy for smoking cessation. However, more research in the topic is needed.