Tobacco has negative public health consequences. In this study we aimed to determine the prevalence of and factors associated with using more than one form of tobacco among people from a low socioeconomic background in Bangladesh in order to inform tobacco control and prevention programs for this population. Inclusion criteria for study subjects were current tobacco users, aged ≥ 15 years who were able and willing to participate. Study subjects were selected randomly from two industrial areas. An enumeration survey was conducted prior to actual data collection to identify the low income population by sex. Exclusion criteria for study subjects were age >65 years and being a temporary migrant. The number of subjects calculated to be needed for the study was 460. Each subject was asked to complete a semi-structured questionnaire that asked about sociodemographic characteristics, use of tobacco and types and potential associated factors. Descriptive statistics were used to summarize sociodemographic characteristics, a chi-square test and Fisher's exact test were used to determine associations between selected factors and tobacco use and logistic regression analysis was used to identify factors associated with using more than one form of tobacco. A total 460 subjects were included in this study 50% were males. The mean age of study subjects was 43.3 years. The most common occupation was housewives followed by service. Of the 460 subjects, 23% used more than one form of tobacco: 97.1% of them male subjects and 2.9% of female subjects. Use of more than one form of tobacco was significantly (p < 0.049) more common among subjects aged <30 years. Significant associations were found between using more than one form of tobacco and no formal education level (p=0.009), marital status (p < 0.031), family history of tobacco use (p < 0.031) and monthly income (p < 0.001). Logistic regression analysis revealed a significant association between being aged <30 years and using more than one form of tobacco [odds ratio (OR)=4.2; 95% confidence interval (CI): 1.53-11.54; p=0.020]. Logistic regression analysis also showed a significance association between having a tobacco related disease and use of more than one form of tobacco (OR=1.7; 95% CI: 1.03-2.82; p=0.040). Younger subjects and those with tobacco associated diseases are more likely to use more than one form of tobacco. Tobacco control programs targeting these groups need to keep this in mind and focus on more than just smoking cessation in this group.