The information derived from social media analytic studies provides valuable sources of information for healthcare stakeholders. However, there is still a lack of research with using social media to identify the lifestyle choices of those dealing with diabetes in order to better understand and design impactful health interventions before an extremity like death occurs due to diabetes. This exploratory study aims to demonstrate how social media can be leveraged as a data source to help us understand the lifestyle choices of those dealing with diabetes. Using two text mining approaches - sentiment analysis and unsupervised topic modeling - food and physiology were topics expressed in both sentiments. Overall, lifestyle related topics accounted for nearly 25% of the topics identified in the corpus of data. There is a pressing need for incorporating predictive modelling approaches to this study in order to quantify our findings and how this knowledge can improve health outcomes from a population perspective.