Nonmedical use of prescription medications is a major global health problem. To understand the extent of the problem and its trajectory, innovative surveillance methods need to be developed. Social media is a potentially useful resource for surveillance, but it presents many challenges from the perspective of natural language processing. In this extended abstract, we summarize the contents of an invited talk on the topic at the 6th International Workshop on Mining Actionable Insights from Social Networks (MAISON). Specifically, we present our ongoing efforts to leverage social media data, via natural language processing and machine learning methods, for obtaining insights about the nonmedical use of prescription medications.