We appreciate the insightful commentary from Drs Mysliwiec and Roth on our article (September 2012),1 and agree that many patients in our study potentially had insomnia that was refractory to pharmacotherapy. Given the retrospective nature of this study and complex nature of the included patients, the evaluation of insomnia was challenging. We acknowledge that our categorization of these patients potentially led to an underestimation of the rate of insomnia in our population. This was, in part, by intent, as we felt that using more stringent diagnostic criteria provided a more realistic assessment of our population. Sleep fragmentation and difficulties initiating and maintaining sleep were nearly universal in our population. By nature of the cohort, several had chronic pain, anxiety, psychiatric disease, polypharmacy, and underlying sleep-disordered breathing that could contribute to their sleep disruptions. Given that most insomnia is comorbid, it would have been reasonable for us to report a much higher prevalence in our cohort. However, we chose to use a narrow definition that reflected those we felt had true insomnia and not just a manifestation of some other, more likely factor (ie, subjective sleep latency ≥30 min during the majority of nights or subjective sleep fragmentation associated with daytime impairment not better explained by sleep disordered breathing, pain, or other more identifiable factors).1