OBJECTIVES: Antidepressants are one of the pharmacologic agents that have been intensively prescribed off-label.  The study aimed to investigate trends of off-label antidepressant uses for insomnia treatment in Medicare population, with emphasis on patient and physician’s socioeconomic characteristics. METHODS: This project proposed a secondary data analysis using National Ambulatory Medical Care Survey 2002 to 2009.  Subjects were derived from US outpatient visits in which at least one insomnia drug prescribed to Medicare beneficiary.  Data trends were graphically plotted and analyzed using weight-adjusted chi-squared analyses. A weighted logistic regression with SAS PROC SURVEYLOGISTICS and ROC curve were conducted to predict the maximum likelihood of off-label antidepressant use associated with patient and physician characteristics.   RESULTS: An estimated of 7.56 billion visits occurred to office based physician in the US from 2002 to 2009, in which 1.78 billion visits (23.52%) were made by Medicare recipients.  There were 82.76 million (4.65%) visits, where at least one commonly used insomnia drugs was prescribed.  Among these, FDA-approved benzodiazepines were prescribed at 11 million visits (13.3%), FDA-approved non-benzodiazepine z-hypnotics were at 35 million visits (31.53%), and off label antidepressants were at 40 million visits (48.61%).   There was significantly increased likelihood of prescribing off-label antidepressants in relationship to patient’s year of visit, gender, age, and physician’s specialty and ownership (P<0.001).  The receiver operational characteristic analysis estimated the ROC curve of predictor variables was 0.779.  CONCLUSIONS: The study revealed high prevalent use of off-label antidepressants for insomnia in Medicare population. Our results also showed various socioeconomics disparities among physician and patients in use of antidepressant for insomnia.   These lead to great concern of potential legal risk of medication errors, side effects, and adverse drug reactions from the unregulated off-label prescriptions.