We examined the occurrence and 1-year course of mixed anxiety-depressive disorder (MAD) in a sample of primary care patients. Participants are part of the Primary Care Anxiety Project, a naturalistic, longitudinal study of anxiety disorders in primary care. Participants completed a questionnaire screening for anxiety symptoms. Those screening positive were invited for an interview to diagnose MAD and DSM-IV Axis I disorders. Participants were then interviewed at 6 and 12 months postintake. Of 1634 participants completing an intake interview, four participants (0.2%) met complete DSM-IV MAD criteria. The adjusted probability of remitting from MAD in 1 year was 80%. Although this was not a prevalence study, results indicate a very low occurrence of MAD across 15 primary care settings. Further, they indicate that this diagnosis may not be stable across time and raise doubts about its utility.