BACKGROUND: Several previous studies have established that miscarriage is a risk factor for depressive symptoms and disorder. By contrast, research on miscarriage as a possible risk factor for anxiety symptoms is inconclusive, and for anxiety disorders, sparse and uninformative. The current study examines the incidence of and relative risk for 3 DSM-III anxiety disorders (obsessive-compulsive disorder [OCD], panic disorder, and phobic disorders) within the 6 months following miscarriage. Adequate diagnostic data on other anxiety disorders were not available. METHOD: Using a cohort design, we tested whether women who miscarry are at increased risk for a first or recurrent episode of an anxiety disorder in the 6 months following loss. The miscarriage cohort consisted of women attending a medical center for spontaneous abortion (N = 229); the comparison group was a population-based cohort of women drawn from the community (N = 230). RESULTS: Among miscarrying women, 3.5% experienced a recurrent episode of OCD, compared with 0.4% of community women (relative risk [RR] = 8.0; 95% confidence interval [CI] = 1.0 to 63.7). The relative risk for noncomorbid panic disorder was substantial (RR = 3.6), albeit not statistically significant (95% CI = 0.8 to 17.2). There was no strong evidence for increased risk for phobic disorders or agoraphobia, combined or considered separately, in the 6 months following loss. Relative risk for all 3 disorders combined was 1.5 (95% CI = 0.9 to 2.3). CONCLUSION: In this first miscarriage cohort study using a concurrent frequency-matched comparison group, miscarriage was a substantial risk factor for an initial or recurrent episode of OCD. Given statistical power limitations of this investigation, the current findings do not preclude a possible contribution of miscarriage to risk for other anxiety disorders.