Purpose To analyze claustrophobia during magnetic resonance (MR) imaging and to explore the potential of the 26-item claustrophobia questionnaire (CLQ) (range, 0-4) as a screening tool in patients scheduled for MR imaging. Materials and Methods The study received institutional review board approval, and patients in the CLQ cohort provided informed consent. A total of 6520 consecutive patients were included. Overall, 4288 patients completed the CLQ before MR imaging (CLQ cohort), and 2232 patients underwent MR imaging without having completed the CLQ (non-CLQ cohort). Claustrophobic events were recorded by the staff. Results The CLQ mean score in patients with claustrophobic events (1.48 ± 0.93) was significantly higher (P < .01) than in the group without claustrophobic events (0.60 ± 0.5). The CLQ cutoff value was 0.16 for men and 0.56 for women. Because of the low prevalence, negative predictive values of CLQ cutoff values (men, 0.99 [573 of 582]; women, 0.97 [745 of 766]) were higher than positive predictive values (men, 0.01 [88 of 582]; women, 0.16 [192 of 1186]). The overall claustrophobic event rate was 9.8% (640 of 6520; 95% confidence interval [CI]: 9.1, 10.6). The CLQ did not induce claustrophobic events because the event rate in the CLQ cohort was significantly lower than that in the non-CLQ cohort, as shown by the adjusted odds ratio of 0.81 (95% CI: 0.68, 0.96). Conclusion The CLQ is a suitable screening tool for the absence of a subsequent claustrophobic event. Furthermore, while it is possible to identify patients with a considerable risk of claustrophobia, prediction in individual patients is not possible. © RSNA, 2016 Online supplemental material is available for this article.