Objective To investigate whether anxiety plays a role in self-recruitment for non-population-based glaucoma screening. Methods In a non-population-based pre-publicised trial, self-recruited Caucasian participants were screened for glaucoma, and also completed the Trait Anxiety Inventory and Shortened Health Anxiety Inventory questionnaires. In pre-publicity for the trial, information on risk factors for glaucoma was given. Participants classified as possible glaucoma cases later underwent a detailed glaucoma investigation. Results Of the 120 total participants (72 females, 48 males), 12 were considered glaucoma suspects at the screening, although only three (2.5%) were ultimately diagnosed with glaucoma. Health anxiety showed significant correlation with trait anxiety (r ¼ 0.525, Po0.001). Trait anxiety was similar for both the genders; this score was significantly below the normal Hungarian population value in women (Po0.001) and at the normal population level in men (P ¼ 0.560). In contrast, health anxiety was significantly higher than for both the 'normal' and 'anxious' reference groups (Po0.001), although smaller than that for hypochondriacs (Po0.001). Participants with pre-existing ocular symptoms, and those who attended because of fear of blindness, had significantly higher trait-and health-anxiety scores (Po0.05 for all comparisons). Conclusion In this screening trial, the health anxiety of the self-recruited participants was significantly above normal, whereas the prevalence of glaucoma was within the usual range for a Caucasian population. This suggests that providing pre-publicity information on risk factors for glaucoma does not necessarily increase the prevalence of glaucoma among self-recruited participants in non-population-based screening, as some individuals may participate on account of elevated health anxiety, rather than because of higher potential risk for glaucoma.