Visual acuity in late adolescence and future psychosis risk in a cohort of 1 million men

Abstract

Background: We aimed to determine whether late adolescent visual impairment is associated with later psychosis. Methods: We conducted a longitudinal cohort study of Swedish male military conscripts aged 18 or 19 years from January 1, 1974, through December 31, 1997 (N = 1 140 710). At conscription, uncorrected and optometry-lens-corrected distance visual acuity was measured. Participants were then followed up to see if they received an inpatient diagnosis of non-affective psychotic disorder, including schizophrenia (N = 10 769). Multivariable Cox modeling was used to estimate differences between groups. Results: After adjustment for confounders, those with severe impairment before optical correction in their best eye (decimal fraction <0.3) had an increased psychosis rate compared to those with normal uncorrected vision (decimal fraction 1.0) (hazard ratio [HR] 1.26, 95% CI 1.16-1.37). Larger interocular visual acuity difference was associated with an increased psychosis rate (adjusted HR 1.49, 95% CI 1.37-1.63 in those with differences >0.5 compared to those with no between eye acuity difference). Individuals with impaired vision that could not be corrected to normal with lenses had highest rates of psychosis (best eye adjusted HR 1.56; 95% CI 1.33-1.82), those with imperfect, but correctable vision also had elevated rates (best eye adjusted HR 1.21; 95% CI 1.15-1.28). Individuals with visual impairment had higher rates of psychosis than their full siblings with normal vision (adjusted HR 1.20, 95% CI 1.07-1.35). Conclusions: Impaired visual acuity is associated with non-affective psychosis. Visual impairment as a phenotype in psychosis requires further consideration.