Several previous studies have investigated an association between leukocyte telomere length (LTL) and schizophrenia (SCZ). However, results have been largely inconsistent, partially due to the relatively small sample sizes in each study and heterogeneity caused by various uncontrolled confounders (e.g., duration of illness or hospitalization, lifetime antipsychotic dose, and LTL assay methods). Here, we investigate the association of LTL with SCZ with the quantitative polymerase chain reaction method in independent cohorts consisting of 1241 patients with SCZ and 1042 controls (the largest independent sample in this field). Furthermore, we examined whether duration of hospitalization and lifetime antipsychotic dose had an effect on LTL in SCZ. In all samples, we observed significantly longer LTL in patients with SCZ relative to controls. In subgroup analyses, we observed that longer telomeres in SCZ were only visible in elderly patients and not in patients under 50 years old. Moreover, significantly longer LTL in elderly patients with SCZ was only specific to those with long-term hospitalization, but not outpatients or those with short-term hospitalization. This may be because the former received more appropriate lifestyle management. Meanwhile, lifetime antipsychotic dose had no effect on LTL. Our findings suggest that consideration of the effect of age and duration of hospitalization on LTL may improve our understanding of controversial results obtained in previous studies of telomeres in SCZ.