In order to assess the effect of long-term versus short-term intravesical chemotherapy in preventing the recurrence of patients with non-muscle-invasive bladder cancer, we searched several databases with words as mesh terms and free text words to find all eligible randomized clinical trials (RCTs) for the comparison of the two strategies of instillation durations. "Observed-Expected events research (O-E)" and "Variance (V)" for calculating hazard ratio (HR) were used in Revman 5.2 software recommended by Cochrane Collabration for data analysis. Sensitivity and subgroup analysis were selected to minish heterogeneity. GRADEpro 3.6 profile recommended by Cochrane Collabration was employed for quality assessment of analyses. Finally, 13 eligible RCTs with 4216 patients were included in this review and 16 comparisons from 13 trials were involved for analysis. The pooled analysis revealed no significant difference between long-term and short-term duration [HR=0.99, 95% CI (0.89, 1.11), P=0.89]. Within the subgroup analysis, patients benefited from long-term instillations with a start regimen of one immediate instillation [HR=0.83, 95% CI (0.69, 1.00), P=0.05]. But patients were not suitable to receive long-term instillations with epirubicin (EPI) [HR=1.01, 95% CI (0.91, 1.13), P=0.78]. The progression rate was not reduced after long-term instillations [HR=0.96, 95% CI (0.66, 1.39), P=0.82]. From our results, patients should not receive introvesical chemotherapy more than half a year. In contrast, patients with one immediate instillation are preferred to have a long-term duration at least one year. Long-term instillations can not reduce the progression rate.