Meta-analyses have shown modest positive associations between diabetes mellitus (DM) and bladder cancer risk, but results are heterogeneous. This might be due to lack of distinction between bladder cancer subtypes, between sexes, and possibly between Type 2 and Type 1 DM (T2DM and T1DM). The relationship of T2DM (and secondarily T1DM) characteristics with risk of bladder cancer subtypes (invasive versus noninvasive) was investigated in the Netherlands Cohort Study. In 1986, 120,852 men and women aged 55-69 years provided information on DM and lifestyle data. After 20.3 years of follow-up, multivariable case-cohort analyses were based on 1020 invasive and 1088 noninvasive bladder cancer cases, and 4267 subcohort members with complete data on DM and confounders. While T2DM was not associated with noninvasive bladder cancer, it was statistically significantly associated with invasive bladder cancer risk: the multivariable-adjusted was HR = 1.57 (95% CI 1.04-2.37), comparing participants with T2DM versus without DM. The association was only significant in women, and women showed a stronger association [HR = 2.19 (95% CI 1.10-4.34)] between T2DM and invasive bladder cancer than men [HR = 1.42 (95% CI 0.88-2.30)]; interaction by sex was nonsignificant. Associations were stronger positive in those whose age at diagnosis of T2DM was 55+ years, and in those diagnosed with T2DM less than five years before baseline. T2DM participants using antidiabetic medication had higher invasive bladder cancer risk than those without DM. Exploratory age-sex-adjusted analyses suggested a positive association between T1DM and invasive bladder cancer, but this was based on few cases. These findings suggest that T2DM and possibly T1DM are positively associated with invasive bladder cancer risk.