Introduction Current use of combined hormonal contraceptives worsens glucose tolerance and increases the risk of type 2 diabetes mellitus at late fertile age, but the impact of their former use on the risk of glucose metabolism disorders is still controversial. Material and methods This was a prospective, longitudinal birth cohort study with long‐term follow‐up consisting of 5889 women. The cohort population has been followed at birth, and at ages of 1, 14, 31 and 46. In total, 3280 (55.7%) women were clinically examined and 2780 also underwent a 2‐h oral glucose tolerance test at age 46. Glucose metabolism indices were analyzed in former combined hormonal contraceptive users (n = 1371) and former progestin‐only contraceptive users (n = 52) and in women with no history of hormonal contraceptive use (n = 253). Results Compared with women with no history of hormonal contraceptive use, those who formerly used combined hormonal contraceptives for over 10 years had an increased risk of prediabetes (odds ratio OR 3.9, 95% confidence interval CI: 1.6–9.2) but not of type 2 diabetes mellitus. Former progestin‐only contraceptive use was not associated with any glucose metabolism disorders. The results persisted after adjusting for socioeconomic status, smoking, alcohol consumption, parity, body mass index and use of cholesterol‐lowering medication. Conclusions Former long‐term use of combined hormonal contraceptives was associated with a significantly increased risk of prediabetes in perimenopausal women, which potentially indicates a need of screening for glucose metabolism disorders in these women. Former long‐term use of CHCs was associated with prediabetes in perimenopausal women, which potentially indicates a need of glucose metabolism screening in these women and emphasizes the importance of considering contraception alternatives other than CHCs at later fertile age.