Objective The use of combined hormonal contraceptives (CHCs) worsens glucose tolerance, but the risk for glucose metabolism disorders remains controversial. Design The study is a prospective longitudinal population-based cohort study. Methods The study was based on a cohort population that comprised 1879 women born in 1966. At age 46, the women answered a questionnaire on contraceptive use and underwent an oral glucose tolerance test. Glucose metabolism indices were evaluated in current CHC (n = 153), progestin-only contraceptive (POC, n = 842), and non-hormonal contraceptive users (n = 884). Results In the entire study population, current CHC use was significantly associated with prediabetes (OR: 2.0, 95% CI: 1.3–3.2) and type 2 diabetes (OR: 3.3, 95% CI: 1.1–9.7) compared to non-hormonal contraceptive use. After 5 years of use, the prediabetes risk increased 2.2-fold (95% CI: 1.3–3.7) and type 2 diabetes risk increased 4.5-fold (95% CI: 1.5–13.5). Compared with the current POC use, current CHC use was significantly associated with prediabetes (OR: 1.9, 95% CI: 1.2–3.0). Current POC use was not associated with any glucose metabolism disorders. The results prevailed after adjusting for BMI and socioeconomic status. Conclusions CHC use in perimenopausal women was associated with a significantly increased risk of glucose metabolism disorders. This association should be considered in women with increased metabolic risk.