Objective: To examine the associations of female reproductive and hormonal factors with incidence of total knee arthroplasty (TKA) for osteoarthritis (OA), and to determine whether the associations differ according to overweight/obesity status. Methods: This study included 22,289 women in the Melbourne Collaborative Cohort Study. Data on age at menarche, pregnancy, parity, years of menstruation, oral contraceptive (OC) use, menopausal status, and hormone replacement therapy (HRT) were collected in 1990-1994. Incidence of TKA during 2001-2013 was determined by linking cohort records to the Australian Orthopaedic Association National Joint Replacement Registry. All analyses were adjusted for age, body mass index (BMI) at midlife, change in BMI (from early reproductive age to midlife), country of birth, physical activity, smoking, and education level. Results: Over the course of 12.7 years, 1,208 TKAs for OA were identified. Ever being pregnant was associated with increased risk of TKA (hazard ratio [HR] 1.32 [95% confidence interval (95% CI) 1.06-1.63]). Parity was positively associated with risk of TKA (P for trend = 0.003). OC users had increased risk of TKA compared with non-users (for OC use of <5 years, HR 1.25 [95% CI 1.08-1.45]; for OC use of ≥5 years, HR 1.17 [95% CI 1.00-1.37]). A 1-year increase in menstruation was associated with a 1% decrease in risk of TKA (HR 0.99 [95% CI 0.97-0.99]). These associations remained significant only in women of normal weight at early reproductive age. Current HRT users had increased risk of TKA compared with non-users (HR 1.37 [95% CI 1.14-1.64]); the association was significant only in non-obese women at midlife. Conclusion: Reproductive and hormonal factors were associated with risk of knee OA. These associations remained significant in women of normal weight at early reproductive age and in non-obese women at midlife. Further work is needed to understand the complex effect of these factors on knee OA.