BACKGROUND: The incidence of multiple sclerosis (MS) is rising in women. OBJECTIVE: To determine whether the use of combined oral contraceptives (COCs) are associated with MS risk and whether this varies by progestin content. METHODS: We conducted a nested case-control study of females ages 14-48 years with incident MS or clinically isolated syndrome (CIS) 2008-2011 from the membership of Kaiser Permanente Southern California. Controls were matched on age, race/ethnicity and membership characteristics. COC use up to ten years prior to symptom onset was obtained from the complete electronic health record. RESULTS: We identified 400 women with incident MS/CIS and 3904 matched controls. Forty- percent of cases and 32% of controls had used COCs prior to symptom onset. The use of COCs was associated with a slightly increased risk of MS/CIS (adjusted OR = 1.52, 95%CI = 1.21-1.91; p<0.001). This risk did not vary by duration of COC use. The association varied by progestin content being more pronounced for levenorgestrol (adjusted OR = 1.75, 95%CI = 1.29-2.37; p