A CIDR synchronization program is an important tool for facilitating ovulation synchronization and timed AI (OvSynch-TAI). The objective of this study was to test the efficacy of reusing CIDR for OvSynch-TAI compared with that for a progesterone-free OvSynch-TAI protocol (NCS) or a breed by oestrus detection (ED) control. A total of 87 does were randomised into 1 of 5 treatments: (1) ED (control, n = 18), (2) NCS (n = 18), (3) CIDR6-New (n = 17), (4) CIDR6–1X (n = 17), and (5) CIDR6–2X (n = 17). Does in the ED group received two 15-mg doses of PGF2α at a 10-d interval and were bred 12 h after the onset of oestrus following the second PGF2α injection. The NCS group received 15 mg of PGF2α on Day 0, 50 μg of gonadotropin-releasing hormone on Day 8, 15 mg of PGF2α on Day 15, and 50 μg of gonadotropin-releasing hormone on Day 18, concurrently with TAI. The CIDR groups (new, 1X-used, or 2X-used) received a P4 device for a 6-d period, and 15 mg of PGF2α was administered at CIDR removal. Does were bred 48 h after CIDR removal and were given 50 μg of gonadotropin-releasing hormone at TAI. The CIDR in the CIDR6–1X group were previously in place for 10 days before use, whereas CIDR in the CIDR6–2X group were previously in place for 16 days before use. Reused CIDR were rinsed in a diluted Nolvasan solution, followed by a clean water rinse, allowed to air dry, and stored in a refrigerator until time of use. The experiment was conducted in 2 replicates. Within each replicate, all treatments were scheduled so that does were bred during the same 2-day period, and all does were inseminated with a single dose of frozen semen using a nonsurgical (transcervical) technique. Blood samples were taken daily in all treatment groups to monitor concentrations of serum progesterone until the time of breeding. Pregnancy was determined by ultrasonography at 54 and 85 days of gestation. Data were analysed using GLM procedures of SAS (SAS Institute Inc., Cary, NC, USA). Data for pregnancy rates were analysed with a model that included effects of treatment, replicate, and their interactions. Data for serum progesterone concentrations were analysed with a model that included the effects of treatment, replicate, day, and their interactions. Means were separated by Duncan’s test. Mean progesterone differed (P < 0.001) with treatment (5.3 ± 0.8bc, 3.5 ± 0.8c, 7.0 ± 0.8ab, 7.9 ± 0.8a, 6.2 ± 0.8ab ng mL–1 for ED, NCS, CIDR6-New, CIDR6–1X, and CIDR6–2X, respectively; least squares mean ± standard error of the mean). Pregnancy rates for the ED, NCS, CIDR6-New, CIDR6–1X, and CIDR6–2X treatment groups were 39 ± 11%bc, 22 ± 11%c, 64 ± 12%ab, 77 ± 12%a, and 57 ± 12%ab, respectively. In conclusion, reused CIDR were as effective as new CIDR for attaining satisfactory pregnancy rates. Timed AI using a once-used CIDR was more effective for establishing pregnancy than ED and NCS treatments. The lower pregnancy rates in the ED and NCS groups were associated with lower mean progesterone levels during the Ovsynch treatment period before breeding. This research was supported by the North Carolina Agricultural Experiment Station.