To determine the influence of the interval between fetal measurements on performance of fetal growth velocity for predicting infants with anthropometric features of fetal growth restriction (FGR). Two hundred seventy-four low-risk women had serial fetal biometry at scheduled intervals. Growth velocity of the fetal abdominal area for each was calculated with 2-, 4-, and 6-week scan intervals in which the second measurement was the last scan before delivery. Fetal abdominal area velocity over a 4-week interval in the early third trimester also was included. Fetal growth restriction was defined as skinfold thickness under the tenth percentile, ponderal index under the 25th percentile, midarm circumference-to-occipitofrontal circumference ratio of under -1 standard deviation (SD). Test performance was expressed as likelihood ratios with 95% confidence intervals (CI). Fetal abdominal area velocity calculated over a 4-week interval predicted FGR with a likelihood ratio of 10.4 (95% CI 3.9, 26) for skinfold thickness; 9.5 (95% CI 4.6, 19) for ponderal index; and 4.7 (2.3, 8.4) for midarm circumference-to-occipitofrontal circumference ratio. Intermeasurement intervals of 6 weeks had a likelihood ratio of 8.5 (95% CI 4, 17) for skinfold thickness; 7.5 (95% CI 3.4, 16.1) for ponderal index; and 14 (6.7, 28) for midarm circumference-to-occipitofrontal circumference ratio. The likelihood ratios for the 2-week interval and the early third trimester 4-week interval were all less than 5. Four- and 6-week measurement intervals were useful for predicting infants with FGR and were superior to a 2-week interval. Fetal growth velocity is influenced by proximity of the last fetal measurement to date of delivery, which adversely affects clinical use of growth velocity for predicting FGR.