Birth defect of any type is undesirable and often pose a negative impact on the health and development of the newborn. Birth defects surveillance with datasets from surveillance health-related programs are useful to predict the pattern of birth defects and take preventive measures. In this study, the total prevalence, perinatal prevalence, and livebirth prevalence of birth defects were compared. Data were obtained from the Birth Defects Surveillance System in Hunan Province, China, 2016-2020. The total prevalence is the number of birth defects (including livebirths, stillbirths, and selective terminations of pregnancy) per 1,000 births (including livebirths and stillbirths). The perinatal prevalence is the number of birth defects (between 28 weeks gestation and 7 days postpartum) per 1,000 births. The livebirth prevalence is the number of liveborn birth defects per 1,000 births (unit: ‰). Underestimated proportion (unit: %) is the reduction level of perinatal prevalence or livebirth prevalence compared to the total prevalence. Prevalence with 95% confidence intervals (CI) was calculated using the log-binomial method. Chi-square tests (χ 2) were used to examine if significant differences existed in prevalence or underestimated proportion between different groups. A total of 847,755 births were included in this study, and 23,420 birth defects were identified, including 14,459 (61.74%) birth defects with gestational age > =28 weeks, and 11,465 (48.95%) birth defects in livebirths. The total prevalence, perinatal prevalence, and livebirth prevalence of birth defects were 27.63‰ (95%CI, 27.27-27.98), 17.06‰ (95%CI, 16.78-17.33), and 13.52‰ (95%CI, 13.28-13.77), respectively, and significant differences existed between them (χ2 = 4798.55, p < 0.01). Compared to the total prevalence, the perinatal prevalence and livebirth prevalence were underestimated by 38.26 and 51.05%, respectively. Significant differences existed between the total prevalence, perinatal prevalence, and livebirth prevalence of birth defects in all subgroups according to year, sex, residence, and maternal age (p < 0.05). Significant differences existed between the total prevalence, perinatal prevalence, and livebirth prevalence for 17 specific defects: congenital heart defect, cleft lip-palate, Down syndrome, talipes equinovarus, hydrocephalus, limb reduction, cleft lip, omphalocele, anal atresia, anencephaly, spina bifida, diaphragmatic hernia, encephalocele, gastroschisis, esophageal atresia, bladder exstrophy, and conjoined twins (p < 0.05). In comparison, no significant difference existed between the total prevalence, perinatal prevalence, and livebirth prevalence for 6 specific defects: polydactyly, other external ear defects, syndactyly, hypospadias, cleft palate, and anotia/microtia (p > 0.05). The total prevalence and livebirth prevalence of birth defects in Hunan Province, China, was not well studied. A systematic study was conducted to compare the total prevalence, perinatal prevalence, and livebirth prevalence of birth defects. The study reveals that significant differences existed between the total prevalence, perinatal prevalence, and livebirth prevalence of birth defects (including many specific defects), and year, sex, residence, and maternal age had significant impacts on it. The outcomes of the study will help to take preventive measures for birth defects as well as benefit the people involving public health and policymakers to improve the current scenario.