Quarterly Provisional Estimates for Selected Birth Indicators, 2016-Quarter 3, 2018
Provisional estimates are based on a snapshot of all vital statistics birth certificate data received and processed by NCHS as of a specified cutoff date. To adjust for the incompleteness of these data, individual records are weighted to independent provisional counts of all births that occurred in each state by month. Provisional estimates closely match final data, but they are subject to revision in future quarterly releases as additional records for that quarter are received. See Technical Notes for more details.
In this release of Quarterly Provisional Estimates, NCHS presents provisional estimates of selected reproductive indicators from birth data for 2016 through the third quarter of 2018. Estimates for 2016 and 2017 are based on final data (1-2), while estimates for 2018 are provisional. The estimates are based on all birth records received and processed by NCHS as of December 9, 2018. Estimates are presented for: general fertility rates, age-specific birth rates, total and low risk cesarean delivery rates, preterm birth rates and other gestational age categories. These indicators were selected based on their importance for public health surveillance as well as the feasibility of producing reliable estimates using available provisional data (3). Future quarterly releases will include additional birth indicators from natality data (4). Quarterly estimates are compared with estimates for the same quarter of the preceding year. For comparability with rates for 12-month periods, the quarterly (3-month) rates have been annualized to present births per year per 1,000 population that would be expected if the quarter-specific rate prevailed for 12 months. In addition, the rates and percentages for a 12-month period ending with each quarter (i.e., 12-month moving average) are presented to account for seasonality. Estimates for the 12-month period ending with the fourth quarter in each year can be interpreted as an annual provisional estimate for that year. Differences are reported only if statistically significant (P<0.05).