Reduce the rate of fetal deaths at 20 or more weeks of gestation — MICH‑01 Data Methodology and Measurement

About the National Data


Baseline: 5.9 fetal deaths at 20 or more weeks of gestation per 1,000 live births and fetal deaths occurred in 2017

Target: 5.7 per 1,000

Number of fetal deaths (20 or more weeks of gestation).
Number of live births plus fetal deaths (20 or more weeks gestation).
Target-setting method
Target-setting method details
Linear trend fitted using weighted least squares and a projection at the 50 percent prediction interval.
Target-setting method justification
Trend data were evaluated for this objective. Using historical data points, a trend line was fitted using weighted least squares, and the trend was projected into the next decade. This method was used because three or more comparable data points were available, the projected value was within the range of possible values, and a projection at the 50 percent prediction interval was selected because the slope of the linear trend was statistically significant.


Methodology notes

A description of the primary measurement used to determine gestational age —obstetric estimate of gestation at delivery (OE)— has been published by NCHS. The OE replaced the measure based on the date of the last normal menses (LMP). This transition was made because of the increasing evidence of the greater validity of the OE compared with the LMP-based measure.

The majority of states require reporting fetal deaths of 20 weeks of gestation or more, or 350 grams delivery weight (roughly equivalent to 20 weeks), or some combination of the two. The number of fetal deaths may be underreported in part because of variations in state reporting requirements.


Comparable HP2020 objective
Retained, which includes core objectives that are continuing from Healthy People 2020 with no change in measurement.


Additional resources about the objective

1. Because Healthy People 2030 objectives have a desired direction (e.g., increase or decrease), the confidence level of a one-sided prediction interval can be used as an indication of how likely a target will be to achieve based on the historical data and fitted trend.