Contact Name
Barbara Dietsch
562-799-5126
bdietsc@WestEd.org
A User’s Guide To Data On Teen Pregnancy
In the 1990s we witnessed substantial declines in teen pregnancy and birth rates in the United States, declines which essentially reversed the steep increases observed in the late 1980s. As we review the recent and long-term trends it is important to distinguish among the various measures used to describe teen pregnancy, so that the data we use are appropriate for the message.
The most commonly-used measure is the teen birth rate, which describes the risk that a teenager will give birth in a given year. To compute this measure, you need the number of births occurring to teenagers, for example those aged 15-19 years, and the number of teenage females in the same age group in the population. The rate is usually expressed per 1,000 teenage females in the specified age group. In 1998, the birth rate, according to preliminary data from the National Center for Health Statistics (NCHS) was 51.1 births per 1,000 females aged 15-19. This rate has declined 18 percent from its recent high, 62.1 in 1991.
When considering trends in teen childbearing, it is advisable to look separately at patterns for younger and older teenagers. The health, social, and economic consequences, and the appropriate interventions and prevention strategies vary substantially between the two groups. The rate for older teenagers has generally been more than 2.5 times the rate for younger teens (in 1998, 82.0 for ages 18-19 compared with 30.4 for ages 15-17 years). Rates fell for both groups in the 1990s, but the declines were much greater for younger teens (down 21 percent) than for older teens (down 13 percent).
During the 1990s, despite the 18-percent decline in the birth rate, the number of births to teenagers fell just 7 percent, reflecting concurrent increases in the teenage population. Projected increases in the teen population mean that further reductions in the number of births to teenagers will depend on sustained and substantial declines in the birth rate.
Another measure often used is the proportion of all births that are to teenagers. Although the data needed to compute this measure are almost always available, the measure is problematic. The proportion can increase or decline because of trends in births and birth rates for older age groups, having nothing to do with teenage childbearing.
What is it about teenage childbearing that has resulted in so much public attention and policy intervention? One big change has been the striking increase in the proportion of births to teenagers that are out-of-wedlock. In 1970, 30 percent of births to teenagers were out-of-wedlock; in 1998, nearly four-fifths were. Three decades ago, many unmarried teenagers who became pregnant married before the birth of the child. The birth rate for unmarried teenagers increased steeply until just recently, more than doubling from 1970 to 1994. As a consequence of these changes, the number of births to unmarried teenagers, a central focus of public policy makers today, rose steeply.
All the measures discussed to this point relate to teenage births. But we know that is just part of the story. Teenage pregnancy and birth rates are not the same. Just over half of teen pregnancies result in a live birth. In order to compute teen pregnancy rates, you need data on three outcomes: live births, induced abortions, and fetal losses. The teenage pregnancy rate is computed by summing these three components and dividing the result by the teen population, in this case females aged 15-19, and then multiplying by 1,000. Because it is more difficult to collect and assemble data on induced abortions and fetal losses, the data on these two outcomes are not as current as data for live births.
In 1996, the most recent year for which all the necessary data are available, the teenage pregnancy rate was 98.7 per 1,000 women aged 15-19, 15 percent lower than its all-time high in 1991, 116.5. The 1996 rate is the lowest in the 20-year period for which a consistent series has been reported by NCHS. Teen pregnancy, birth, and abortion rates have all declined in the 1990s.
We can track changes in key measures of teenage pregnancy and childbearing fairly readily at the national level. Monitoring trends at the state level is more limited. We can compile the number of births to teenagers and compute teen birth rates; we can also look at the percent of teen births that are out-of-wedlock. All of these measures can be very useful to state program planners and policy makers. However, we cannot compute birth rates for unmarried teens because we do not have reliable estimates of unmarried teenagers at the state level. (The population estimates we use for computing birth rates for all teenagers are usually from the U.S. Census Bureau, although many states now produce their own population estimates as well.) We cannot routinely compute teen pregnancy rates at the state level because of the difficulty in assembling complete and timely abortion data and the difficulty in estimating fetal losses. These are important data limitations, because about 3 in 10 teen pregnancies result in an induced abortion, and about 1 in 6 end in a fetal loss according to data recently reported by NCHS.
Data limitations are even more challenging for monitoring trends in teen childbearing at the local or community level, for which reliable population estimates are not available even for the basic teen birth rate. Thus, you often find people discussing the proportion of all births in their area that occurred to teenagers or the percent of teen births that were out-of-wedlock – measures for which data are available from the birth registration system itself. However, as noted earlier, changes in the proportion of all births occurring to teens can reflect changes in birth rates for older women which are unrelated to teen childbearing, and reliance on these changes can lead to misinterpretation of what is really happening in teen childbearing.
In summary, having and using the appropriate measure and data item are vitally important as we monitor trends in teenage pregnancy and birth rates, but it is clear that in many cases what is needed is not going to be available. Keeping data limitations in mind is critical as you track teen pregnancy or birth patterns in your state or community. Often with national statistics as a backdrop and context for your community effort, you can compile some comparable measures for your locality, such as the teen birth rate or the proportion of teen births that are out-of-wedlock to help evaluate the success of prevention programs.
Detailed information on teenage pregnancy and birth patterns is available at the NCHS homepage: http://www.cdc.gov/nchs/about/major/natality/natality.htm.
Stephanie J. Ventura is Research Statistician, Division of Vital Statistics, National Center for Health Statistics.
