|
|
| |||||||||||||||||||||||||||||||||||||||
Background On January 1, 2000, Texas began enforcement of a law that requires physicians to notify a parent of a minor child seeking an abortion at least 48 hours before the procedure.
Methods We assessed changes in the rates in Texas of abortions and births (events per 1000 age-specific population) before enforcement of the parental notification law (1998 to 1999) and after enforcement (2000 to 2002). We did this by comparing the rate changes among minors 15 to 17 years of age at the time of conception (i.e., those who were subject to the law) with those of teens 18 years of age at the time of conception (i.e., those who were not subject to the law).
Results After enforcement of the law, abortion rates fell by 11 percent among 15-year-olds (rate ratio, 0.89; 95 percent confidence interval, 0.83 to 0.94), 20 percent among 16-year-olds (rate ratio, 0.80; 95 percent confidence interval, 0.76 to 0.85), and 16 percent among 17-year-olds (rate ratio 0.84; 95 percent confidence interval, 0.80 to 0.87), relative to the rates among 18-year-olds. Among the subgroup of minors 17.50 to 17.74 years of age at the time of conception (who would have been subject to the parental notification law in early pregnancy), birth rates rose by 4 percent relative to those of teens 18.00 to 18.24 years of age (rate ratio, 1.04; 95 percent confidence interval, 1.00 to 1.08). The adjusted odds ratio for having an abortion after 12 weeks' gestation among minors 17.50 to 17.74 years of age as compared with 18-year-olds was 1.34 (95 percent confidence interval, 1.10 to 1.62).
Conclusions The Texas parental notification law was associated with a decline in abortion rates among minors from 15 to 17 years of age. It was also associated with increased birth rates and rates of abortion during the second trimester among a subgroup of minors who were 17.50 to 17.74 years of age at the time of conception.
The effect of parental involvement laws is uncertain. Some research suggests that such laws lower rates of abortion among minors but have little effect on birth rates.4,5,6,7,8 Other studies suggest that the observed decline in abortions is spurious, since abortions to minors obtained outside their state of residence are often not recorded.9,10,11,12 Moreover, results from previous research may be less relevant to the probable effect of such laws today, because of the increased distance a minor has to travel in order to obtain an abortion in states without parental involvement laws. The number of states that enforce parental involvement laws has doubled since 1991.2,13 Given the more restrictive environment, such laws may be more likely to lower abortion rates and increase birth rates and may increase the likelihood that an abortion will be performed later in pregnancy.
Texas is geographically the largest and most populous state to enforce a parental involvement law. In addition, distances to abortion providers in states that do not enforce such laws are substantial, especially for minors in the heavily populated southeastern region of the state. Thus, Texas offers a good opportunity to assess the effect of parental involvement laws on abortions, births, and the timing of abortion in a more restrictive environment. In this study, we analyzed changes in the rates of abortions, births, and the timing of abortion among minors before and after enforcement of Texas's parental notification statute.
Methods
After receiving approval from the institutional review board at the Texas Department of State Health Services, we obtained all individual birth and abortion certificates from the period from 1997 to 2003. We limited all analyses to Texas residents who were 15 to 19 years of age who conceived during the period from August 1, 1997, to July 31, 1999, and from January 1, 2000, to December 31, 2002 (a total of 394,015 certificates). These data provide information on the date of birth of the person who had an abortion or gave birth, the date of the abortion or birth, and a clinical estimate of gestation in weeks. We estimated the date of conception by subtracting the gestational age from the date of abortion or birth. We also estimated the age at the time of conception by subtracting gestational age from the age at the time of the abortion or birth.
One limitation of the data from abortion certificates is that the reporting of the birth date of the mother is sometimes incomplete. For instance, 3419 of induced-termination certificates (24.1 percent) lacked the patient's exact date of birth in 1997, as did 1278 of the certificates (9.9 percent) in 1998, 389 (3.0 percent) in 1999, 44 (0.4 percent) in 2000, 911 (8.5 percent) in 2001, 2 (0.02 percent) in 2002, and 2 (0.02 percent) in 2003. In contrast, the mother's date of birth is well recorded on birth certificates: less than 1 percent lacked this information in each year. We excluded cases for which data on the exact date of birth were missing. Because more cases lacked data in the years before the parental notification law went into effect than after its introduction, we underestimate the absolute reduction in abortions associated with the law. However, because the proportion that is missing is distributed equally by age, the relative change in the rate of abortions among minors before and after enforcement of the law, as compared with the relative change among older teens, should not be substantively affected by the missing cases (Table 1 of the Supplementary Appendix, which is available with the full text of this article at www.nejm.org). To address the possibility of bias associated with these exclusions further, we also performed a secondary analysis limiting the sample to one year before and one year after enforcement of the law, a period during which there were fewer missing data.
Statistical Analysis
Abortion rates were calculated as the number of abortions per 1000 age-specific population. Rate ratios for abortions and births were calculated for various age groups by dividing the rate before enforcement of the parental notification law by the rate after its enforcement. Because abortion rates and birth rates among teenagers have been declining nationally and in Texas since 1991,14,15,16,17 we calculated the rate ratios for abortions and births for teens who were 18 years of age at the time of conception and were therefore not subject to the parental notification requirements; these teens were used as the referents (i.e., the rate ratios that would have been expected in the absence of a parental notification law).
We estimated the changes in abortion and birth rates associated with the law by dividing the rate ratio for teens under the age of 18 by the rate ratio for 18-year-olds to calculate relative rate ratios. A relative rate ratio of 1 indicates no association between the law and the rates of abortions and births for minors. We used the delta method to estimate the confidence interval of the natural logarithm of the rate ratio (Section 1 of the Supplementary Appendix).18
Before enforcement of the parental notification law, there were large differences in the rates of abortions and births between younger and older teens. To help compensate for these differences, we also performed analyses comparing changes in outcomes among minors who were at least 17 years 6 months but less than 17 years 9 months of age at the time of conception (17.50 to 17.74 years) with teens who were less than 3 months beyond their 18th birthday at the time of conception (18.00 to 18.24 years). Minors who conceived when they were at least 17 years 9 months but less than 18 years at the time of conception (17.75 to 17.99 years) were unlikely to have been affected by the law, since most would have been 18 years of age by the time that an abortion would have been performed.
We used logistic regression to analyze the association between Texas's parental notification law and the probability that a minor's pregnancy would end in abortion. An odds ratio of less than 1 indicates that the law is associated with a decreased likelihood that a minor, as compared with an 18-year-old, would have an abortion. We also analyzed whether the law was associated with an increased probability that an abortion, if performed, would occur after 12 weeks' gestation (a second-trimester abortion).19,20 Odds ratios were adjusted for race and ethnic background (as self-reported by teens), marital status, previous abortions, previous live births, and health service region of residence, all of which were reported on both the birth and induced-termination certificates.
Results
Abortions and Births
There were, on average, 14,644 abortions and 65,055 births per year among teens in Texas in the two years before enforcement of the parental notification law (Table 1). Abortions were most common among non-Hispanic white teens (41.5 percent), whereas most births were to Hispanic teens (53.5 percent).
|
|
|
Pregnancy Resolution and Abortion Timing
The odds that a pregnancy would end in abortion after enforcement of the law fell significantly more among minors who were 17.50 to 17.74 years of age at the time of conception than among 18-year-olds (adjusted odds ratio, 0.77; 95 percent confidence interval, 0.71 to 0.84) (Table 4). Among minors who were 17.70 to 17.80 years of age at the time of conception, there was an increase in the proportion of abortions performed after 12 weeks' gestation after enforcement of the law (years 2000–2002) (Figure 1). There was no analogous rise in second-trimester abortions among 17-year-olds two years before enforcement of the law (1998–1999). The adjusted odds of having an abortion after 12 weeks increased among teens who were 17.50 to 17.74 years of age at the time of conception relative to those who were 18 years of age (odds ratio, 1.34; 95 percent confidence interval, 1.10 to 1.62) (Table 4). However, no significant increase in the odds of second-trimester abortion was observed among younger 17-year-olds (among whom parental notification requirements would have been in force throughout the time that abortion was an option).
|
|
On the basis of birth and abortion records, we found that enforcement of Texas's parental notification law was associated with a decline in abortion rates among all minors. In association with the law, there was also a decrease in the odds that a pregnancy would result in an abortion and an increase in the odds of a second-trimester abortion among older minors who were 17.50 to 17.74 years of age at the time of conception.
We also found that among this older group of minors, the abortion rate of non-Hispanic white minors and Hispanic minors fell after enforcement of the law, whereas the abortion rate of non-Hispanic black minors did not. This finding was consistent with racial and ethnic differences in communication between minors and parents about the use of reproductive health services.21,22 In a recent survey of minors at family planning clinics, 53.1 percent of Hispanics, 58.1 of non-Hispanic whites, and 75.9 of non-Hispanic blacks reported that their parents knew of their use of the clinics' services.21 In an earlier study, researchers reported that parents of black girls were more likely to know that the minor sought an abortion than were the parents of white girls.23
Results from previous studies have indicated that the abortion rate of minors decreases 13 to 42 percent after enforcement of a parental notification or consent law, if abortions are measured according to state of occurrence.5,8,9,10,12 Only a few studies have reported a decline associated with these laws when abortions are measured according to state of residence.4,7,11 Researchers have also analyzed changes in birth rates associated with parental notification or consent laws.6,7,10,11 Birth rates should rise if abortion rates fall, unless minors increase the use of contraception or decrease sexual activity in response to a parental notification or consent law. However, no previous study has reported a change in birth rates associated with such laws.6,7,10,11
In each of these studies, researchers have relied on information regarding the age of minors at the time of birth and thus have been unable to distinguish teens who conceived at the age of 18 years and also gave birth at the age of 18 years from teens who conceived at the age of 17 years (and were subject to the law) but gave birth at the age of 18 years. This discrepancy may have biased previous estimates of the effect of parental involvement laws on birth rates toward a finding of no increase.6,7,10,11 A strength of our study was the availability of information from birth and abortion certificates that allowed us to calculate age at the time of conception and thus to distinguish teens who were subject to the law at the time of conception from those who were not. Moreover, the relatively large number of abortions and births to teens in Texas allowed us to focus on the behavior of 17-year-olds. Older minors are more likely to be affected by parental involvement laws than are younger minors, because they are less likely to communicate with their parents about their abortion.23 The demonstrated decrease in abortion rates and increase in birth rates suggest that Texas's parental notification law induced a rise in unintended childbearing among this subgroup of older minors.
A potential limitation of our study and of previous analyses of parental involvement laws is the possibility that the decline in abortion rates associated with the law might have been overestimated if there had been an increase in abortions in other states to minors from Texas that were not recorded by the Texas Department of State Health Services. On the basis of information that was recorded on abortion certificates from the health department, less than 0.3 percent of abortions to minors who were residents of Texas were obtained in another state from 2000 to 2002. The low number, however, could mean that abortion certificates of Texas residents that were recorded in other states were not sent to the Texas health department by the health departments in other states. To assess this possibility, we obtained data on the number of abortions that Texas residents had in neighboring states, as recorded by the respective state health departments. In 2000, there were 13 abortions obtained by Texas minors in New Mexico, 5 in Oklahoma, and 5 in Arkansas (Henshaw S: personal communication). A high rate of reporting is supported by the observation that in 2000 the numbers of abortions that were reported by the state health departments in New Mexico, Oklahoma, and Arkansas are very close to those reported by the survey of abortion providers conducted by the Alan Guttmacher Institute, which is widely accepted to provide an accurate estimate of abortions.16,24 Louisiana does not report the state of residence on the abortion certificate. However, the state has enforced a parental consent law since 1978, which makes it unlikely that minors would seek abortions in Louisiana in response to the parental notification law in Texas. Finally, there is little evidence to suggest that the decline in abortions associated with the law was due to the underreporting of abortions for minors that were performed in Mexico (Section 5 of the Supplementary Appendix).
We limited our analysis to Texas and used 18-year-olds as the comparison group because similar data were not available from bordering states. An analysis of data from another state would allow minors of the same age to be used as a comparison group, but an important limitation of such an analysis would be that other unmeasured determinants of the numbers of abortions and births (which are likely to vary across states) could confound the estimates. However, published data from the Centers for Disease Control and Prevention relating to changes in abortion rates in states that did not have changes in the status of parental notification laws appear to support our conclusions. For instance, the abortion rate of 17-year-olds fell by 25 percent in Texas from 1999 to 2000 but rose by 12 percent in Louisiana, New Mexico, and Oklahoma; overall abortion rates fell by 6 percent in the 41 states (and the District of Columbia) that report such data and had no change in their parental notification laws during this period (Table 5 of the Supplementary Appendix).16,25
In summary, the enforcement of a parental notification law in Texas has been associated with lower abortion rates among minors and increased birth rates among older minors who were 17.50 to 17.74 years of age at the time of conception. We also found an increase in second-trimester terminations among teens who did not reach the age of 18 years until after the first trimester of pregnancy. Although the observational design of the study makes it impossible to confirm causality, these findings are relevant to an assessment of the likely effect of pending legislation to extend such laws.
Supported by a grant (1-R03-HD49359-01) from the National Institute of Child Health and Human Development to the National Bureau of Economic Research.
No potential conflict of interest relevant to this article was reported.
Source Information
From Baruch College (T.J.) and the Graduate Center (S.C.), City University of New York; and the National Bureau of Economic Research (T.J., R.K.) — both in New York; and the University of Illinois at Chicago, Chicago (R.K.).
Address reprint requests to Dr. Joyce at the National Bureau of Economic Research, 365 Fifth Ave., 5th Fl., New York, NY 10016, or at ted_joyce{at}baruch.cuny.edu.
References
| |||||||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
