Judicial Bypass for Minors Seeking Abortions in Arkansas Versus Other States

Abstract

Objectives. To describe demographic characteristics and abortion timing of minors in Arkansas who obtained an abortion through a judicial bypass, and to contrast the frequency of judicial bypass in other states in the United States.

Methods. We used individual-level data on all abortions to minors (n = 2624) performed in Arkansas from 2005 to 2014 and aggregated data from 10 additional states on abortions obtained through a judicial bypass. We characterized correlates of the judicial bypass and its association with the timing of abortion.

Results. Minors in Arkansas who used the courts were more likely to be 17 years of age, nonresidents of Arkansas, Hispanic, and with fewer previous pregnancies than their younger, resident, non-Hispanic White, and multigravida counterparts; 7.7% of abortions to minors were obtained via a bypass in 11 states, but only 2% if Texas and Arkansas are excluded.

Conclusions. The demographics of minors who obtain an abortion through a judicial bypass differ significantly from those who have parental consent, and there is widespread variation in the proportion of bypass cases across states.

The US Supreme Court legalized abortion in January 1973, but did not resolve the procedure by which minors could obtain an abortion without parental consent until Bellotti v Baird in 1979.¹ The ruling allowed states to require that minors obtain consent of a parent before terminating a pregnancy. However, minors had the right to petition the court for permission to bypass parental consent or notification. If the court deemed the minor sufficiently mature to make the decision, or if parental involvement would not be in the best interest of the minor, the abortion could go forward without parental involvement. Currently 37 states require parental involvement, either consent or notification, before an abortion can be performed.²

Although there is a large literature on the impact of parental involvement laws on reproductive outcomes, there is relatively little work on minors who obtain an abortion through a judicial bypass.³ Early studies from Massachusetts and Minnesota reported that older minors were the most likely to file a petition to the courts for an abortion without parental involvement, and a more recent survey in Ohio reported that 95% of minors using a judicial bypass for an abortion were aged at least 16 years.^4–6 The first population-based description of minors obtaining an abortion through a judicial bypass used vital statistics from Arkansas from 2005 to 2007.⁷ In 2005, Arkansas changed its parental involvement law from requiring parental notification to requiring notarized written consent of a parent or guardian. The legislation also required that abortions to minors obtained through a judicial bypass be recorded on the induced termination of pregnancy certificate collected by the state health department. An early analysis of these data indicated that minors who obtained an abortion via a judicial bypass were older, less likely to be state residents, and terminated their pregnancies earlier than did their peers who had parental consent.⁷

Court proceedings in bypass cases are strictly confidential given the age of minors and the sensitivity of the decision, which might explain the lack of academic research. However, 13 states currently require that abortion facilities report to state health departments whether parents were notified or consented, or whether the abortion was obtained through a judicial bypass procedure.⁸ Seven of the 13 states report these totals as part of their annual report of vital statistics. Arkansas is the only state that provides researchers a de-identified file of induced termination records.

We make several contributions to the literature. First, we used 10 years of individual-level data from Arkansas to contrast the characteristics of minors who obtained an abortion through a judicial bypass to those with parental consent. The large number of cases provided more detailed breakdowns by age, race/ethnicity, and state residency than has been possible previously. The data also permitted an investigation as to whether the bypass procedure was associated with an increased risk of late-term abortions. Although the risk of complications from abortion is small, it rises significantly during the second trimester.^9,10 Second, we present trends in number of abortions to minors, and the proportion of those obtained through a judicial bypass over the past 10 years. In Arkansas, as well as in many other states, the political environment has changed substantially since 2010 with the increasing number of conservative state legislatures. Coincident with this change is a dramatic increase in abortion restrictions at the state level.¹¹ Lastly, we extended the analysis to 10 additional states that provided aggregate statistics on abortions obtained by minors through a judicial bypass. These newly compiled data provide the first comparison of the number and proportion of abortions obtained through a judicial bypass across states and over time.

METHODS

We analyzed a de-identified file with individual records on all abortions from 2005 through 2014 obtained from the Arkansas Department of Health. The Arkansas Department of Health collects data on induced terminations of pregnancy as part of its vital record system. Information includes the minor’s age, race/ethnicity, marital status, educational attainment, previous births, previous induced and spontaneous abortions, state of residence, and the procedure that was used to terminate the pregnancy. Since August 2005, the induced termination certificate completed by providers includes whether the abortion required parental consent and, if so, whether consent is obtained and, if not, whether the termination was obtained through a judicial bypass. Our analytical data file included 2624 abortions to minors who terminated in Arkansas from August 2005 to December 2014.

In each year, there were abortions with no indication of either parental consent or a judicial bypass. This can occur if the minor is emancipated through marriage or if, before the current pregnancy, the minor is aged 17 years and a judge rules the minor able to join the military with parental permission, or if a judge rules that emancipation is in the best interest of the minor.¹² There were 157 abortions (6.0%) to minors aged 17 years in which there was no indication of parental consent or a bypass, but who may have been emancipated. There were 155 cases (5.9%) in which the minor was younger than 17 years with no possibility of emancipation and yet there was no indication as to whether consent was obtained or a bypass procedure approved. We assumed all abortions to minors with no indication of a bypass did not receive court approval for an abortion without parental consent. As a robustness check, we dropped these cases and re-estimated the statistical models.

We also collected data on the number of abortions obtained through a judicial bypass from 10 of the 13 states, not including Arkansas, that collect and make available such data. Seven states report these data on the Web sites on which they post vital statistics (Alabama, Arizona, Georgia, Idaho, Oklahoma, South Dakota, and Wisconsin). Data from 3 other states were from officials from the state’s vital statistics department (Kansas, South Carolina, and Texas). We divided these counts by the total number of abortions to minors performed in the state to estimate the proportion of abortions obtained via a bypass. In some cases, the number of abortions obtained through a judicial bypass was fewer than 5. We used 3 as the count to estimate the ratio of judicial bypass cases to total number of abortions to minors when the outcome was left-censored.

We contrasted the characteristics of minors in Arkansas who obtained an induced abortion through a judicial bypass to those who did not. We estimated the unadjusted and adjusted odds ratios of obtaining an abortion via a bypass for each characteristic. We used ordinary least squares to estimate the association between the gestational age at termination in weeks and whether the minor obtained permission for the abortion through the courts. We used logistic regressions to further test for an association between the bypass procedure and abortions performed at 13 weeks’ gestation or later, and 16 weeks’ gestation or later. In all regressions, we adjusted standard errors for a general form of heteroscedasticity.

RESULTS

There were 2624 abortions to minors in Arkansas from August 2005 through December 2014. Of these, 262 (10%) were obtained through a judicial bypass (Table 1). Forty-three percent of abortions were to minors who were aged 17 years, but they accounted for 58% of those who used the bypass procedure (Table 1; adjusted odds ratio [AOR] = 4.20; 95% confidence interval [CI] = 2.73, 6.46). By contrast, 28% of abortions were to minors who were aged 15 years or younger, but only 11% of those who used the bypass. Four percent of minors who obtained an abortion in Arkansas were Hispanic, but they accounted for 9% of all minors who obtained an abortion via judicial bypass (Table 1; AOR = 2.99; 95% CI = 1.77, 5.04). Twenty percent of abortions to minors performed in Arkansas were to nonresidents of the state, but they represented 24% of all minors whose abortions were obtained via a judicial bypass (Table 1; AOR = 1.42; 95% CI = 1.04, 1.95). Six percent of minors had a previous induced abortion and another 6% had had a previous pregnancy that resulted in a live birth or a spontaneous abortion. Minors with previous pregnancies were less likely to use the bypass procedure. Less than 1% of minors were married when they obtained an abortion.

TABLE 1—

Characteristics of Women Aged 12–17 Years With Reported Abortion by Judicial Bypass Status: Arkansas, 2005–2014

		Judicial Bypass
Characteristic	Overall Proportion (n = 2624)	Yes, Proportion (n = 262)	No, Proportion (n = 2362)	OR (95% CI)	AOR (95% CI; n = 2623)a
Age, y
≤ 15	0.28	0.11	0.30	1 (Ref)	1 (Ref)
16	0.30	0.32	0.29	3.01 (1.93, 4.67)	3.02 (1.93, 4.73)
17	0.43	0.58	0.41	3.92 (2.59, 5.94)	4.20 (2.73, 6.46)
Race/ethnicity
Non-Hispanic White	0.52	0.53	0.52	1 (Ref)	1 (Ref)
Non-Hispanic Black	0.41	0.34	0.41	0.81 (0.61, 1.07)	0.87 (0.65, 1.17)
Hispanic	0.04	0.09	0.03	2.65 (1.61, 4.37)	2.99 (1.77, 5.04)
Other/missing	0.03	0.04	0.03	1.30 (0.68, 2.51)	1.41 (0.72, 2.79)
Married	0.01	0.00	0.01	. . .	. . .
State nonresident	0.20	0.24	0.20	1.30 (0.96, 1.76)	1.42 (1.04, 1.95)
Previous pregnancy
None	0.88	0.92	0.88	1 (Ref)	1 (Ref)
Induced abortion	0.06	0.04	0.06	0.61 (0.32, 1.18)	0.53 (0.28, 1.02)
Otherb	0.06	0.04	0.06	0.65 (0.35, 1.21)	0.46 (0.24, 0.89)
Pseudo-R2					0.06

Note. AOR = adjusted odds ratio; CI = confidence interval; OR = odds ratio. In each case, the dependent variable is 1 if the abortion was obtained via judicial bypass, and zero otherwise. The AORs control for each characteristic listed in the table as well as indicator variables for each year from 2005 to 2014 (not shown). Heteroscedasticity-consistent standard errors were used to create the 95% CIs.

^aPseudo-R².

^bPrevious pregnancy that resulted in a live birth or a spontaneous abortion.

Bypass and Length of Pregnancy

The mean gestational age at termination for minors who used the judicial bypass was 7.9 weeks, 0.79 weeks (roughly 5 days) earlier than minors who had parental consent (Table 2; P < .01). Differences in gestational age, however, were greater by race/ethnicity. After adjustment, on average, non-Hispanic Black minors terminated a pregnancy 2.27 weeks later than their non-Hispanic White peers (Table 2; P < .01). Non-Hispanic White minors terminated their pregnancy 6 days earlier, on average, than Hispanics. The largest difference in mean gestational age was between resident and nonresident minors of Arkansas. The latter accounted for 20% of all abortions to minors obtained in the state, and, on average, they terminated 3.6 weeks later than resident minors (Table 2; P < .001).

TABLE 2—

Relationship Between Delay in Abortion and Judicial Bypass Status Among Women Aged 12–17 Years With Reported Abortion: Arkansas, 2005–2014

	Gestational Age (in Weeks; n = 2615)		Gestational Age ≥ 13 Weeks (n = 2614)		Gestational Age ≥ 16 Weeks (n = 2614)
Variable	Outcome Mean	Adjusted Difference (95% CI)	Outcome Mean	AOR (95% CI)	Outcome Mean	AOR (95% CI)
Overall	8.64		0.22		0.11
Judicial bypass	7.88	−0.79 (−1.29, 0.28)	0.15	0.58 (0.38, 0.87)	0.06	0.47 (0.27, 0.84)
Age, y
≤ 15	9.24	1 (Ref)	0.26	1 (Ref)	0.14	1 (Ref)
16	8.59	−0.45 (−0.85, 0.05)	0.22	0.90 (0.69, 1.16)	0.11	0.85 (0.61, 1.19)
17	8.28	−0.51 (−0.89, 0.13)	0.19	0.83 (0.64, 1.06)	0.09	0.75 (0.54, 1.04)
Race/ethnicity
Non-Hispanic White	7.51	1 (Ref)	0.13	1 (Ref)	0.05	1 (Ref)
Non-Hispanic Black	10.14	2.27 (1.94, 2.60)	0.34	3.20 (2.57, 3.98)	0.18	3.54 (2.60, 4.81)
Hispanic	8.13	0.88 (0.08, 1.68)	0.17	1.69 (0.91, 3.13)	0.09	2.27 (1.05, 4.94)
Other/missing	8.55	0.77 (−0.08, 1.62)	0.18	1.24 (0.66, 2.33)	0.12	2.19 (1.07, 4.51)
State nonresident	11.69	3.56 (3.15, 3.98)	0.51	6.29 (5.08, 7.79)	0.30	6.62 (5.06, 8.67)
Previous pregnancy
None	8.64	1 (Ref)	0.22	1 (Ref)	0.11	1 (Ref)
Induced abortion	8.49	−0.13 (−0.76, 0.51)	0.21	0.99 (0.64, 1.53)	0.09	0.90 (0.48, 1.66)
Othera	8.78	0.19 (−0.50, 0.89)	0.22	1.00 (0.64, 1.57)	0.13	1.34 (0.77, 2.34)
Pseudo-R2		0.19		0.16		0.18

Note. AOR = adjusted odds ratio; CI = confidence interval. We estimated adjusted mean difference in gestational age by ordinary least squares. We used logistic regression to estimate the AORs for the dichotomous dependent variables that are 1 if the abortion occurred at 13 weeks’ or more gestation, or 1 if the abortion was performed at 16 or more weeks’ gestation. The independent variable of interest is judicial bypass, which equals 1 if the abortion was obtained via a judicial bypass. The AOR control for each characteristic listed in the table as well as indicator variables for each year from 2005 to 2014 (not shown). Heteroscedasticity-consistent standard errors were used to create the 95% CIs.

^aPrevious pregnancy that resulted in a live birth or a spontaneous abortion.

Twenty-two percent of minors terminated their pregnancy at 13 weeks’ gestation or later and 11% terminated at 16 weeks’ gestation or later (Table 2). The odds of terminating a pregnancy at 13 weeks’ gestation or later was significantly smaller among minors who used the judicial bypass than those who did not (Table 2; AOR = 0.58; 95% CI = 0.38, 0.87). The odds ratio of terminating at 16 weeks or later was even smaller, less than half as large among minors who used the courts (Table 2; AOR = 0.47; 95% CI = 0.27, 0.84).

Differences in delayed abortion by race/ethnicity were stark. Thirty-four percent of abortions to non-Hispanic Black minors were at 13 weeks’ gestation or later and 18% were at 16 weeks or later. Figures for non-Hispanic Whites were 13% and 5%, respectively. These differences indicated that the odds that a non-Hispanic Black minor terminated a pregnancy in the second trimester were more than 3 times greater than that of her non-Hispanic White counterpart (Table 2; AOR = 3.20; 95% CI = 2.57, 3.98). State residency revealed the most dramatic differences in the timing of abortion. Fifty-one percent of nonresident minors terminated in the second trimester and 30% terminated at 16 weeks or later yielding an adjusted odds that was more than 6 times greater than the odds for state residents (Table 2; AOR = 6.29; 95% CI = 5.08, 7.79; and AOR = 6.62; 95% CI = 5.06, 8.67, respectively).

Approximately 12% of abortions had no indication as to whether the procedure was with parental consent or via a bypass and the percentage of cases missing an indication of judicial bypass was substantially less in more recent years. We dropped these cases and re-estimated the coefficients in Table 2. The findings were very similar, suggesting that the missing cases were largely unrelated to the outcomes. The results are available upon request.

Variation Across States and Over Time

Figure A (available as a supplement to the online version of this article at http://www.ajph.org) shows the substantial variation in the proportion of bypass cases between 2010 and 2012 among minors in 11 states for which such data were available. Texas had the highest at 15.7% followed by Arkansas at 10.7%. Georgia had the lowest at 0.21% followed by Alabama at 0.48%. The average for all 11 states was 7.7% of abortions to minors obtained via a bypass but only 2.0% if Texas and Arkansas were excluded.

Figure 1 shows the trends of abortion to minors in the past 10 years in Arkansas alone. The annual number of abortions obtained by minors in Arkansas fell steadily from 2006 whereas the proportion of abortions to minors obtained through a bypass remained relatively stable. A complete list of abortions to minors and those obtained through a judicial bypass is presented in Table A (available as a supplement to the online version of this article at http://www.ajph.org).

FIGURE 1—

Number of Abortions to Minors Aged 12–17 Years in Arkansas and the Percentage of Abortions Obtained Through Judicial Bypass: 2005–2014

Note. The reported numbers are from the authors’ calculations, based on the de-identified Arkansas microdata. 2005 is annualized on the basis of data from August to December 2005.

DISCUSSION

We used 10 years of data on induced abortion in Arkansas to characterize minors who obtained an abortion through a judicial bypass. We showed that minors who used the courts to avoid involving their parents were more likely to be aged 17 years, Hispanic, and with fewer previous pregnancies than their younger, White, and multigravida counterparts. We also showed that minors who used the bypass procedure were more likely to terminate their pregnancies at earlier gestations. The larger sample extends previous work by analyzing many more judicial bypass cases, documenting the number and proportion of bypass cases in 11 states, and chronicling changes in bypass cases in Arkansas over a 10-year period during which the number of abortion facilities in the state fell by 50%.^7,13,14

A major advantage of a large sample of individual abortion records is that it permits more detailed breakdowns than are available from the Centers for Disease Control and Prevention’s (CDC’s) abortion surveillance system. According to the CDC, for example, 11% of abortions to all women occurred at 13 weeks or later and 4.8% at 16 weeks’ gestation or later, among the 38 states that reported data in 2010. The figures for Arkansas in 2010 were almost identical, 11.2% and 4.7%, respectively.¹⁵ Further stratification by age and gestational age was not available from the CDC. Yet a more detailed breakdown was possible with the Arkansas data, which revealed substantial variation in the timing of pregnancy termination by age and race/ethnicity. First, minors in Arkansas terminated significantly later than older women. Twenty-two percent of minors terminated their pregnancies at 13 weeks or later, and 11% at 16 weeks’ gestation or later. When we stratified the data by race/ethnicity, we demonstrated that the delay in pregnancy termination among minors was mainly driven by non-Hispanic Blacks, of which 34% had a second-trimester abortion compared with 13% of non-Hispanic White minors.

One reason why minors terminate later than older women is that they are late to recognize their pregnancy. In a survey of 1209 abortion patients, 57% of minors who would have preferred to terminate earlier said it took time to recognize their pregnancy.¹⁶ In another survey, women who had a second-trimester abortion reported that ambivalence about abortion and uncertainty about their last menstrual period were factors related to their second-trimester abortion. Although the survey included women of all ages (n = 398), adolescents made up a quarter of the sample.¹⁷ Delay in terminating a pregnancy increases not only the risk of major complications such as hemorrhage and infection but also the cost of an abortion and limits the availability of providers who can perform abortions later in pregnancy.^10,16,17 As of 2010 in the United States, the median cost of an abortion at 10 weeks was $470, but $1500 at 20 weeks.¹³

Another surprising finding is that 525 nonresident minors obtained an abortion in Arkansas during the study period and 63, or 12%, went through the Arkansas courts to avoid parental involvement. Minors came mostly from Tennessee (n = 202; 38%) and Mississippi (n = 121; 23%). Both states require parental consent, and both have judicial bypass procedures. A major reason minors from these 2 states came to Arkansas could be related to the availability of abortion services at later gestational ages. No abortion clinics in Tennessee or Mississippi provide induced termination after 16 weeks’ gestation (Rachel Jones, director of the Guttmacher Institute’s abortion provider survey, e-mail correspondence, March 31, 2017).

Forty-one percent of abortions to minors from Mississippi and Tennessee performed in Arkansas occurred at 16 weeks’ gestation or later. The figure for resident minors of Arkansas was 6%. Such late terminations represent a potential public health consequence of limited abortion services.¹⁰

Although there were only 100 abortions to Hispanics during the study period, too few from which to generalize, they nevertheless were more likely to use the bypass procedure. Gallup Poll data indicate that Hispanics are more likely to support major restrictions for abortion than are Whites.¹⁸ The bypass procedure may allow Hispanic minors to avoid involving their parents for fear of disappointing them or being pressured to carry to term, 2 major reasons why minors do not involve their parents in states without such laws.^19,20

There was substantial variation across states in the proportion of abortions obtained via a bypass. There were very few abortions obtained through a bypass in Alabama, Georgia, Kansas, Oklahoma, and South Carolina. NARAL Pro-Choice America gave each of these states a grade of F on its annual report card of states’ support for pro-choice policies in 2012. Arkansas and Texas, however, are also given grades of F even though 10% to 15% of abortions to minors in these states were obtained via a bypass.²¹ A possible explanation for the relatively large proportion of bypass cases in Texas is Jane’s Due Process, a nonprofit advocacy organization that provides support and legal advice to minors seeking an abortion.²² Without such independent support groups, personnel at local district courts may be poorly informed about the bypass procedure and, thus, ill prepared to assist minors trying to petition the courts.²³

An important limitation of our study, despite data on the population of minors who obtained an abortion in Arkansas, was that induced termination of pregnancy certificates lack information beyond basic demographics. We had no information on the number of denied petitions, the income level, or family structure of the minor’s household. Nor did we know why the minors sought a bypass or how much time it took from pregnancy recognition to termination. Survey data from minors at abortion clinics suggested that fear of parental reaction, including an effort to prevent the abortion, was one reason for not involving their parents.^19,20 Interviews with judges who handled bypass cases and a small survey of minors who used the bypass to abort suggested that they were mature and clear about their decision.^4,24 These characterizations are consistent with our finding that minors in Arkansas who aborted without parental consent were older and terminated earlier than those who involved their parents.

Another concern when one is using vital statistics on induced abortion is the completeness of reporting. This may be particularly true of bypass cases given their sensitivity and concern for confidentiality. The general quality of abortion surveillance by state health departments varies widely. The number of abortions as reported by numerous state health departments is between 45% and 89% of those reported by Guttmacher Institute.^13,15,25,26 In our sample, however, the total number of abortions performed in Arkansas was 96% of those recorded by the Guttmacher Institute in 2008, 2010, and 2011. Similarly, the reported number of abortions in the other 10 states that recorded bypass cases were 94% of the totals reported by the Guttmacher Institute for the same 3 years.^13,15,25,26 This concordance does not eliminate the possibility of underreporting of bypass cases, but provides some indication of the quality in abortion surveillance in our sample states.

Overall, data from Arkansas underscore the vulnerability of pregnant minors. That 10% of minors in the state feel compelled to use the bypass procedure might point to distress or alienation within families. Twenty-two percent of all minors and 34% of Black non-Hispanic minors who abort a pregnancy in Arkansas terminate in the second trimester. Such late terminations suggest shortcomings in sex education, access to contraception, and the availability of trusting adults in many girls’ lives and awareness of the court bypass option. That many minors late in pregnancy travel from other states to obtain abortions in Arkansas implies that these shortcomings may be present across states.