- Email: [email protected]
Follow-up of pregnant teens at a hospital-based clinic
JOURNAL OF ADOLESCENT HEALTH 1995;17:193-197
FELLOWSHIP
FORUM
Follow-up of Pregnant Teens at a Hospital-Based Clinic S A R A F R I M F O R M A N , M . D . , M A R Y M. A R U D A , R.N., M . S . N . , C - P . N . P . , S. J E A N E M A N S , M . D . , A N D E L I Z A B E T H R. W O O D S , M . D . , M . P . H .
Purpose: Our study investigates the factors associated with length of time to referral appointments for pregnant teens diagnosed in a hospital-based adolescent clinic. Methods: Ninety-six pregnancies diagnosed during a six-month period at a hospital-based adolescent clinic were followed prospectively. Demographic data, plans for pregnancy, last menstrual period (LMP), and w h o accompanied patient to the appointment were recorded by providers. Appointment at referral site was confirmed by the patients' primary provider, contact with the patient, and/or referral site. The time from first positive pregnancy test to kept referral appointment was determined. Results: Of the 96 diagnosed pregnancies, data were available for 94 patients. The mean age was 17.6 ± 2.0 years. The average gestation was 7.4 _~ 4.2 w e e k s (range 4-27 weeks) at initial visit and 11.7 _+ 4.9 weeks (5-30 weeks) at referral site. The time to referral appointment was significantly shorter for patients w h o planned to terminate compared with those w h o planned to continue with the pregnancy (22.4 ~ 16.6 vs. 35.7 ± 24.5 days, P = 0.0042). However, the final decision to continue vs. terminate the pregnancy was a stronger predictor of the interval to kept referral appointment (R 2 = 0.177) than the initial plan. Those w h o planned to tell more people about their pregnancy had a significantly
From tile Division ,,~ AdolesceHt/'~,m~ty Adldt Medicine, Childrc~F~ Hospital, Boston, and Departme~t ,,~ Pcdiatrws, Harvard Medical School, Cambridge, MA. Address correspondence rout ret,itlt requests to: Sara Frith Formml, M.D., Divisiotl of Adolescent/'~'olmx' Adult Medicim', Chihtrc~l'> Ho> pital, Boston, M A 02115. Manuscript accepted November 11, I994 Supported in part by Proh:ct #MCI-MA 259195 from the Mater~lat and Child Health Bureau (Tith' V. Social Security Act), Health Re sources and Services Admim,tratilm. Departme~lt i!f Health mid Htotlatl Services.
higher rate of continuing the pregnancy than those terminating. Conclusion: Pregnant teens may need multiple followup appointments to facilitate connection to referral care. Close follow-up and communication with referral sites needed to optimize care of pregnant teens because of the delay until teens access pregnancy services. KEY WORDS:
Adolescence Teen pregnancy Pregnancy services Pregnancy decisions
Introduction
Each year, a p p r o x i m a t e l y one million adolescents b e c o m e p r e g n a n t and 500,000 live births occur to adolescents in the United States. (1) Referral and follow-up for p r e g n a n t adolescents remains a challenge, especially w h e n adolescents are referred to different health care sites for abortion a n d prenatal services. Adolescents m a y hesitate to seek services early, leading to increased m o r b i d i t y with late terminations and late prenatal care. In addition, adolescents are at higher risk for being u n i n s u r e d compared to older w o m e n and therefore are less likely to receive prenatal care (2). This m a y lead to an increased risk of p o o r o u t c o m e s for adolescent pregnancies. Few p r e v i o u s studies h a v e e x a m i n e d factors contributing to timely referral care, a n d thus w e u n d e r took to d e t e r m i n e the time f r o m first p r e g n a n c y test to kept referral a p p o i n t m e n t for p r e g n a n t teens. O u r goals were to follow p r e g n a n t adolescents until a
© Societyfor Ad~}lescentMedicine, 1 ~ Published bv El~.evierSciencelnc, ~:; Avenue~,~the \mericas, New 5ork, NY Ill010
1054-139X/95/$9.50 SSDI 1054-139X195100004-C
194
FORMANET AL
decision for outcome had been made, to determine referral intervals, and to examine factors associated with the time interval from first p r e g n a n c y test to kept referral appointment.
Methods
All pregnancies diagnosed at the Adolescent Clinic at an urban Children's Hospital in Boston between August 1, 1991 and January, 31, 1992 were prospectively studied. Pregnancy was confirmed by an ICON II (Hybritech) urine pregnancy test. Providers were asked to complete a survey form for every pregnant patient, which included patient demographic information, date of last menstrual period (LMP), plans for the pregnancy, who the patient planned to tell about the pregnancy, and w h o acc o m p a n i e d the patient to the appointment. A follow-up form was filled out for each subsequent visit or telephone conversation until a follow-up appointment was kept at the referral site. The time interval from the visit to the adolescent clinic for the first positive pregnancy test to the first kept referral site appointment was determined by personal contact with the patient or referral site by an adolescent clinic provider. The date of referral a p p o i n t m e n t was recorded and the gestational age at the first referral site visit was calculated from gestational age at the first adolescent clinic visit. The dating of pregnancy gestation at first adolescent clinic visit was determined by LMP in conjunction with bimanual exam at the first visit, or by pelvic or transvaginal ultrasound where indicated. Missing information on patients was obtained by chart review, contact with referral sites, or by subsequent patient contact by providers. This study was a p p r o v e d by the H u m a n Investigation Committee as a quality i m p r o v e m e n t project. The sample was divided into two groups, one group consisted of those w h o terminated the pregnancy, and a second group, all of w h o m planned to continue their pregnancies. The eleven young w o m e n who either had spontaneous abortions (9) or ectopic pregnancies (2) were included in the "planning to keep" group. This decision was m a d e because the majority of teens w h o had spontaneous abortions and ectopic pregnancies had initially planned to continue their pregnancy. During the s t u d y period the average time until next available a p p o i n t m e n t for abortion services was 2 weeks and for prenatal care was 3 weeks. Data analysis was p e r f o r m e d with the SAS statistical package. Fisher's exact test (2 tailed) or X2 test
JOURNAL OF ADOLESCENTHEALTH Vol. 17, No. 3
was used for all categoric variables. For continuous variables the student's t-test for two i n d e p e n d e n t means was p e r f o r m e d w h e n comparison of groups had equal variances (as assessed b y the F-test result) and an approximation of the student's t-test was p e r f o r m e d w h e n comparison groups had unequal variances. All results are reported as m e a n + sd. Pearson correlation coefficients (r) were used to assess the associations between the interval between positive p r e g n a n c y test and kept referral appointment and other interval level variables. This analysis was followed by a stepwise multiple regression analysis.
Results
Ninety-six pregnancies were diagnosed during the six months of the study, however, d e m o g r a p h i c data were not available on two patients. A description of the patients is p r o v i d e d in Tables 1 and 2. The outcome of the pregnancy was k n o w n for 88/96 pregnancies. In spite of extensive attempts to contact patients, eight were lost to follow-up owing to factors such as lack of a telephone or w r o n g address. Three additional patients had no information available other than the pregnancy outcome and thus were included in the "lost to follow-up" group. Race, age, and insurance status had no significant effect on the outcome of the pregnancy (Tables 1
Table 1. Female Adolescents' Decisions to Terminate or Continue Pregnancy Final Choice Terminated Pregnancy
Continued Pregnancy*
N
%**
N
%
28 6 3 2
42.4 46.2 23.1 100.0
38 7 10 0
57.6 53.8 76.9 0.0
NS
20 10 9
46.5 33.3 42.9
23 20 12
53.5 66.7 57.1
NS
26 2 8
74.3 6.1 42.1
9 31 11
25.7 93.9 57.9
.001
Race
Black White Latino Other htsurance Status
Private Medicaid Self pay Plan at first visit
Terminate Continue pregnancy Don't know"
*Includes keeping child, adoption, ectopic pregnancy and miscarriage. **Row percentage.
September 1995
FOLLOW-UPOF PREGNANTTEENS
Table 2. Mean Differences in Health Care Behavior Between Female Adolescents Who Terminated Versus Chose to Continue Their Pregnancies Terminated Pregnancy
Age School grade Interval from pregnancy to positive test (days) Gestation at first visit (weeks) Interval from positive test to referral appointment (days) Gestation at referral appointment (weeks) Number of people planning to tell about pregnancy Number of missed referral appointments
Continued Pregnancy
X
SD
X
SD
p
17.6 10.6
2.2 2.0
17.6 10.5
1.8 1.5
0.82 0.88
52.9
26.6
49.6
23.9
0.52
76
3.6
7.3
4.7
0.66
22.4
16.6
35.7
24.5
11.1
3.8
12.2
5.7
0.30
1.7
I.(~
2.9
1.8
0.0(131
0.14
0.54
0.38
0.62
(1.28
0.0042
and 2). There was a moderate association (contingency coefficient = 0.52) between the initial decision to terminate or continue the p r e g n a n c y and the final outcome (Table 1). Those w h o were undecided at their first visit were slightly more likely to continue the p r e g n a n c y than to terminate. Those w h o continued the p r e g n a n c y planned to tell more (p -< 0.00311 people about the p r e g n a n c y at the initial visit than those w h o terminated the p r e g n a n c y (Table 2). Planning to tell a boyfriend at the first visit did not have significant impact on continuing the pregnancy. Of 60 patients w h o planned to tell a boyfriend at first visit, 40 (67%) continued p r e g n a n c y and 20 (33%) chose to terminate (p = 0.098). There was no relationship between patients planning to tell their m o t h e r about the p r e g n a n c y at the time of the first visit and the o u t c o m e of the pregnancy. There was no significant difference in gestational age at the initial clinic visit between those patients w h o chose to terminate c o m p a r e d to those w h o chose to continue the p r e g n a n c y to term (Table 2). The referral interval from first positive p r e g n a n c y test to first referral site a p p o i n t m e n t was significantly less for those w h o chose termination compared to all others. Those patients w h o planned to tell their mothers at the first visit had significantly more visits to the adolescent clinic (0.8 _+ 0.6 visits vs. 0.3 -+ 0.5 visits, p = 0.00191, but no shorter interval to referral appointment. The interval from first positive p r e g n a n c y test to kept referral a p p o i n t m e n t was significantly corre-
195
lated with gestational age at referral site, n u m b e r of visits to the adolescent clinic prior to keep referral appointment, and n u m b e r of missed referral appointments (Table 3). The interval to kept referral a p p o i n t m e n t was not associated with age, school grade, interval from patients self-report of date of LMP to positive p r e g n a n c y test, gestational age at first adolescent clinic visit, or n u m b e r of telephone calls until kept referral appointment. Stepwise multiple regression analysis was used to determine the influence of plans at the initial p r e g n a n c y test, final decision to terminate or continue the pregnancy, and age of the adolescent on the interval to kept referral appointment. Carrying the p r e g n a n c y to term vs. termination explained 17.7% of the variation in the interval to first kept appointment. Neither of the other two variables accounted for a significant a m o u n t of additional variation in the regression model. Of 11 patients w h o were lost to follow-up, 10 had insurance information recorded. Five of these patients were u n i n s u r e d and five were receiving Medicaid, which was significantly different from the total pregnant population (p = 0.0014). Those lost to follow-up represented 19 % of the total u n i n s u r e d / M e d i c a i d population but only 10% of the total population.
Discussion This s t u d y focused on the follow-up of p r e g n a n t adolescents diagnosed at an urban adolescent clinic in which referral to an outside source was necessary for p r e g n a n c y termination or prenatal care. In genTable 3. Pearson Correlation Coefficients Among Interval from Positive Pregnancy Test to Referral Appointment and Other Variables Interval to Appointment r School grade Age Interval from last menstrual period to positive test Gestational age at referral site Number of visits at adolescent clinic until kept referral appointment Number of telephone calls until kept referral appointment Number of missed referral appointments
p
-.10 -.11
.39 .32
-.05
.67
.51
.0001
.41
.0008
.17
.24
.77
.0001
196
FORMAN ET AL
eral, the length of time for an initial obstetric appointment at our referring hospital is 3 weeks compared to 2 weeks for a first trimester abortion and 3-4 weeks for a second trimester abortion. Adolescent women who decided to continue their pregnancies took more than a month _+ 24 days to keep their first obstetric appointment compared to about 3 weeks for those who choose to terminate the pregnancy. This interval was strongly associated with the number of missed appointments, suggesting that it was not entirely due to differences in the time it took to get an appointment. The final decision to either carry the pregnancy to term vs. terminate the pregnancy was a more important correlate of the length of time to kept appointment than the patient's plan at the time of a positive pregnancy test. However, a higher percentage of adolescents who initially planned to carry to term actually continued their pregnancies than patients who planned to terminate who actually received an abortion. The time interval to kept appointment was not influenced by the patient's age, gestational age at first clinic visit, or the number of follow-up telephone calls made to the patient. Those who planned to tell more people about their pregnancy at their first visit tended to keep the pregnancy. The decision for an abortion in adolescents may be kept private and may be less influenced by others. Several studies have examined the influence of family (especially parents) and friends on the decision-making process when pregnancy is diagnosed. In certain of these studies, family and friends appear to be more influential for those teens who chose to continue than those who chose to terminate (3,4,5). In other studies the association is less clear (5,6). In our study, the majority of teens stated a choice for outcome of their pregnancy at the first visit. The observation that the initial plan tended to be what the adolescent ultimately decided is consistent with another study showing consistency of plan in pregnant Hispanic adolescents (7). Those who were undecided at the first visit were slightly more likely to keep the pregnancy. This correlation can assist physicians in predicting what an adolescent likely needs for services. However, a proportion of our patients did change their decision for pregnancy outcome, which may have implications for general medical treatment. There are some important limitations to our study. The referral interval may be influenced by the local availability of follow-up care. In our area, the waiting time for a prenatal care appointment
JOURNAL OF ADOLESCENT HEALTH Vol. 17, No. 3
can be several weeks, which is usually longer than for abortion services. The availability of services may partially explain the significantly shorter time for termination of a pregnancy than for prenatal care. The overall time interval for referral noted in this study may not be generally applicable to regions with shorter waiting periods for appointments. In addition, the average age of our population as well as the race/ethnic and socioeconomic characteristics, may be different from other clinic populations. However, these variables did not significantly alter the referral interval in our population. Another potential bias is that having providers complete data collections forms may have resulted in a change in practice standards. However, we did not observe a change in the data across the study period. Adolescent pregnancy remains an important problem for the clinician providing health care to teens. Spontaneous abortion and ectopic pregnancy are not uncommon outcomes. Given the significant time interval from a positive pregnancy test to the first referral obstetric appointment, health care providers should provide early prenatal counseling, prescribe prenatal vitamins (8), and perform a complete pelvic examination with screening for sexually transmitted diseases (9,10,11). Many teens need multiple follow-up appointments to facilitate connection to referral care. Alternative solutions are to develop liaisons with prenatal care sites that allow walk-in or next-day appointments for newly diagnosed pregnant adolescents. Close follow-up, frequent appointments, and communication with the referral sites is necessary to optimize the teen's chosen outcome. We would like to thank Robert H. DuRant, Ph.D. for his review and editing and Ruth Connors for her secretarial support during the preparation of this manuscript.
References 1. Facts in Brief, Teenage Sexual and Reproductive Behavior in the United States. The Alan Guttmacher Institute, June 1, 1991. 2. Joseph CL. Identification of factors associated with delayed antenatal care. J Nat Med Assoc 1989;81:57-63. 3. Ortiz CG, Nuttall EV. Adolescent pregnancy: effects of family support, education, and religion on the decision to carry or terminate among Puerto Rican teenagers. Adolescence 1987;88:897-917. 4. Bracken MB, Klerman LV, Dracken M. Abortion, adoption or motherhood: an empirical study of decision making during pregnancy. Am J Obstet Gynecol 1978;130:251. 5. Zabin LS, Hirsc MB, Emerson MR, et al. To whom do innercity minors talk about their pregnancies? Adolescents' com
September 1995
munication with parents and parent surrogates. Faro Plan Persp 1992;24:148-73. 6. Rosen RH. Adolescent pregnancy decision making: are parents important? Adolescence 1980;15:43. 7. Berger DK, Kyman W, Perey G, et al. Hispanic adolescent pregnancy tester: a comparative analysis of negative testers childbearers and aborters. Adolescence 1991 Winter;26:951~2. 8. Centers for Disease Control & Prevention. Recommendations for use of folic acid to reduce the number of cases of spina bifida and other neural tube defects. MMWR Sept 11, 1992;41:RR-14:1-7.
FOLLOW-UP OF PREGNANT TEENS
197
9. Alger LS, Lovchik JC, Hebel JR, et al. The association of chlamydia trachomatis, Neisseria gonorrhoeae and group B streptococci with preterm rupture of the membranes and pregnancy outcome. Am J Obstet Gynecol 1988;159:397-404. 10. Cohen I, Veille JC, Calkins BM. Improved pregnancy outcome following successful treatment of chlamydial infection. JAMA 1990;263:3160-163. 11. Pomeranz AJ, Matson SC, Nelson DB. Delay in obstetrical care in newly diagnosed teenage pregnancy. Clin Pediatr 1991:661-63.
FELLOWSHIP
FORUM
Follow-up of Pregnant Teens at a Hospital-Based Clinic S A R A F R I M F O R M A N , M . D . , M A R Y M. A R U D A , R.N., M . S . N . , C - P . N . P . , S. J E A N E M A N S , M . D . , A N D E L I Z A B E T H R. W O O D S , M . D . , M . P . H .
Purpose: Our study investigates the factors associated with length of time to referral appointments for pregnant teens diagnosed in a hospital-based adolescent clinic. Methods: Ninety-six pregnancies diagnosed during a six-month period at a hospital-based adolescent clinic were followed prospectively. Demographic data, plans for pregnancy, last menstrual period (LMP), and w h o accompanied patient to the appointment were recorded by providers. Appointment at referral site was confirmed by the patients' primary provider, contact with the patient, and/or referral site. The time from first positive pregnancy test to kept referral appointment was determined. Results: Of the 96 diagnosed pregnancies, data were available for 94 patients. The mean age was 17.6 ± 2.0 years. The average gestation was 7.4 _~ 4.2 w e e k s (range 4-27 weeks) at initial visit and 11.7 _+ 4.9 weeks (5-30 weeks) at referral site. The time to referral appointment was significantly shorter for patients w h o planned to terminate compared with those w h o planned to continue with the pregnancy (22.4 ~ 16.6 vs. 35.7 ± 24.5 days, P = 0.0042). However, the final decision to continue vs. terminate the pregnancy was a stronger predictor of the interval to kept referral appointment (R 2 = 0.177) than the initial plan. Those w h o planned to tell more people about their pregnancy had a significantly
From tile Division ,,~ AdolesceHt/'~,m~ty Adldt Medicine, Childrc~F~ Hospital, Boston, and Departme~t ,,~ Pcdiatrws, Harvard Medical School, Cambridge, MA. Address correspondence rout ret,itlt requests to: Sara Frith Formml, M.D., Divisiotl of Adolescent/'~'olmx' Adult Medicim', Chihtrc~l'> Ho> pital, Boston, M A 02115. Manuscript accepted November 11, I994 Supported in part by Proh:ct #MCI-MA 259195 from the Mater~lat and Child Health Bureau (Tith' V. Social Security Act), Health Re sources and Services Admim,tratilm. Departme~lt i!f Health mid Htotlatl Services.
higher rate of continuing the pregnancy than those terminating. Conclusion: Pregnant teens may need multiple followup appointments to facilitate connection to referral care. Close follow-up and communication with referral sites needed to optimize care of pregnant teens because of the delay until teens access pregnancy services. KEY WORDS:
Adolescence Teen pregnancy Pregnancy services Pregnancy decisions
Introduction
Each year, a p p r o x i m a t e l y one million adolescents b e c o m e p r e g n a n t and 500,000 live births occur to adolescents in the United States. (1) Referral and follow-up for p r e g n a n t adolescents remains a challenge, especially w h e n adolescents are referred to different health care sites for abortion a n d prenatal services. Adolescents m a y hesitate to seek services early, leading to increased m o r b i d i t y with late terminations and late prenatal care. In addition, adolescents are at higher risk for being u n i n s u r e d compared to older w o m e n and therefore are less likely to receive prenatal care (2). This m a y lead to an increased risk of p o o r o u t c o m e s for adolescent pregnancies. Few p r e v i o u s studies h a v e e x a m i n e d factors contributing to timely referral care, a n d thus w e u n d e r took to d e t e r m i n e the time f r o m first p r e g n a n c y test to kept referral a p p o i n t m e n t for p r e g n a n t teens. O u r goals were to follow p r e g n a n t adolescents until a
© Societyfor Ad~}lescentMedicine, 1 ~ Published bv El~.evierSciencelnc, ~:; Avenue~,~the \mericas, New 5ork, NY Ill010
1054-139X/95/$9.50 SSDI 1054-139X195100004-C
194
FORMANET AL
decision for outcome had been made, to determine referral intervals, and to examine factors associated with the time interval from first p r e g n a n c y test to kept referral appointment.
Methods
All pregnancies diagnosed at the Adolescent Clinic at an urban Children's Hospital in Boston between August 1, 1991 and January, 31, 1992 were prospectively studied. Pregnancy was confirmed by an ICON II (Hybritech) urine pregnancy test. Providers were asked to complete a survey form for every pregnant patient, which included patient demographic information, date of last menstrual period (LMP), plans for the pregnancy, who the patient planned to tell about the pregnancy, and w h o acc o m p a n i e d the patient to the appointment. A follow-up form was filled out for each subsequent visit or telephone conversation until a follow-up appointment was kept at the referral site. The time interval from the visit to the adolescent clinic for the first positive pregnancy test to the first kept referral site appointment was determined by personal contact with the patient or referral site by an adolescent clinic provider. The date of referral a p p o i n t m e n t was recorded and the gestational age at the first referral site visit was calculated from gestational age at the first adolescent clinic visit. The dating of pregnancy gestation at first adolescent clinic visit was determined by LMP in conjunction with bimanual exam at the first visit, or by pelvic or transvaginal ultrasound where indicated. Missing information on patients was obtained by chart review, contact with referral sites, or by subsequent patient contact by providers. This study was a p p r o v e d by the H u m a n Investigation Committee as a quality i m p r o v e m e n t project. The sample was divided into two groups, one group consisted of those w h o terminated the pregnancy, and a second group, all of w h o m planned to continue their pregnancies. The eleven young w o m e n who either had spontaneous abortions (9) or ectopic pregnancies (2) were included in the "planning to keep" group. This decision was m a d e because the majority of teens w h o had spontaneous abortions and ectopic pregnancies had initially planned to continue their pregnancy. During the s t u d y period the average time until next available a p p o i n t m e n t for abortion services was 2 weeks and for prenatal care was 3 weeks. Data analysis was p e r f o r m e d with the SAS statistical package. Fisher's exact test (2 tailed) or X2 test
JOURNAL OF ADOLESCENTHEALTH Vol. 17, No. 3
was used for all categoric variables. For continuous variables the student's t-test for two i n d e p e n d e n t means was p e r f o r m e d w h e n comparison of groups had equal variances (as assessed b y the F-test result) and an approximation of the student's t-test was p e r f o r m e d w h e n comparison groups had unequal variances. All results are reported as m e a n + sd. Pearson correlation coefficients (r) were used to assess the associations between the interval between positive p r e g n a n c y test and kept referral appointment and other interval level variables. This analysis was followed by a stepwise multiple regression analysis.
Results
Ninety-six pregnancies were diagnosed during the six months of the study, however, d e m o g r a p h i c data were not available on two patients. A description of the patients is p r o v i d e d in Tables 1 and 2. The outcome of the pregnancy was k n o w n for 88/96 pregnancies. In spite of extensive attempts to contact patients, eight were lost to follow-up owing to factors such as lack of a telephone or w r o n g address. Three additional patients had no information available other than the pregnancy outcome and thus were included in the "lost to follow-up" group. Race, age, and insurance status had no significant effect on the outcome of the pregnancy (Tables 1
Table 1. Female Adolescents' Decisions to Terminate or Continue Pregnancy Final Choice Terminated Pregnancy
Continued Pregnancy*
N
%**
N
%
28 6 3 2
42.4 46.2 23.1 100.0
38 7 10 0
57.6 53.8 76.9 0.0
NS
20 10 9
46.5 33.3 42.9
23 20 12
53.5 66.7 57.1
NS
26 2 8
74.3 6.1 42.1
9 31 11
25.7 93.9 57.9
.001
Race
Black White Latino Other htsurance Status
Private Medicaid Self pay Plan at first visit
Terminate Continue pregnancy Don't know"
*Includes keeping child, adoption, ectopic pregnancy and miscarriage. **Row percentage.
September 1995
FOLLOW-UPOF PREGNANTTEENS
Table 2. Mean Differences in Health Care Behavior Between Female Adolescents Who Terminated Versus Chose to Continue Their Pregnancies Terminated Pregnancy
Age School grade Interval from pregnancy to positive test (days) Gestation at first visit (weeks) Interval from positive test to referral appointment (days) Gestation at referral appointment (weeks) Number of people planning to tell about pregnancy Number of missed referral appointments
Continued Pregnancy
X
SD
X
SD
p
17.6 10.6
2.2 2.0
17.6 10.5
1.8 1.5
0.82 0.88
52.9
26.6
49.6
23.9
0.52
76
3.6
7.3
4.7
0.66
22.4
16.6
35.7
24.5
11.1
3.8
12.2
5.7
0.30
1.7
I.(~
2.9
1.8
0.0(131
0.14
0.54
0.38
0.62
(1.28
0.0042
and 2). There was a moderate association (contingency coefficient = 0.52) between the initial decision to terminate or continue the p r e g n a n c y and the final outcome (Table 1). Those w h o were undecided at their first visit were slightly more likely to continue the p r e g n a n c y than to terminate. Those w h o continued the p r e g n a n c y planned to tell more (p -< 0.00311 people about the p r e g n a n c y at the initial visit than those w h o terminated the p r e g n a n c y (Table 2). Planning to tell a boyfriend at the first visit did not have significant impact on continuing the pregnancy. Of 60 patients w h o planned to tell a boyfriend at first visit, 40 (67%) continued p r e g n a n c y and 20 (33%) chose to terminate (p = 0.098). There was no relationship between patients planning to tell their m o t h e r about the p r e g n a n c y at the time of the first visit and the o u t c o m e of the pregnancy. There was no significant difference in gestational age at the initial clinic visit between those patients w h o chose to terminate c o m p a r e d to those w h o chose to continue the p r e g n a n c y to term (Table 2). The referral interval from first positive p r e g n a n c y test to first referral site a p p o i n t m e n t was significantly less for those w h o chose termination compared to all others. Those patients w h o planned to tell their mothers at the first visit had significantly more visits to the adolescent clinic (0.8 _+ 0.6 visits vs. 0.3 -+ 0.5 visits, p = 0.00191, but no shorter interval to referral appointment. The interval from first positive p r e g n a n c y test to kept referral a p p o i n t m e n t was significantly corre-
195
lated with gestational age at referral site, n u m b e r of visits to the adolescent clinic prior to keep referral appointment, and n u m b e r of missed referral appointments (Table 3). The interval to kept referral a p p o i n t m e n t was not associated with age, school grade, interval from patients self-report of date of LMP to positive p r e g n a n c y test, gestational age at first adolescent clinic visit, or n u m b e r of telephone calls until kept referral appointment. Stepwise multiple regression analysis was used to determine the influence of plans at the initial p r e g n a n c y test, final decision to terminate or continue the pregnancy, and age of the adolescent on the interval to kept referral appointment. Carrying the p r e g n a n c y to term vs. termination explained 17.7% of the variation in the interval to first kept appointment. Neither of the other two variables accounted for a significant a m o u n t of additional variation in the regression model. Of 11 patients w h o were lost to follow-up, 10 had insurance information recorded. Five of these patients were u n i n s u r e d and five were receiving Medicaid, which was significantly different from the total pregnant population (p = 0.0014). Those lost to follow-up represented 19 % of the total u n i n s u r e d / M e d i c a i d population but only 10% of the total population.
Discussion This s t u d y focused on the follow-up of p r e g n a n t adolescents diagnosed at an urban adolescent clinic in which referral to an outside source was necessary for p r e g n a n c y termination or prenatal care. In genTable 3. Pearson Correlation Coefficients Among Interval from Positive Pregnancy Test to Referral Appointment and Other Variables Interval to Appointment r School grade Age Interval from last menstrual period to positive test Gestational age at referral site Number of visits at adolescent clinic until kept referral appointment Number of telephone calls until kept referral appointment Number of missed referral appointments
p
-.10 -.11
.39 .32
-.05
.67
.51
.0001
.41
.0008
.17
.24
.77
.0001
196
FORMAN ET AL
eral, the length of time for an initial obstetric appointment at our referring hospital is 3 weeks compared to 2 weeks for a first trimester abortion and 3-4 weeks for a second trimester abortion. Adolescent women who decided to continue their pregnancies took more than a month _+ 24 days to keep their first obstetric appointment compared to about 3 weeks for those who choose to terminate the pregnancy. This interval was strongly associated with the number of missed appointments, suggesting that it was not entirely due to differences in the time it took to get an appointment. The final decision to either carry the pregnancy to term vs. terminate the pregnancy was a more important correlate of the length of time to kept appointment than the patient's plan at the time of a positive pregnancy test. However, a higher percentage of adolescents who initially planned to carry to term actually continued their pregnancies than patients who planned to terminate who actually received an abortion. The time interval to kept appointment was not influenced by the patient's age, gestational age at first clinic visit, or the number of follow-up telephone calls made to the patient. Those who planned to tell more people about their pregnancy at their first visit tended to keep the pregnancy. The decision for an abortion in adolescents may be kept private and may be less influenced by others. Several studies have examined the influence of family (especially parents) and friends on the decision-making process when pregnancy is diagnosed. In certain of these studies, family and friends appear to be more influential for those teens who chose to continue than those who chose to terminate (3,4,5). In other studies the association is less clear (5,6). In our study, the majority of teens stated a choice for outcome of their pregnancy at the first visit. The observation that the initial plan tended to be what the adolescent ultimately decided is consistent with another study showing consistency of plan in pregnant Hispanic adolescents (7). Those who were undecided at the first visit were slightly more likely to keep the pregnancy. This correlation can assist physicians in predicting what an adolescent likely needs for services. However, a proportion of our patients did change their decision for pregnancy outcome, which may have implications for general medical treatment. There are some important limitations to our study. The referral interval may be influenced by the local availability of follow-up care. In our area, the waiting time for a prenatal care appointment
JOURNAL OF ADOLESCENT HEALTH Vol. 17, No. 3
can be several weeks, which is usually longer than for abortion services. The availability of services may partially explain the significantly shorter time for termination of a pregnancy than for prenatal care. The overall time interval for referral noted in this study may not be generally applicable to regions with shorter waiting periods for appointments. In addition, the average age of our population as well as the race/ethnic and socioeconomic characteristics, may be different from other clinic populations. However, these variables did not significantly alter the referral interval in our population. Another potential bias is that having providers complete data collections forms may have resulted in a change in practice standards. However, we did not observe a change in the data across the study period. Adolescent pregnancy remains an important problem for the clinician providing health care to teens. Spontaneous abortion and ectopic pregnancy are not uncommon outcomes. Given the significant time interval from a positive pregnancy test to the first referral obstetric appointment, health care providers should provide early prenatal counseling, prescribe prenatal vitamins (8), and perform a complete pelvic examination with screening for sexually transmitted diseases (9,10,11). Many teens need multiple follow-up appointments to facilitate connection to referral care. Alternative solutions are to develop liaisons with prenatal care sites that allow walk-in or next-day appointments for newly diagnosed pregnant adolescents. Close follow-up, frequent appointments, and communication with the referral sites is necessary to optimize the teen's chosen outcome. We would like to thank Robert H. DuRant, Ph.D. for his review and editing and Ruth Connors for her secretarial support during the preparation of this manuscript.
References 1. Facts in Brief, Teenage Sexual and Reproductive Behavior in the United States. The Alan Guttmacher Institute, June 1, 1991. 2. Joseph CL. Identification of factors associated with delayed antenatal care. J Nat Med Assoc 1989;81:57-63. 3. Ortiz CG, Nuttall EV. Adolescent pregnancy: effects of family support, education, and religion on the decision to carry or terminate among Puerto Rican teenagers. Adolescence 1987;88:897-917. 4. Bracken MB, Klerman LV, Dracken M. Abortion, adoption or motherhood: an empirical study of decision making during pregnancy. Am J Obstet Gynecol 1978;130:251. 5. Zabin LS, Hirsc MB, Emerson MR, et al. To whom do innercity minors talk about their pregnancies? Adolescents' com
September 1995
munication with parents and parent surrogates. Faro Plan Persp 1992;24:148-73. 6. Rosen RH. Adolescent pregnancy decision making: are parents important? Adolescence 1980;15:43. 7. Berger DK, Kyman W, Perey G, et al. Hispanic adolescent pregnancy tester: a comparative analysis of negative testers childbearers and aborters. Adolescence 1991 Winter;26:951~2. 8. Centers for Disease Control & Prevention. Recommendations for use of folic acid to reduce the number of cases of spina bifida and other neural tube defects. MMWR Sept 11, 1992;41:RR-14:1-7.
FOLLOW-UP OF PREGNANT TEENS
197
9. Alger LS, Lovchik JC, Hebel JR, et al. The association of chlamydia trachomatis, Neisseria gonorrhoeae and group B streptococci with preterm rupture of the membranes and pregnancy outcome. Am J Obstet Gynecol 1988;159:397-404. 10. Cohen I, Veille JC, Calkins BM. Improved pregnancy outcome following successful treatment of chlamydial infection. JAMA 1990;263:3160-163. 11. Pomeranz AJ, Matson SC, Nelson DB. Delay in obstetrical care in newly diagnosed teenage pregnancy. Clin Pediatr 1991:661-63.