- Email: [email protected]
Early adolescent pregnancy: A comparative study of pregnancy outcome in young adolescents and mature women
JOURNAL OF ADOLESCENT HEALTH CARE 1984;5:167-171
Early Adolescent Pregnancy: A Comparative Study of Pregnancy Outcome in Young Adolescents and Mature Women THERESA O. SCHOLL, Ph.D., M . P . H . , E D M U N D DECKER, B.S., ROBERT J. KARP, M.D., GEOFFREY GREENE, M.S., M . P . H . , A N D MARIE DE SALES, S.S.N.D.
Thirty-two adolescents were matched to mature controls for a study of variables k n o w n to influence the course and outcome of pregnancy. These included: ethnicity; clinic payment status; year of delivery; parity; and presence or absence of alcohol and tobacco use. The comparative factors examined included: age of menarche; gynecologic age; admission hemoglobin; prepregnant weight and height; weight gain during pregnancy; complications of pregnancy; length of gestation; birth weight; 1- and 5minute apgar scores; and the presence or absence of congenital defects. There was no significant difference between young adolescents and matched controls in infant birth weight or apgar scores when the above confounding factors were controlied. Young gravidae, however, had significantly shorter gestations, earlier menarche, lower hemoglobin levels, and poorer weight gain during pregnancy than mature women. Multiple-regression analysis suggests that 1) pregnancy weight gain was associated with the trimester the gravida enrolled in the Supplemental Food Program for Women, Infants, and Children (WIC) (p = 0.008) and maternal stature (p = 0.012); 2) length o f gestation was associated with maternal stature (p = 0.022) and prepregnant weight (p = 0.011); and 3) maternal hemoglobin was associated with birth weight (p = 0.087). Alternative interpretations are discussed.
From the Department of Family Practice, University of Medicine and Denistry of New Jersey-School of Osteopathic Medicine (T.O.S., E.D., R.].K., G.G.) St. John the Baptist Prenatal Clinic (M.S.), and the Area Health Education Center (G.G.). This work was supported in part by Grant #13.995from the Office of Adolescent Pregnancy Programs. Address reprint requests to: Dr. Theresa Scholl, Department of Family Practice, Division of Community Medicine, University of Medicine and Dentistry of New Jersey-New Jersey School of Osteopathic Medicine, 300 Broadway, Camden, NJ 08103. Manuscript accepted October 11, 1983.
KEY WORDS:
Early adolescent pregnancy Maternal nutritional status Pregnancy outcome Birth weight Most studies of pregnancy in adolescents have shown a higher risk of complications and poor outcomes than in older gravidae (1-3). Despite some evidence to the contrary, it has been believed that the excess of materna ! and fetal problems experienced by the adolescent was an affect of biological immaturity (4-6). While adolescent pregnancy occurs in almost every ethnic, social, and economic grouping, adolescents who give birth are likely to come from minority backgrounds characterized by poverty and disadvantage (7,8). Recent studies that have imposed controls for these factors suggest that while the older adolescent fares no worse than the older gravida of the same SES (7-10), their outcomes are generally not as good as those of more privileged w o m e n (il). In younger adolescents (~ 15 years), the effects of early maternity on the course and outcome of pregnancy are still in question (12). The present study was designed to compare differences in pregnancy outcomes and maternal factors in young teenagers and mature women from the same social and economic circumstances after placement of controls for several confounding variables that also influence the course and outcome of pregnancy.
Patients and Methods The study group consisted of w o m e n enrolled for prenatal care at St. John the Baptist Prenatal Clinic,
© Society for Adolescent Medicine, 1984 Published by Elsevier Science Publishing Co., Inc., 52 Vanderbilt Ave., New York, NY 10017
167 0197-0070/84/$3.00
168
SCHOLL ET AL.
JOURNAL OF ADOLESCENT HEALTH CARE Vol. 5, No. 3
Camden, New Jersey, who gave birth to a live infant during 1980. St. John's is a prenatal center that provides services to low-income Hispanic and Black w o m e n from the inner-city. Most are welfare recipients and approximately 90% have been defined as "at nutritional risk" and enrolled in the Supplemental Food Program for Women, Infants, and Children (WIC). Data on all the St. John's obstetric patients were collected using a standard format. Data on pregnancy course and outcome were retrospectively abstracted from the obstetrical records of 32 adolescents, aged 15 years or younger at the time of their registration for prenatal care, and 32 controls aged 20 years or older. Young adolescents and controls were matched for several factors reported to be associated with pregnancy course and outcome (13). These included ethnicity, clinic paymen t status, year of delivery, parity, and the presence or absence of maternal smoking and drinking at registration. All of the young adolescents with available records were ma.tched to a control according to these six criteria. Maternal factors examined in the study included age of menarche, gynecological age, hemoglobin level at registration, prepregnant weight and height, weight gain during pregnancy, and complications of pregnancy (e. g., hypertension, eclampsia, preeclampsia, diabetes ) . Fetal factor s included length of gestation, birth weight, apgar score (1- and 5-minute), and presence or absence of congenital defects. Comparisons for maternal and fetal factors between young adolescents and controls were m a d e using t-tests (continuous data) and z-tests, Fisher's Exact Test, and the Mantel-Haenszel test (14) (categorical data). Factors associated with significant differences between adolescents a n d controls (excluding menarche and gynecologic age) were Studied using multiple-regression analysis after adjusting for the six control variables and maternal age. In addition, the weight gain regression was used as a con-
trol for the length of gestation and the gestation regression was used as a control for pregnancy weight gain. Other variables were chosen using an all-possible-subsets routine. Regressions were tested for statistical significance after adjusting for the control variables, maternal age, and all other variables in the'model.
Results Analyses involving maternal factors are shown in Table 1. There were significant differences between the adolescent and older gravidae in age of menarche (11.3 versus 12.7 years), in hemoglobin on admission (11.4 mg/100 ml versus 11.9 mg/100 ml), in gynecologic age (3.2 versus 8.8 years), and in low pregnancy weight gain (< 7 kilos). In addition, the adolescents tended to have lower prepregnant weight, shorter stature, and fewer risk factors than the mature gravidae, although these differences were not statistically significant. Because hemoglobin levels decline as gestation progresses, adjustments were made for the effect of gestation at registration on hemoglobin by analysis of covariance. After adjustment, the significantly lower hemoglobin levels in young adolescent gravidae persisted. However, this difference is probably not clinically significant. Poor pregnancy weight gain may be confounded by length of gestation (ile., early delivery may curtail gains). After stratifying by length of gestation (Mantel'Haenszel Test) differences between young adolescents and controls remained significant. Fetal outcomes are shown in Table 2. There were nosignificant differences between the infants born to adolescents and to the older women for birth weight and 1- and 5-minute apgar scores. On the other hand, adolescents had significantly shorter gestations. Cross-classification of birth weight and gestation suggest that all infants born to adolescents were ap-
Table 1. Maternal Characteristics--32 Young Adolescents (-< 15 years) versus 32 Mature Gravidae (>- 20 years) Maternal Factors Age at menarche ~ Hemoglobin on admission (rag/100 ml)~' Gynecological age c Prepregnan! weight (kilos) Height (cm) Complications of pregnancy Low pregnancy weight gain (< 7 kilos)a ap < 0.001, t-test bp ~ 0.05, t-test cp K 0.01, t-test ap K 0.05, z-test
Adolescents (,X _+ SD) 1113 (+ 11.4 (± 3.2 (± 52.6 (± 157.8 (±
Percent
113) 0.9) 1.34) 7.0) 8.7)
Mature Women (X + SD) 12.7 1i.9 8.8 55.8 160.9
6.25 (2/32) 15.7 (5/32)
Percent
(-----1.8) (± Q.9) (± 2.6) (± 8.9) (± 8.1) 12.5 (4/32) 0 (0/32)
July 1984
OUTCOME IN ADOLESCENTPREGNANCY
169
Table 2. Fetal Outcomes: 32 Young Adolescents (-< 15 years) versus 32 Mature Gravidae (-> 20 years) Outcomes
Adolescents (X _+ SD)
Birth weight (g) Birth weight < 2500 g (%) Apgar Scores: 1 minute 5 minute Births -< 36 Weeks (%)" Congenital Defects (%)b
Percent
Mature Women (,X ± SD)
3205.7 (± 372.4)
Percent
3195.8 (_+ 466.1) 6.25 (2/32)
3.1 (1/32)
8.2 (_+ 1.3) 9.0 (± 0.3)
7.8 (± 1.5) 8.9 (± 0.8) 21.9 (7/32) 6.25 (2/30)
0 (0/32) 0 (0/32)
ap < 0.01, z-test bCongenital defects include: Bell's palsy (one case) and congenital hip dislocation (one case) propriate for gestational age, whereas one small-forgestational-age infant was born to a control. Table 3 shows the factors associated with gestation, pregnancy weight gain, a n d hemoglobin at registration. For pregnancy weight gain (R2 = 32.2), significant additional factors included the trimester the gravida enrolled in WIC (p = 0.008) and the maternal height at registration (p = 0.012). Maternal prepregnant weight a n d maternal risk factors entered but did not retain significance after adjusting for the variables already included. Significant prognostic factors for gestation ( R 2 = 22.9) included maternal prepregnant weight (p = 0.011) and maternal height (p = 0.022). For hemoglobin (R2 -- 26.6), the only additional factor that was of borderline significance was infant birth weight (p = 0.087). Gynecologic age was tested in all models for effects. Preliminary univariable analyses s h o w e d a significant association with gestation (r = 0.33, p = 0.045) but the effect was not significant after adjustment.
Discussion The present s t u d y examined the relationship between maternal age, pregnancy outcome, and maternal factors using the data from registrants at an
inner-city prenatal facility. Care was taken to provide controls for several characteristics that m a y potentially confound outcomes a n d lead to spurious differences.
Pregnancy Outcome We found no significant differences in birth weight or apgar scores between young adolescents and older w o m e n w h o received prenatal care. These results agree with Rothenberg's s t u d y (9)of older adolescents showing that outcomes were no worse for young mothers in comparison to older mothers w h e n the social and behavioral factors affecting pregnancy were controlled. We found shortened gestation in our y o u n g adolescent women. The consistency of this observation across studies where outcomes have been determined by different m e t h o d s (5,6,15,16) makes bias seem unlikely, but does not exclude the possibility. Studies reporting an increased incidence of infants below 2500 g in y o u n g adolescents m a y be reporting an increased incidence of shortened gestation, a longrecognized cause of low birth weight (17). All of the infants born to our y o u n g adolescents had birth weights appropriate for gestation. The two infants weighing less t h a n 2500 g were born to adolescent mothers at -< 36 weeks gestation. Multipleregression analysis suggested that increased mater-
Table 3. Factors Associated with Gestation, Pregnancy Weight Gain, and Hemoglobin at Registration Dependent variable
Variables in model~
Pregnancy weight gain
TrimesterWIC supplementation Maternal height
Gestation
Maternal height Prepregnant weight
Hemoglobin on admission
Birth weight
Estimated coefficient(/3)
Standard error (/~)
p
-2.43 0.240
0.887 0.091
0.008 0.012
- 0.100 0.118
0.042 0.044
0.02 0.01
0.0001
0.08
0.0004
R2 (model)
32.2 22.9 26.6
aControl variables included: ethnicity; clinic payment status; year of delivery; parity; presence or absence of drinking, smoking at registration; maternal age; weight gain (for gestation model); and gestation (for weight gain model).
170
SCHOLL ET AL.
nal height and reduced maternal prepregnant weight are associated with shortened gestation. These prognostic factors are in agreement with reported maternal characteristics associated with preterm birth (18). Maternal infection has also been implicated in preterm birth for older women, but its effect on young adolescent gestation is unknown. Maternal Characteristics
There were a number of differences in maternal characteristics when young adolescent gravidae were compared to their older controls. Most of these seem to reflect underlying differences in maternal nutritional status, growth, and development. Young mothers menstruate earlier than older gravidae (2,15). That is, in part, an artifact of a truncated (i.e., incomplete) maternal age distribution. However, it may be inferred that many young adolescents are biologically advanced for chronological age, with a larger amount of postmenarcheal growth than later-maturing girls (19,20). As a group, adolescents tend to have irregular eating habits and consume a greater proportion of calories as snacks with a lower density of nutrients than older w o m e n (21); moreover, adolescents have higher nutritional requirements (22). Most dietary studies of adolescent gravidae have reported low calorie intakes and low intakes of micronutrients; many have reported biochemical evidence of poor nutritional status (23,24). Therefore, our finding that poor pregnancy weight gain and lower hemoglobin levels on registration were more common in young adolescent gravidae w h e n compared to the older gravidae with similar social and economic circumstances is not surprising. Regression analyses suggest that pregnancy weight gain is associated with the trimester the gravida enrolled in WIC. Thus, one important element in attaining adequate pregnancy weight gain in an urban poor population may be an early start in a supplementation program. While evidence is scanty, there are some indications that younger adolescents may be more responsive to nutrition supplementation during pregnancy than older women. For instance, Paige has shown a 269 g increase in the birth weight of infants born to young adolescents (< 16 years) who were supplemented with Sustacal (25). The increase obtained in older adolescents (87 g) was comparable to previous reports in adults (26). An alternative interpretation is that the relationship demonstrated in our study is not a causal one. That is, women who seek prenatal care and obtain WIC
JOURNAL OF ADOLESCENT HEALTH CARE Vol. 5, No. 3
early in pregnancy may differ in other unmeasured factors associated with increased pregnancy weight gain in comparison to women who do not seek care until later in their pregnancy. Maternal stature appears to be associated with pregnancy weight gain. Shorter women, because of their reduced body size, may normally gain less weight during pregnancy than taller women (27). In the present study, there is the additional concern that the effect on weight gain may be influenced by adolescent growth during the pregnancy. A recent record-based report suggested that young adolescent mothers are not fully mature but continue to grow from one pregnancy to the next (15). Although differences in our study were not statistically significant, the smaller body size of young gravidae (3.2 kilos less in prepregnant weight and 2.3 cm less in height than controls) is consistent with this and with data reported by Garn and Petzold (27). In summary, our data indicate little difference in pregnancy outcome between young adolescents and mature women who receive prenatal care except for the length of gestation. On the other hand, young adolescent gravidae had lower hemoglobin levels on admission and a greater frequency of poor weight gain than older w o m e n living under similar social and economic conditions. These indicators of poor nutritional status may be common to all young low SES teenagers or may be a condition that is exacerbated by childbearing in early adolescence. We thank the staff of St. John the Baptist Prenatal Clinic, Diocese of Camden for facilitating our work and Sharon Pavelich for assisting with the research.
References 1. Clark J. Adolescent obstetrics: Obstetric and sociologic implications. Clin Obstet Gynecol 1971;14:1026-36. 2. Hassan HM, Falls FH. The young primipara. Am J Obstet Gynecol 1964;88:256-69. 3. Dott AB, Fort AT. Medical and social factors affecting early teenage pregnancy. Am J Obstet Gynecol 1976;125:532-6. 4. Hutchins FL, Kendall N, Rubino J. Experience with teenage pregnancy. Obstet Gynecol 1979;54:1-5. 5. Dwyer JF. Teenage pregnancy. Am J Obstet Gynecol 1976;118:373-6. 6. Battaglia FC, Frazier TM, Hellegers AE. Obstetric and pediatric complications of juvenile pregnancy. Pediatrics 1963;32:902-10. 7. Millman S, Mosher WD. Selected demographic characteristics of teenage wives and mothers. Advance Data (DHEW Publ. No. (PHS) 80-1250) 1980;61. 8. Venture SJ. Teenage childbearing: United States 1966-1975. Monthly Vital Statistics Report (DHEW Publ. No. (HRA) 771120) 1977;25/15. 9. Rothenberg PB, Varga PE. The relationship between age of
July 1984
10. 11.
12. 13.
14.
15.
16.
17. 18. 19.
mother and child health and development. Am J Publ Health 1981;71:810-17. Osbourne GK, Howat RC, Jordan MM. The obstetric outcome of teenage pregnancy. Br J Obstet Gynecol 1981;88:215-21. Department of Health and Human Services. Factors associated with low birth weight: United States, 1976. DHEW Publ. No. (PHS) 80-1915, National Center for Health Statistics, Hyattsville, MD, 1980. Hollingsworth DR, Kessler AK. Teenage pregnancy solutions are evolving. N Engl J Med 1980;303:516-18. Bloom AD. Guidelines for studies of human populations exposed to mutagenic and reproductive hazards. March of Dimes Birth Defects Foundation, 1981. Mantel N, Haenzel W. Statistical aspects of the analysis of data from retrospective studies of disease. J Nail Cancer Inst 1959;22:719-23. Sukanich AC. Measures of physical maturity and outcome of pregnancy in girls less than 16 years of age. Unpublished dissertation. Graduate School of Public Health, University of Pittsburgh, Pittsburgh, Pennsylvania, 1981. Horon IL, Strobino DM, MacDonald H. Birthweights among infants born to adolescent and young adult women. Am J Obstet Gynecol 1983;146:444-9. Lubchenko LO. Assessment of gestational age and development at birth. Ped Clin North Am 1970;17:125-45. Berkowitz GS. An epidemiologic study of preterm delivery. Am J Epidemiol 1981;113:81-6. Roche AF, Davila GH. Later adolescent growth in stature. Pediatrics 1972;50:874-6.
OUTCOME IN ADOLESCENT PREGNANCY
171
20. Faust MS. Somatic development of adolescent girls. Monographs of the Society for Research in Child Development 1977;169/42:1. 21. Nelson M. Dietary practices of adolescents. In: Winick M, ed., Adolescent Nutrition. New York, John Wiley and Sons: 1982;35-44. 22. National Research Council. Recommended Daily Allowances, 1980, (9th Edition). National Academy of Science, Washington, 1980. 23. Cherry F, Bennett E, Bazzano GS, et al. Plasma zinc in hypertension, toxemia, and other reproductive variables in adolescent pregnancy. Amer J Clin Nutr 1981;34:2367-75. 24. Carruth BR. Smoking and pregnancy outcome of adolescents. J Adol Health Care 1981;2:115-20. 25. Paige DM, Cordano A, Mellits ED, et al. Nutritional supplementation of pregnant adolescents. J Adol Health Care 1981;1:261-5. 26. Rush D. Nutritional services during pregnancy and birthweight: A retrospective matched pair analysis. Can Med Assoc J 1981;125:567-75. 27. Gueri M, Jutsum P, Sorhaindo B. Anthropometric assessment of nutritional status in pregnant women: A reference table of weight-for-height by week of pregnancy. Am J Clin Nutr 1982;35:609-16. 28. Garn SM, Petzold AS. Characteristics of the mother and child in teenage pregnancy. Am J Dis Child 1983;137:365-8.
Early Adolescent Pregnancy: A Comparative Study of Pregnancy Outcome in Young Adolescents and Mature Women THERESA O. SCHOLL, Ph.D., M . P . H . , E D M U N D DECKER, B.S., ROBERT J. KARP, M.D., GEOFFREY GREENE, M.S., M . P . H . , A N D MARIE DE SALES, S.S.N.D.
Thirty-two adolescents were matched to mature controls for a study of variables k n o w n to influence the course and outcome of pregnancy. These included: ethnicity; clinic payment status; year of delivery; parity; and presence or absence of alcohol and tobacco use. The comparative factors examined included: age of menarche; gynecologic age; admission hemoglobin; prepregnant weight and height; weight gain during pregnancy; complications of pregnancy; length of gestation; birth weight; 1- and 5minute apgar scores; and the presence or absence of congenital defects. There was no significant difference between young adolescents and matched controls in infant birth weight or apgar scores when the above confounding factors were controlied. Young gravidae, however, had significantly shorter gestations, earlier menarche, lower hemoglobin levels, and poorer weight gain during pregnancy than mature women. Multiple-regression analysis suggests that 1) pregnancy weight gain was associated with the trimester the gravida enrolled in the Supplemental Food Program for Women, Infants, and Children (WIC) (p = 0.008) and maternal stature (p = 0.012); 2) length o f gestation was associated with maternal stature (p = 0.022) and prepregnant weight (p = 0.011); and 3) maternal hemoglobin was associated with birth weight (p = 0.087). Alternative interpretations are discussed.
From the Department of Family Practice, University of Medicine and Denistry of New Jersey-School of Osteopathic Medicine (T.O.S., E.D., R.].K., G.G.) St. John the Baptist Prenatal Clinic (M.S.), and the Area Health Education Center (G.G.). This work was supported in part by Grant #13.995from the Office of Adolescent Pregnancy Programs. Address reprint requests to: Dr. Theresa Scholl, Department of Family Practice, Division of Community Medicine, University of Medicine and Dentistry of New Jersey-New Jersey School of Osteopathic Medicine, 300 Broadway, Camden, NJ 08103. Manuscript accepted October 11, 1983.
KEY WORDS:
Early adolescent pregnancy Maternal nutritional status Pregnancy outcome Birth weight Most studies of pregnancy in adolescents have shown a higher risk of complications and poor outcomes than in older gravidae (1-3). Despite some evidence to the contrary, it has been believed that the excess of materna ! and fetal problems experienced by the adolescent was an affect of biological immaturity (4-6). While adolescent pregnancy occurs in almost every ethnic, social, and economic grouping, adolescents who give birth are likely to come from minority backgrounds characterized by poverty and disadvantage (7,8). Recent studies that have imposed controls for these factors suggest that while the older adolescent fares no worse than the older gravida of the same SES (7-10), their outcomes are generally not as good as those of more privileged w o m e n (il). In younger adolescents (~ 15 years), the effects of early maternity on the course and outcome of pregnancy are still in question (12). The present study was designed to compare differences in pregnancy outcomes and maternal factors in young teenagers and mature women from the same social and economic circumstances after placement of controls for several confounding variables that also influence the course and outcome of pregnancy.
Patients and Methods The study group consisted of w o m e n enrolled for prenatal care at St. John the Baptist Prenatal Clinic,
© Society for Adolescent Medicine, 1984 Published by Elsevier Science Publishing Co., Inc., 52 Vanderbilt Ave., New York, NY 10017
167 0197-0070/84/$3.00
168
SCHOLL ET AL.
JOURNAL OF ADOLESCENT HEALTH CARE Vol. 5, No. 3
Camden, New Jersey, who gave birth to a live infant during 1980. St. John's is a prenatal center that provides services to low-income Hispanic and Black w o m e n from the inner-city. Most are welfare recipients and approximately 90% have been defined as "at nutritional risk" and enrolled in the Supplemental Food Program for Women, Infants, and Children (WIC). Data on all the St. John's obstetric patients were collected using a standard format. Data on pregnancy course and outcome were retrospectively abstracted from the obstetrical records of 32 adolescents, aged 15 years or younger at the time of their registration for prenatal care, and 32 controls aged 20 years or older. Young adolescents and controls were matched for several factors reported to be associated with pregnancy course and outcome (13). These included ethnicity, clinic paymen t status, year of delivery, parity, and the presence or absence of maternal smoking and drinking at registration. All of the young adolescents with available records were ma.tched to a control according to these six criteria. Maternal factors examined in the study included age of menarche, gynecological age, hemoglobin level at registration, prepregnant weight and height, weight gain during pregnancy, and complications of pregnancy (e. g., hypertension, eclampsia, preeclampsia, diabetes ) . Fetal factor s included length of gestation, birth weight, apgar score (1- and 5-minute), and presence or absence of congenital defects. Comparisons for maternal and fetal factors between young adolescents and controls were m a d e using t-tests (continuous data) and z-tests, Fisher's Exact Test, and the Mantel-Haenszel test (14) (categorical data). Factors associated with significant differences between adolescents a n d controls (excluding menarche and gynecologic age) were Studied using multiple-regression analysis after adjusting for the six control variables and maternal age. In addition, the weight gain regression was used as a con-
trol for the length of gestation and the gestation regression was used as a control for pregnancy weight gain. Other variables were chosen using an all-possible-subsets routine. Regressions were tested for statistical significance after adjusting for the control variables, maternal age, and all other variables in the'model.
Results Analyses involving maternal factors are shown in Table 1. There were significant differences between the adolescent and older gravidae in age of menarche (11.3 versus 12.7 years), in hemoglobin on admission (11.4 mg/100 ml versus 11.9 mg/100 ml), in gynecologic age (3.2 versus 8.8 years), and in low pregnancy weight gain (< 7 kilos). In addition, the adolescents tended to have lower prepregnant weight, shorter stature, and fewer risk factors than the mature gravidae, although these differences were not statistically significant. Because hemoglobin levels decline as gestation progresses, adjustments were made for the effect of gestation at registration on hemoglobin by analysis of covariance. After adjustment, the significantly lower hemoglobin levels in young adolescent gravidae persisted. However, this difference is probably not clinically significant. Poor pregnancy weight gain may be confounded by length of gestation (ile., early delivery may curtail gains). After stratifying by length of gestation (Mantel'Haenszel Test) differences between young adolescents and controls remained significant. Fetal outcomes are shown in Table 2. There were nosignificant differences between the infants born to adolescents and to the older women for birth weight and 1- and 5-minute apgar scores. On the other hand, adolescents had significantly shorter gestations. Cross-classification of birth weight and gestation suggest that all infants born to adolescents were ap-
Table 1. Maternal Characteristics--32 Young Adolescents (-< 15 years) versus 32 Mature Gravidae (>- 20 years) Maternal Factors Age at menarche ~ Hemoglobin on admission (rag/100 ml)~' Gynecological age c Prepregnan! weight (kilos) Height (cm) Complications of pregnancy Low pregnancy weight gain (< 7 kilos)a ap < 0.001, t-test bp ~ 0.05, t-test cp K 0.01, t-test ap K 0.05, z-test
Adolescents (,X _+ SD) 1113 (+ 11.4 (± 3.2 (± 52.6 (± 157.8 (±
Percent
113) 0.9) 1.34) 7.0) 8.7)
Mature Women (X + SD) 12.7 1i.9 8.8 55.8 160.9
6.25 (2/32) 15.7 (5/32)
Percent
(-----1.8) (± Q.9) (± 2.6) (± 8.9) (± 8.1) 12.5 (4/32) 0 (0/32)
July 1984
OUTCOME IN ADOLESCENTPREGNANCY
169
Table 2. Fetal Outcomes: 32 Young Adolescents (-< 15 years) versus 32 Mature Gravidae (-> 20 years) Outcomes
Adolescents (X _+ SD)
Birth weight (g) Birth weight < 2500 g (%) Apgar Scores: 1 minute 5 minute Births -< 36 Weeks (%)" Congenital Defects (%)b
Percent
Mature Women (,X ± SD)
3205.7 (± 372.4)
Percent
3195.8 (_+ 466.1) 6.25 (2/32)
3.1 (1/32)
8.2 (_+ 1.3) 9.0 (± 0.3)
7.8 (± 1.5) 8.9 (± 0.8) 21.9 (7/32) 6.25 (2/30)
0 (0/32) 0 (0/32)
ap < 0.01, z-test bCongenital defects include: Bell's palsy (one case) and congenital hip dislocation (one case) propriate for gestational age, whereas one small-forgestational-age infant was born to a control. Table 3 shows the factors associated with gestation, pregnancy weight gain, a n d hemoglobin at registration. For pregnancy weight gain (R2 = 32.2), significant additional factors included the trimester the gravida enrolled in WIC (p = 0.008) and the maternal height at registration (p = 0.012). Maternal prepregnant weight a n d maternal risk factors entered but did not retain significance after adjusting for the variables already included. Significant prognostic factors for gestation ( R 2 = 22.9) included maternal prepregnant weight (p = 0.011) and maternal height (p = 0.022). For hemoglobin (R2 -- 26.6), the only additional factor that was of borderline significance was infant birth weight (p = 0.087). Gynecologic age was tested in all models for effects. Preliminary univariable analyses s h o w e d a significant association with gestation (r = 0.33, p = 0.045) but the effect was not significant after adjustment.
Discussion The present s t u d y examined the relationship between maternal age, pregnancy outcome, and maternal factors using the data from registrants at an
inner-city prenatal facility. Care was taken to provide controls for several characteristics that m a y potentially confound outcomes a n d lead to spurious differences.
Pregnancy Outcome We found no significant differences in birth weight or apgar scores between young adolescents and older w o m e n w h o received prenatal care. These results agree with Rothenberg's s t u d y (9)of older adolescents showing that outcomes were no worse for young mothers in comparison to older mothers w h e n the social and behavioral factors affecting pregnancy were controlled. We found shortened gestation in our y o u n g adolescent women. The consistency of this observation across studies where outcomes have been determined by different m e t h o d s (5,6,15,16) makes bias seem unlikely, but does not exclude the possibility. Studies reporting an increased incidence of infants below 2500 g in y o u n g adolescents m a y be reporting an increased incidence of shortened gestation, a longrecognized cause of low birth weight (17). All of the infants born to our y o u n g adolescents had birth weights appropriate for gestation. The two infants weighing less t h a n 2500 g were born to adolescent mothers at -< 36 weeks gestation. Multipleregression analysis suggested that increased mater-
Table 3. Factors Associated with Gestation, Pregnancy Weight Gain, and Hemoglobin at Registration Dependent variable
Variables in model~
Pregnancy weight gain
TrimesterWIC supplementation Maternal height
Gestation
Maternal height Prepregnant weight
Hemoglobin on admission
Birth weight
Estimated coefficient(/3)
Standard error (/~)
p
-2.43 0.240
0.887 0.091
0.008 0.012
- 0.100 0.118
0.042 0.044
0.02 0.01
0.0001
0.08
0.0004
R2 (model)
32.2 22.9 26.6
aControl variables included: ethnicity; clinic payment status; year of delivery; parity; presence or absence of drinking, smoking at registration; maternal age; weight gain (for gestation model); and gestation (for weight gain model).
170
SCHOLL ET AL.
nal height and reduced maternal prepregnant weight are associated with shortened gestation. These prognostic factors are in agreement with reported maternal characteristics associated with preterm birth (18). Maternal infection has also been implicated in preterm birth for older women, but its effect on young adolescent gestation is unknown. Maternal Characteristics
There were a number of differences in maternal characteristics when young adolescent gravidae were compared to their older controls. Most of these seem to reflect underlying differences in maternal nutritional status, growth, and development. Young mothers menstruate earlier than older gravidae (2,15). That is, in part, an artifact of a truncated (i.e., incomplete) maternal age distribution. However, it may be inferred that many young adolescents are biologically advanced for chronological age, with a larger amount of postmenarcheal growth than later-maturing girls (19,20). As a group, adolescents tend to have irregular eating habits and consume a greater proportion of calories as snacks with a lower density of nutrients than older w o m e n (21); moreover, adolescents have higher nutritional requirements (22). Most dietary studies of adolescent gravidae have reported low calorie intakes and low intakes of micronutrients; many have reported biochemical evidence of poor nutritional status (23,24). Therefore, our finding that poor pregnancy weight gain and lower hemoglobin levels on registration were more common in young adolescent gravidae w h e n compared to the older gravidae with similar social and economic circumstances is not surprising. Regression analyses suggest that pregnancy weight gain is associated with the trimester the gravida enrolled in WIC. Thus, one important element in attaining adequate pregnancy weight gain in an urban poor population may be an early start in a supplementation program. While evidence is scanty, there are some indications that younger adolescents may be more responsive to nutrition supplementation during pregnancy than older women. For instance, Paige has shown a 269 g increase in the birth weight of infants born to young adolescents (< 16 years) who were supplemented with Sustacal (25). The increase obtained in older adolescents (87 g) was comparable to previous reports in adults (26). An alternative interpretation is that the relationship demonstrated in our study is not a causal one. That is, women who seek prenatal care and obtain WIC
JOURNAL OF ADOLESCENT HEALTH CARE Vol. 5, No. 3
early in pregnancy may differ in other unmeasured factors associated with increased pregnancy weight gain in comparison to women who do not seek care until later in their pregnancy. Maternal stature appears to be associated with pregnancy weight gain. Shorter women, because of their reduced body size, may normally gain less weight during pregnancy than taller women (27). In the present study, there is the additional concern that the effect on weight gain may be influenced by adolescent growth during the pregnancy. A recent record-based report suggested that young adolescent mothers are not fully mature but continue to grow from one pregnancy to the next (15). Although differences in our study were not statistically significant, the smaller body size of young gravidae (3.2 kilos less in prepregnant weight and 2.3 cm less in height than controls) is consistent with this and with data reported by Garn and Petzold (27). In summary, our data indicate little difference in pregnancy outcome between young adolescents and mature women who receive prenatal care except for the length of gestation. On the other hand, young adolescent gravidae had lower hemoglobin levels on admission and a greater frequency of poor weight gain than older w o m e n living under similar social and economic conditions. These indicators of poor nutritional status may be common to all young low SES teenagers or may be a condition that is exacerbated by childbearing in early adolescence. We thank the staff of St. John the Baptist Prenatal Clinic, Diocese of Camden for facilitating our work and Sharon Pavelich for assisting with the research.
References 1. Clark J. Adolescent obstetrics: Obstetric and sociologic implications. Clin Obstet Gynecol 1971;14:1026-36. 2. Hassan HM, Falls FH. The young primipara. Am J Obstet Gynecol 1964;88:256-69. 3. Dott AB, Fort AT. Medical and social factors affecting early teenage pregnancy. Am J Obstet Gynecol 1976;125:532-6. 4. Hutchins FL, Kendall N, Rubino J. Experience with teenage pregnancy. Obstet Gynecol 1979;54:1-5. 5. Dwyer JF. Teenage pregnancy. Am J Obstet Gynecol 1976;118:373-6. 6. Battaglia FC, Frazier TM, Hellegers AE. Obstetric and pediatric complications of juvenile pregnancy. Pediatrics 1963;32:902-10. 7. Millman S, Mosher WD. Selected demographic characteristics of teenage wives and mothers. Advance Data (DHEW Publ. No. (PHS) 80-1250) 1980;61. 8. Venture SJ. Teenage childbearing: United States 1966-1975. Monthly Vital Statistics Report (DHEW Publ. No. (HRA) 771120) 1977;25/15. 9. Rothenberg PB, Varga PE. The relationship between age of
July 1984
10. 11.
12. 13.
14.
15.
16.
17. 18. 19.
mother and child health and development. Am J Publ Health 1981;71:810-17. Osbourne GK, Howat RC, Jordan MM. The obstetric outcome of teenage pregnancy. Br J Obstet Gynecol 1981;88:215-21. Department of Health and Human Services. Factors associated with low birth weight: United States, 1976. DHEW Publ. No. (PHS) 80-1915, National Center for Health Statistics, Hyattsville, MD, 1980. Hollingsworth DR, Kessler AK. Teenage pregnancy solutions are evolving. N Engl J Med 1980;303:516-18. Bloom AD. Guidelines for studies of human populations exposed to mutagenic and reproductive hazards. March of Dimes Birth Defects Foundation, 1981. Mantel N, Haenzel W. Statistical aspects of the analysis of data from retrospective studies of disease. J Nail Cancer Inst 1959;22:719-23. Sukanich AC. Measures of physical maturity and outcome of pregnancy in girls less than 16 years of age. Unpublished dissertation. Graduate School of Public Health, University of Pittsburgh, Pittsburgh, Pennsylvania, 1981. Horon IL, Strobino DM, MacDonald H. Birthweights among infants born to adolescent and young adult women. Am J Obstet Gynecol 1983;146:444-9. Lubchenko LO. Assessment of gestational age and development at birth. Ped Clin North Am 1970;17:125-45. Berkowitz GS. An epidemiologic study of preterm delivery. Am J Epidemiol 1981;113:81-6. Roche AF, Davila GH. Later adolescent growth in stature. Pediatrics 1972;50:874-6.
OUTCOME IN ADOLESCENT PREGNANCY
171
20. Faust MS. Somatic development of adolescent girls. Monographs of the Society for Research in Child Development 1977;169/42:1. 21. Nelson M. Dietary practices of adolescents. In: Winick M, ed., Adolescent Nutrition. New York, John Wiley and Sons: 1982;35-44. 22. National Research Council. Recommended Daily Allowances, 1980, (9th Edition). National Academy of Science, Washington, 1980. 23. Cherry F, Bennett E, Bazzano GS, et al. Plasma zinc in hypertension, toxemia, and other reproductive variables in adolescent pregnancy. Amer J Clin Nutr 1981;34:2367-75. 24. Carruth BR. Smoking and pregnancy outcome of adolescents. J Adol Health Care 1981;2:115-20. 25. Paige DM, Cordano A, Mellits ED, et al. Nutritional supplementation of pregnant adolescents. J Adol Health Care 1981;1:261-5. 26. Rush D. Nutritional services during pregnancy and birthweight: A retrospective matched pair analysis. Can Med Assoc J 1981;125:567-75. 27. Gueri M, Jutsum P, Sorhaindo B. Anthropometric assessment of nutritional status in pregnant women: A reference table of weight-for-height by week of pregnancy. Am J Clin Nutr 1982;35:609-16. 28. Garn SM, Petzold AS. Characteristics of the mother and child in teenage pregnancy. Am J Dis Child 1983;137:365-8.