Materials and Methods
A newly created unique database was used that linked maternal and neonatal/infant hospital discharge records to birth and death certificate records. The linkage of vital statistics data was established for all civilian hospitals that report to the California Office of Statewide Health Planning and Development during 1992–1993. This database did not include deliveries in military facilities, home deliveries, out-of-state deliveries, and deliveries at birthing centers not reporting to the California Office of Statewide Health Planning and Development. The linkage method successfully linked 98.9% of maternal and 98.6% of neonatal/infant hospital discharge records with birth and death records, an overall linkage of 97.9%. This generated a database of more than 1.16 million deliveries. Using SAS software (SAS Institute, Cary, NC) we searched the database, utilizing codes from International Classification of Diseases, Ninth Revision14 (ICD-9). This resulted in a specific data set for statistical analysis.
The linked database was searched with respect to multiple demographics and antepartum, intrapartum, and postpartum diagnoses. The database initially was sampled for all women age 40 or older on the day of delivery (study population). These records then were examined for ICD-9 and Current Procedural Terminology codes relating to pregnancy outcomes.15 A control group of patients was obtained by retrieving records of all delivering patients who were 20–29 years of age on the day of delivery. Data for this second group were likewise examined for particular ICD-9 and Current Procedural Terminology codes. Each of these groups then was divided into a nulliparous subgroup and a multiparous subgroup.
We compared race, mode of delivery, mean birth weight, and gestational age between the study and control groups stratified by parity. We ascertained the incidence of several birth outcomes as well as maternal complications. Because racial and payer type differences existed, we calculated adjusted odds ratios (ORs) comparing deliveries of women age 40 or older with deliveries of women age 20–29 stratified by parity. Race was divided into white, Hispanic, black, Asian, and other. Payer types were MediCal, private insurance, health maintenance organization, self, and other. We used the following birth outcomes and pregnancy complications: birth trauma, birth asphyxia, fetal growth restriction (FGR), intraventricular hemorrhage, infant death, neonatal death, malpresentation, fetal disproportion, obstructed labor, abnormal forces of labor, prolonged labor, preeclampsia, chronic hypertension, maternal diabetes, gestational diabetes, prematurity (less than 37 weeks' gestation), and postterm delivery (over 42 weeks' gestation). These outcomes and complications of pregnancy were defined according to ICD-9 and birth certification data.
The entire population of patients whose discharge information was available and who delivered during the period of the study included approximately 1,160,000 women. There were 24,032 patients in the study population age 40 or older (nulliparas, 4777; multiparas, 19,255), representing 2.1% of the total population. The control population (age 20–29) was composed of 642,525 patients (nulliparas, 258,900; multiparas, 383,625), or 55% of the entire population. In Figure 1, the age distribution of the study population of all patients age 40 or older is demonstrated. The majority of patients age 40 or older at the time of delivery are in their early 40s.
In Table 1, the racial distribution is given for both the nulliparous and multiparous subgroups in each age group. The most significant finding was the high percentage (64%) of white women age 40 or older (64%) undergoing a first delivery. In the other groups the racial distribution was more uniform. Cesarean and operative vaginal delivery (forceps and vacuum) rates were significantly higher in the older nulliparous group compared with the control group and the older multiparous group.
Table 2 displays the various pregnancy outcomes as determined from discharge records and birth certificates for both groups and subgroups. There were statistically significant increases in the rates of most complications of pregnancy among the nulliparous women age 40 or older compared with the other groups. These increases were seen in both underlying maternal medical complications and complications of labor.
In Table 3, the newborn complications are presented for both divisions of each group, with adjusted ORs, adjusting for race and payer source. The rate of birth trauma was decreased in the older nulliparous group of patients compared with the control patients, most likely because of the increased cesarean delivery rate in the older group. In addition, infant and neonatal death rates were not increased among the older nulliparous patients but were increased among the multiparous patients compared with the control group.
This study shows that first-time mothers who are giving birth at age 40 or older are at high risk for some form of operative delivery (our findings: cesarean delivery, 47.0%; operative vaginal delivery, 14.2%), compared with younger nulliparous women (cesarean delivery, 22.5%; operative vaginal delivery, 12.9%). The multiparous women in our study who were age 40 or older had significant increases in rates of operative deliveries (cesarean delivery, 29.6%; operative vaginal delivery, 6.3%) compared with younger controls (17.8 and 4.6%, respectively). The significantly increased operative delivery risk may be explained largely by the antepartum or intrapartum complications such as malpresentation and abnormal forces of labor, but other nonmeasured factors must be taken into account. Older nulliparous women frequently have a long history of infertility and become pregnant with the aid of assisted reproductive technology. Couples' demands for perfect pregnancy outcomes may cause providers to deliver the infants early because of anxiety concerning stillbirth. The mean gestational age at delivery for the older study population was lower than for the control population and may reflect this desire for optimal outcome. The degree to which physician anxiety and patient anxiety add to the increased operative delivery rate is unknown. As the number of women having their first child at age 40 or over continues to grow, it is to be hoped that anxiety will decrease and the cesarean delivery rate will decrease as well. The data obtained in our study may assist health care providers in counseling patients concerning their expectations of pregnancy outcomes.
The cesarean delivery rate of 47% for the study's nulliparous women age 40 or older is twice the rate for the control group and 50% more than the rate for the multiparous women age 40 or over. In the literature there are only limited data concerning cesarean delivery rates for nulliparous women age 40 or older.7,12,13,16 Bianco et al7 reported a 39% cesarean delivery rate for nulliparous women age 40 or older in a population of 607 patients, and these authors concluded that although there was an increase in antepartum and intrapartum complications, the neonatal outcome overall did not appear to be affected by maternal age. More data exist concerning cesarean delivery rates among nulliparous women age 35 or older.9,10,17 Edge and Laros9 reported a 40% cesarean delivery rate for nulliparous patients age 35 or over, a rate that could be explained only partially by presence of gestational complications. The data were collected over 15 years, and changes in physician practice patterns could not be taken into account easily, especially given that the cesarean delivery rate has increased from approximately 5% to 24% during that period.9 Prysak et al10 reported a slightly higher cesarean delivery rate of 44% in 890 nulliparous women age 35 or older and likewise reported an increase in antepartum, intrapartum, and newborn complications, but overall the perinatal outcomes were good. Our study finding of one of the highest cesarean delivery rates reported may be more accurate for current conditions. Furthermore, because our population comprises the vast majority of patients who delivered in California over a 2-year period, a true population cesarean delivery rate can be determined. Most earlier studies report individual hospital cesarean delivery rates.
Virtually every complication of pregnancy was increased in older nulliparous patients compared with both older multiparous and control patients (Table 2). The most striking differences were in the areas of malpresentation and dystocia. The diagnosis of dystocia is physician derived, and often there is inadequate documentation of an appropriate trial of labor. The effects of patient and physician anxiety on what is considered an appropriate trial of labor cannot be measured in our, or any, database. Malpresentation, on the other hand, is determined more easily and thus may represent a true and measurable cause for an increase in the cesarean delivery rate in older nulliparous patients. The rate of preeclampsia was increased in older nulliparous patients by 60% compared with control nulliparous patients and double the rate for older multiparous patients (Table 2). Another pregnancy-related disease, gestational diabetes, was increased dramatically (four-fold) in both older nulliparous and multiparous patients compared with controls, which suggests that maternal age, not parity, is involved in the increase in this disease.
Certain underlying maternal diseases were increased in the older women, compared with the controls. Chronic hypertension was increased five-fold in older nulliparous patients and nine-fold in older multiparous patients, compared with control nulliparous and multiparous women, respectively. The effect that hypertension may have had on pregnancy outcome is unknown but hypertension may have led to the increase in preeclampsia. Placenta previa was increased more than eight-fold in the older nulliparous group, compared with control nulliparous patients. Multiparity has long been known to increase risk for placenta previa, and our data support this (Table 2). The large increase in the rate of placenta previa that was observed in our study's nulliparous patients age 40 or over is unprecedented and currently without explanation. The rate of pregestational diabetes was increased in the older nulliparous group as well as in the older multiparous group, compared with controls. Similar increases in pregnancy complications in the older nulliparous patient have been noted previously. Cnattingius et al8 examined records from a large population of Swedish mothers and found that multiple indicators of poor pregnancy outcome were increased as the age of the mother increased from 20–24 to 30–34 years and compared with mothers over age 40. Their outcomes included small for gestational age, low birth weight, preterm delivery, and fetal death. These investigators8 found that uncomplicated first pregnancies in women over age 40 still were associated with poor pregnancy outcomes, even when underlying medical conditions were taken into account.
Neonatal complications were largely increased in the older patients in our study, with a 50% increase in birth asphyxia, a 40–80% increase in FGR, and a 70–100% increase in intraventricular hemorrhage, compared with control patients (Table 3). These results are consistent with those reported in the older literature, which cited increases in perinatal morbidity and mortality.1–5 In spite of these increases in complications, the vast majority of neonatal outcomes were good, with no increase in either neonatal or infant death among the older nulliparous patients and a slight increase in these rates among the older multiparous patients, compared with controls.
The racial distribution of our study population was different than that of the control population. The total number of white women having children in California during these 2 years was roughly equal to the total number of Hispanic women having children (Table 1). In the study population, however, the percentage of nulliparous white women was increased dramatically (62%), which suggests that this group is more likely to delay childbearing than are all other racial groups. The racial distribution of the older multiparous patients was much closer to that of the control population, which suggests that there was a more usual racial distribution compared with the older nulliparous population. All analyses of pregnancy outcomes were risk adjusted for race and payer type to remove the effect that these two factors have on outcome.
Mean gestational age for the study population at delivery (39.1 weeks) was statistically significantly lower than that for the control population (39.8 weeks), by 4–5 days. Gestational age was lower even when parity was taken into account (Table 1). The causes are unknown but may relate to underlying maternal or fetal problems such as diabetes, chronic hypertension, and fetal distress. It is not known whether this earlier gestational age of delivery is clinically significant. The differences in birth weight between groups are similar (Table 1), with multiparous patients having larger infants, compared with nulliparous patients, a finding that is consistent with findings reported in the literature. The mean birth weights of infants born to the multiparous patients were almost exactly the same for the two age groups, and this would suggest that maternal age may not affect birth weight as much as other factors. The nulliparous patients age 40 or older were associated with the lowest birth weights and the highest cesarean delivery rates.
One of the limitations of this study is that the data were taken from maternal and infant/neonatal hospital discharge summaries that were matched to birth and death certificates. One criticism of this database is that quality of diagnosis reporting can vary. The majority of chart abstracting is performed by medical record personnel within each hospital. Although most of the data, including maternal demographics and primary birth outcomes such as gestational age and birth weight, are reliable, other more obscure outcomes may be missed by abstractors. Major outcomes such as cesarean or operative vaginal delivery are likely to be recorded correctly because of increases in hospital reimbursement with more complicated ICD-9 coding and procedural coding. In addition, the California birth certificate has been shown to be a reliable source of insurance information and other interview questions, compared with direct patient interviews.18 Also, the linked database allows for confirmation of certain major outcomes in both the hospital discharge record and the birth certificate record, and records were found to be matched correctly in more than 97% of cases. Furthermore, the fact that many of the outcome rates in our study were similar to those found in studies involving smaller populations suggests that there was a reasonable degree of reliability.
Additional work is needed in the area of complications of pregnancy in older women and birth outcomes of their offspring so that better management protocols may be developed. As the body of reported data grows, older women will have more information with which to make decisions regarding childbearing.
1. Kane SH. Advanced age and the primigravida. Obstet Gynecol 1967;29:409–14.
2. Hansen JP. Older maternal age and pregnancy outcome: A review of the literature. Obstet Gynecol Surv 1986;41:726–42.
3. Kim DS, Dorchester W, Freeman RC. Advanced maternal age: The mature gravida. Am J Obstet Gynecol 1985;152:7–12.
4. Kessler I, Lancet M, Borenstein R, Steinmetz A. The problem of the older primipara. Obstet Gynecol 1980;56:165–9.
5. Naeye R. Maternal age, obstetric complications, and the outcome of pregnancy. Obstet Gynecol 1983;61:210–6.
6. Hook EB. Rates of chromosomal abnormalities at different maternal ages. Obstet Gynecol 1981;58:282–5.
7. Bianco A, Stone J, Lynch L, Lapinski R, Berkowitz G, Berkowitz RL. Pregnancy outcome at age 40 and older. Obstet Gynecol 1996;87:917–22.
8. Cnattingius S, Forman MR, Berendes HW, Isotalo L. Delayed childbearing and risk of adverse perinatal outcome: A population-based study. JAMA 1992;268:886–90.
9. Edge V, Laros RK. Pregnancy outcome in nulliparous women aged 35 or older. Am J Obstet Gynecol 1993;168:1881–5.
10. Prysak M, Lorenz RP, Kisly A. Pregnancy outcome in nulliparous women 35 years and older. Obstet Gynecol 1995;85:65–70.
11. Ventura SJ. Trends and variations in first births to older women, United States, 1970–86. Vital Health Stat 1989;21:1–27.
12. Spellacy WN, Miller SJ, Winegar A. Pregnancy after 40 years of age. Obstet Gynecol 1986;68:450–4.
13. Lehmann DK, Chism J. Pregnancy outcome in medically complicated and uncomplicated patients aged 40 years or older. Am J Obstet Gynecol 1987;157:738–42.
14. World Health Organization. International Classification of Diseases. 9th rev. Geneva: World Health Organization, 1997.
15. Physician's Current Procedural Terminology CPT 1998. 2nd ed. Chicago: American Medical Association, 1998.
16. Ekblad U, Vilpa T. Pregnancy in women over forty. Ann Chir Gynaecol Suppl 1994;208:68–71.
17. Gordon D, Milberg J, Daling J, Hickok D. Advanced maternal age as a risk factor for cesarean delivery. Obstet Gynecol 1991;77:493–7.
18. Braveman P, Pearl M, Egerter S, Marchi K, Williams R. Validity of insurance information on California birth certificates. Am J Public Health 1998;88:813–6.