Secondary Logo

Journal Logo

Original Research

Temporal Trends of Preterm Birth Subtypes and Neonatal Outcomes

Barros, Fernando C. PhD1; Vélez, Maria del Pilar MD2

Author Information
doi: 10.1097/01.AOG.0000215984.36989.5e
  • Free

Preterm neonates, even those near term, are at a higher risk of severe morbidity and mortality,1,2 and the observed increase in the prevalence of preterm births in some developed and middle-income countries3–8 is a reason for great concern.

Preterm births do not comprise a single entity, but have different subtypes9 with diverse determinants,10 and preventive measures need to be developed taking into account this diversity. Villar and colleagues11 analyzed the neonatal outcomes of the subtypes of preterm births and found that the risks are quite different for each group. The subtypes are based on the presence or absence of maternal medical or obstetric complications and on the way of initiation of labor or delivery—spontaneous labor, labor after premature rupture of membranes (PROM), or induction of labor or elective cesarean (induction or elective cesarean).

In this article we describe the temporal trends of the different subtypes of preterm births in Latin America, taking advantage of a large database covering births occurring in the region in the last 2 decades. We also analyze the temporal trends in neonatal morbidity and mortality for each subgroup.


Data were collected with the Perinatal Information System, developed in 1983 by the Pan American Health Organization/World Health Organization Latin American Center for Perinatology and Human Development (Simini F. Perinatal information system (SIP): a clinical database in Latin America and the Caribbean [comment]. Lancet 1999;354:75).12 The Perinatal Information System is a clinical data registration system with a computerized program for data entry, statistical analysis, and report processing. From the first antenatal visit until discharge of both mother and newborn, the attending physicians or nurses record data in the perinatal clinical record in check-box format. These data include demographic and biologic information related to maternal characteristics, prenatal care, labor management, maternal complications during pregnancy, delivery, and the puerperium and neonatal characteristics and outcomes. In each maternity hospital after discharge of both mother and neonate, the data are entered with help of the Perinatal Information System software. Periodically, the institutions send their databases to the Center, where further data entry, quality control, and validation are performed.

From 1985 to 2003 the database received information from pregnancies and births from Uruguay (26.0%), Argentina (24.8%), Peru (9.5%), Colombia (8.6%), Honduras (8.2%), Paraguay (6.9%), El Salvador (4.2%), Chile (2.8%), Bolivia (2.3%), Costa Rica (2.2%), Panama (1.4%), Nicaragua (1.2%), Brazil (0.8%), Ecuador (0.6%), Mexico (0.4%), and Venezuela (0.1%). Information from the countries are not meant to be representative, because they are collected and sent to the Center by the maternity hospitals who choose to do so. Data from 51 hospitals are included in the study. The outcomes of the study were preterm births with the different subtypes, small for gestational age, neonatal morbidity, and neonatal mortality before discharge.

Gestational age was calculated using the first day of the last normal menstrual period (LMP) or estimated by obstetric ultrasonography obtained before 20 weeks of gestation when LMP was not reliable or not available. In the absence of both menstrual and ultrasound information, gestational age was estimated by physical and neurologic assessments of the newborn.13 Preterm births were those with a gestational age of less than 37 completed weeks. Newborns were considered to be alive if they presented any sign of life irrespective of their birth weight. Only neonates born alive from single pregnancies and without malformations were included in the study.

The subtypes of preterm births were classified according to the presence of severe obstetric and medical complications during pregnancy (yes or no) and the way of initiation of labor or delivery. Obstetric and medical complication were considered as the presence of at least 1 of the following conditions: chronic hypertension, preeclampsia, eclampsia, hemorrhage during second or third trimester of pregnancy but before labor; any cardiopathy, renal disease, severe anemia; and hemolytic disease in the present pregnancy (rhesus-isoimmunization) considered as Rh sensitization. For valid comparison with the publication of Villar and colleagues,11 we did not include pregestational diabetes mellitus and gestational diabetes in the subgroups, because they are associated with an increase in fetal weight in one third of the cases, contrary to the other pathologies.

The ways of initiation of preterm labor or delivery were spontaneous initiation of labor, labor after premature rupture of amniotic membranes (occurring at least 1 hour before the onset of sustained labor), and induction of labor or elective cesarean delivery, either medical induction due to obstetric complications or due to other indication, such as breech or transverse presentation, previous cesarean, suspected small for gestational age fetus (SGA), or possible errors in the calculation of the gestational age. The variable “suspected SGA,” however, was included in the Perinatal Information System database after 1993, and was not available for the years preceding this date.

Small for gestational age neonates were those below the 10th percentile of birth weight for gestational age and sex of the Williams curves.14 For neonatal morbidity, we considered all neonates registered by the doctor or nurse in charge as presenting any sort of morbidity during the hospital stay. Diseases occurring after discharge were not registered and therefore not studied. Likewise, neonatal mortality was registered only before hospital discharge or referral to another hospital or to an intensive care unit.

Maternal education was considered as the total numbers of schooling completed with success. Maternal age was defined as completed years at time of birth. Information on smoking was available only after 1991; mothers were categorized as smokers or nonsmokers. For marital status, mothers were categorized as having a stable union or not.

Univariable analyses using χ2 test for linear trend were performed comparing the prevalences of preterm subgroups and other outcomes with the different periods. Multivariable analysis using logistic regression was used to adjust for possible confounders. The statistical software package SPSS 12.0 (SPSS Inc, Chicago, IL) was used for all data analyses. The study was ethically approved by the Pan American Health Organization.


The database included 291,496 singleton liveborns without malformations in the period 1985–1990, 622,050 during 1991–1995, and 805,745 from 1996 to 2003, totaling 1,719,291 neonates. Eighty-four percent of this increase was due to an increase in the number of cases sent to the Center by the hospitals that participated from the beginning, and 15.8% corresponded to new hospitals entering after 1985–1990.

In the 19 years covered by the study (1985–2003) there were 149,284 preterm births (9.3%) among the 1,605,275 singleton births without congenital malformations and with known gestational age. Information on gestational age could not be obtained in 6.7% of the cases.

Table 1 shows the trends in the distribution of the preterm subtypes in the study period. Spontaneous labor without maternal complication was by far the most prevalent group, representing almost 60% of the cases of preterm births and presented a slight relative decrease in the period. The second most prevalent subtype was PROM without maternal complications, representing 15% of the cases. It is interesting to note that the cases of preterm birth after induction or elective cesarean have increased in the last 2 decades, from around 10% in the years 1985–1990 to 18.5% in recent years. The increase was observed both for neonates with less than 32 weeks of gestation and for those between 32 and 36 weeks (not shown).

Table 1
Table 1:
Prevalence of Preterm Births by Subgroups and Year of Birth From the Perinatal Information System Database of Singleton Liveborn Births Without Malformations

Mothers of preterm neonates presented levels of education significantly lower than those of term neonates—mean (± standard deviation) of 7.0 years of schooling (± 3.7) compared with 7.4 years (± 3.6), respectively (P < .001). Those mothers with missing gestational age information were still less educated—5.7 years (± 3.8). Among the preterm subgroups, the mothers with the lowest level of education were those who had spontaneous labor without medical complications (6.8 years of schooling, ± 3.5), whereas the group with the highest level of education was that of induction or elective cesarean with maternal complications—7.6 years of schooling (± 3.8). Smoking was more prevalent among the preterm subgroups of spontaneous labor without maternal complications (13.0%), PROM with maternal complications (14.1%), and PROM without maternal complications (17.0%), in comparison with 11.2% for mothers of term neonates. In addition, multiparity (4 or more previous live births) was more frequent among mothers of preterm neonates of all subgroups than for mothers of term neonates.

For the different preterm subgroups, mean birth weight was lower for the PROM subgroup—2,231 g (± 709), followed by the group of induction or cesarean, with a mean of 2,311 g (± 761) and then by the spontaneous labor group, with 2,541 g (± 741), compared with 3,268 g (± 573) for the term neonates. The same trends were observed for gestational age—PROM 33.1 weeks (± 3.2), induction or cesarean 33.7 weeks (± 2.8), and spontaneous labor 33.9 weeks (± 2.9), compared with 39.2 weeks (± 1.2) for term newborns. All differences were highly significant (P < .001). Within the subgroups, mean birth weight and gestational age were always lower when the mothers presented medical complications. This was especially notable for the subgroups of induction or cesarean, where the neonates whose mothers presented medical complications weighed almost 300 g less than those without medical complications: 2,143 g (± 724) and 2,414 g (± 766), respectively (P < .001). Also, the neonates belonging to the spontaneous labor without maternal complications group were bigger or had longer gestation than those from all other subgroups.

Regarding being small for gestational age, for the whole population—term and preterm neonates—there was a decrease of 28% in SGA in the last years (1996–2003) in comparison with the previous periods The reduction of SGA was observed in all preterm subgroups (Table 2).

Table 2
Table 2:
Prevalence of Small for Gestational Age Among Term and Preterm Subgroups From the Perinatal Information System Database of Singleton Liveborn Births Without Malformation

In relation to neonatal morbidity, there was a decrease in the overall prevalence in recent years, from 13.6% to 9.0%, due to a 44% decrease in morbidity among term neonates (Table 3). For the preterm neonates, however, there was an increase in morbidity, especially in the subgroups with maternal complications. The induction or elective cesarean group for other causes was the only group in which there was no increase in morbidity. The subgroup of spontaneous labor without maternal complication presented the lowest prevalence of morbidity.

Table 3
Table 3:
Prevalence of Neonatal Morbidity From the Perinatal Information System Database of Singleton Liveborn Births Without Malformation

Neonatal mortality before hospital discharge decreased almost 40% in the study period, and this was much more marked for term neonates (58%) than for the preterms (28%). Because of this, the participation of preterm neonates in the overall mortality increased from 70% to 80%. Table 4 shows that neonates born after preterm spontaneous labor without maternal complication were those with the lowest mortality rates, and that for each subgroup − spontaneous labor, PROM and induction/elective cesarean − the neonatal mortality was much higher among neonates whose mothers presented complications.

Table 4
Table 4:
Neonatal Mortality Before Hospital Discharge per 1,000 Among Term and Preterm Subgroups From the Perinatal Information System Database of Singleton Liveborn Births Without Malformations

A logistic regression analysis was performed to adjust for possible confounding variables, using neonatal mortality before discharge as the outcome. The subgroups of preterm neonates were compared, with the group of spontaneous preterm labor without maternal complications being used as the reference. This group was chosen because it is by far the most frequent and has the lowest neonatal mortality rate. Table 5 shows the nonadjusted odds ratios and the odds ratios adjusted for maternal education, age, smoking during pregnancy, marital status, parity, and interpregnancy interval. After adjustment all odds ratios remained significantly different from the comparison group, increasing in the groups of PROM and induction or elective cesarean with maternal complications. Among the preterm neonates born after PROM without maternal complications the odds ratio remained unaltered.

Table 5
Table 5:
Nonadjusted and Adjusted Odds Ratios for Neonatal Mortality Through Logistic Regression From the Perinatal Information System Database of Singleton Liveborn Births Without Malformations


The trends over time in the prevalence of subtypes of preterm births in this study are noteworthy. Although spontaneous preterm labor without maternal complications was by far the most important type of preterm birth, representing almost 60% of the cases, the relative growth of preterm births due to induction or elective cesarean, either with or without medical complications, was very marked, this subgroup representing 18.5% of the preterm birth in recent years. When we selected only the hospitals starting from the beginning of the study and reanalyzed the prevalences of the preterm subgroups by year, the results were essentially the same as for the whole sample (data not shown). We therefore concluded that the changes over time could not be attributed to an inclusion bias.

It should be considered, however, that the study has limitations, and that the main one is that the database is not population-based, but was created with information provided by a large number of maternity hospitals of many Latin America countries in the last 19 years. Therefore, the findings could not be generalized for other more developed countries and should be addressed with caution even for Latin America, because only 2 countries—Argentina and Uruguay—were responsible for one half the cases. Also, some important variables, such as family income, housing, work status, were not available and therefore could not be addressed in the analysis. In addition, the diagnosis of neonatal mortality was restricted to the deaths occurring during the hospital stay soon after birth, and if the newborn died after being transferred to an intensive care unit or after being sent home, the information was not collected by the system.

Even with these drawbacks, this is an important source of information on maternal and newborn health in Latin American hospitals in the last 2 decades and has been used previously in the evaluation of other aspects of reproductive health.15–20

It is also important to mention that in the subgroup of induction or cesarean without medical complications, the 62% relative increase during the study period, from 6.8% to 11.0%, was not due to an increase in cases of previous cesareans or breech or transverse presentation. In relation to suspected SGA, this information was not available in the first years of the study, but there was an increase in the prevalence of clinical diagnosis of SGA from 8.2% in 1994–1995 to 15.2% for the period 1996–2003. However, a large part of the causes for the increase in the subgroup of induction or elective cesareans without complications could not be elicited with the available information. The increasing importance of this subgroup indicates that they should be included in future analyses of preterm subgroups and should not be described separately.11

The high prevalence of SGA that we found among the preterm neonates born after induction or elective cesareans with maternal complications was also described in the Latin American population studied by Villar and colleagues11 and in a large European study.21

Examining the sociodemographic characteristics of the preterm subgroups, it is observed that the mothers of the most prevalent subgroup, spontaneous labor without complications, presented poverty-related characteristics—less education, younger maternal age, living without a partner, and multiparity. This makes the prevention of preterm births in this category more difficult, because some of these risk factors are not easily modified. Fortunately, in our study population there were clear improvements in maternal education in the 2 decades covered by the study. The prevalence of adolescent pregnancies, marital status, and multiparity did not change, however.

On the other hand, the observed higher prevalence of smoking in the PROM preterm subgroup is in accordance with previous information that tobacco smoking independently increases the risk of PROM.22 A positive aspect is that the prevalence of smoking decreased in recent years in our population, from 22% between 1991 and 1995 to 12.4% between 1996 and 2003 (data not shown in the tables).

Although the preterm neonates born after spontaneous labor without medical complications have the most favorable birth weights and gestational ages and the lowest morbidity and mortality rates among the preterm subgroups, their high prevalence makes them responsible for one half the cases of morbidity and deaths of the preterm neonates. On the other hand, neonates of the subgroup of PROM without maternal complications, although presenting mortality rates twice as high as those of the subgroup of spontaneous labor without medical complications, even after adjusting for confounding variables, were the second most important subgroup, with nearly 20% of the overall morbidity and mortality.

One important finding of this study is the growing importance of the subgroups of induction or elective cesarean with and without maternal complications in preterm morbidity and mortality. These subgroups were responsible for 13.4% of the preterm deaths in the years 1985–1990, and this proportion rose to 21.2% in 1996–2003. Because cesarean and preterm births are increasing markedly in the region,4,8 their role in preterm morbidity and mortality is expected to increase even more.

This study identified trends in the prevalence of the different subtypes of preterm births in Latin America, especially related to early interruption of pregnancy. The importance of spontaneous preterm labor and of PROM, both without maternal complications, as the main determinants of preterm births was also emphasized. Due to the dynamic nature of the different subtypes of preterm births, it is necessary to develop a permanent surveillance system to monitor their trends, and at the same time efforts should be concentrated on identifying evidence-based preventive measures for each subtype.


1. Save the children. State of the world's newborns report. Washington (DC): Save the Children; 2001.
2. Kramer MS, Demissie K, Yang H, Platt RW, Sauve R, Liston R. The contribution of mild and moderate preterm birth to infant mortality. Fetal and Infant Health Study Group of the Canadian Perinatal Surveillance System. JAMA 2000;284:843–9.
3. Ananth CV, Joseph KS, Oyelese Y, Demissie K, Vintzileos AM. Trends in preterm birth and perinatal mortality among singletons: United States, 1989 through 2000. Obstet Gynecol 2005;105:1084–91.
4. Barros FC, Victora CG, Barros AJ, Santos IS, Albernaz E, Matijasevich A, et al. The challenge of reducing neonatal mortality in middle-income countries: findings from three Brazilian birth cohorts in 1982, 1993, and 2004. Lancet 2005;365:847–54.
5. Arias E, MacDorman MF, Strobino DM, Guyer B. Annual summary of vital statistics—2002. Pediatrics 2003;112:1215–30.
6. Demissie K, Rhoads GG, Ananth CV, Alexander GR, Kramer MS, Kogan MD, et al. Trends in preterm birth and neonatal mortality among blacks and whites in the United States from 1989 to 1997. Am J Epidemiol 2001;154:307–15.
7. Joseph KS, Kramer MS, Marcoux S, Ohlsson A, Wen SW, Allen A, et al. Determinants of preterm birth rates in Canada from 1981 through 1983 and from 1992 through 1994. N Engl J Med 1998;339:1434–9.
8. Bettiol H, Rona RJ, Chinn S, Goldani M, Barbieri MA. Factors associated with preterm births in southeast Brazil: a comparison of two birth cohorts born 15 years apart. Paediatr Perinat Epidemiol 2000;14:30–8.
9. Savitz DA, Blackmore CA, Thorp JM. Epidemiologic characteristics of preterm delivery: etiologic heterogeneity. Am J Obstet Gynecol 1991;164:467–71.
10. Berkowitz GS, Blackmore-Prince C, Lapinski RH, Savitz DA. Risk factors for preterm birth subtypes. Epidemiology 1998;9:279–85.
11. Villar J, Abalos E, Carroli G, Giordano D, Wojdyla D, Piaggio G, et al. Heterogeneity of perinatal outcomes in the preterm delivery syndrome. Obstet Gynecol 2004;104:78–87.
12. Diaz-Rossello JL. Health services research, outcomes, and perinatal information systems. Curr Opin Pediatr 1998;10:117–22.
13. Capurro H, Konichezky S, Fonseca D, Caldeyro-Barcia R. A simplified method for diagnosis of gestational age in the newborn infant. J Pediatr 1978;93:120–2.
14. Williams RL, Creasy RK, Cunningham GC, Hawes WE, Norris FD, Tashiro M. Fetal growth and perinatal viability in California. Obstet Gynecol 1982;59:624–32.
15. Conde-Agudelo A, Belizan JM, Lammers C. Maternal-perinatal morbidity and mortality associated with adolescent pregnancy in Latin America: cross-sectional study. Am J Obstet Gynecol 2005;192:342–9.
16. Conde-Agudelo A, Belizan JM, Breman R, Brockman SC, Rosas-Bermudez A. Effect of the interpregnancy interval after an abortion on maternal and perinatal health in Latin America. Int J Gynaecol Obstet 2005;89:S34–40.
17. Conde-Agudelo A, Belizan JM, Norton MH, Rosas-Bermudez A. Effect of the interpregnancy interval on perinatal outcomes in Latin America. Obstet Gynecol 2005;106:359–66.
18. Conde-Agudelo A, Belizan JM, Diaz-Rossello JL. Epidemiology of fetal death in Latin America. Acta Obstet Gynecol Scand 2000;79:371–8.
19. Conde-Agudelo A, Belizan JM, Lindmark G. Maternal morbidity and mortality associated with multiple gestations. Obstet Gynecol 2000;95:899–904.
20. Conde-Agudelo A, Belizan JM. Maternal morbidity and mortality associated with interpregnancy interval: cross sectional study. BMJ 2000;321:1255–9.
21. Zeitlin J, Ancel PY, Saurel-Cubizolles Papiernik E. The relationship between intrauterine growth restriction and preterm delivery: an empirical approach using data from a European case–control study. BJOG 2000;107:750–8.
22. Parry S, Strauss JF 3rd. Premature rupture of the fetal membranes. N Engl J Med 1998;338:663–70.
© 2006 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.