SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Factors Affecting Total Counts of Preterm Births
  5. Risk Factors for Preterm Birth
  6. Possible Classifications for Preterm Birth
  7. A Public Health Perspective on Preterm Birth
  8. References

Preterm birth is the major clinical problem associated with perinatal mortality, serious neonatal morbidity and moderate to severe childhood disability in prosperous countries. Its prevalence is affected by the way in which gestational age is assessed, by national differences in the registration of births, associated practices, such as burial costs, or maternity benefits, which encourage or discourage registration, and by the perceived viability of extremely preterm infants. Despite these uncertainties, there is reliable evidence that preterm births are increasing, especially births before 28 weeks gestation. Contributing factors include births following assisted reproductive therapy and ovulation induction, especially multiple births, and the increasing proportion of births among women >34 years. On the other hand, improvements in neonatal care have substantially increased the survival of preterm infants during the last 15 years. There is wider acceptance of the importance of infection as a factor in preterm birth, and increasing recognition that processes leading to preterm birth may be initiated in very early pregnancy (the initiation of pre-eclampsia, major birth defects, premature placental separation), or even prior to pregnancy (prior pregnancy losses). It is unclear whether the familiar clinical presentations of preterm labour and birth reflect different pathophysiological processes. The pathways which link those processes to the consistent pattern of social differences in the probability of preterm birth have prompted new research approaches but in 2002 ‘the stubborn challenge of preterm birth’ remains just that.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Factors Affecting Total Counts of Preterm Births
  5. Risk Factors for Preterm Birth
  6. Possible Classifications for Preterm Birth
  7. A Public Health Perspective on Preterm Birth
  8. References

Preterm birth is the major clinical problem associated with perinatal mortality, serious neonatal morbidity and moderate to severe childhood disability in wealthy countries. This is often summarised by contrasting the pattern of preterm births, which comprise 6–10% of all births in Western countries, with that of preterm deaths, which comprise more than two-thirds of all perinatal deaths. This seems unequivocal enough, especially when seen in the light of the World Health Organisation's (WHO) clear definition of a preterm birth as birth before 37 completed weeks of gestation, or fewer than 259 days since the first day of the woman's last menstrual period.

Factors Affecting Total Counts of Preterm Births

  1. Top of page
  2. Abstract
  3. Introduction
  4. Factors Affecting Total Counts of Preterm Births
  5. Risk Factors for Preterm Birth
  6. Possible Classifications for Preterm Birth
  7. A Public Health Perspective on Preterm Birth
  8. References

The one problematic component in the WHO definition is the boundary between a spontaneous abortion and a birth. The point in pregnancy when the delivery of a fetus is counted as a birth rather than as some other pregnancy outcome differs across countries both by the formal rules, which govern the registration of births and by less formal processes. Factors, which are recognised to have an impact on the proportion of preterm births in a defined population of births are:

  • national or regional criteria for the registration of a fetal death (ranging from 16 weeks to 28 weeks)
  • under-registration of stillbirths and live births close to the registration boundary
  • under-registration of live births when a stillbirth at the same gestation would not require registration
  • the perception that some extremely small—or extremely preterm—live born infants are non-viable
  • the exclusion of legal terminations of pregnancy from birth registration even when the duration of the pregnancy would require registration had a birth occurred.

Other cultural and social factors, which influence the completeness of registration include:

  • the provision of maternity benefits to a mother after a birth but not after a spontaneous abortion, which may increase the likelihood of complete registration
  • the requirement for burial or funeral costs after a birth, which may have the opposite effect
  • differential hospital charges following a ‘miscarriage’ and a ‘birth’, which may be a disincentive for complete registration of extremely preterm births.

Emerging factors which will reduce the proportion of births, which are preterm include early termination of pregnancy following prenatal diagnosis of a birth defect—though this varies with the specific birth defect, and fetal reduction of twin pregnancies to a single pregnancy prior to the registration boundary. There is pressure in some countries to exclude births from birth registration if the death of the fetus is believed to have occurred prior to the registration boundary even though the pregnancy continued past that point.

Impact of Registration Rules and Informal Practices on Preterm Births and Deaths

These factors make a small, but not negligible, difference to the proportion of preterm births. If all births from 20 to 23 weeks were missing from birth registrations, it would lead to a relative reduction in preterm births of 4%. There is a much bigger impact on preterm deaths where excluding those same births results in a relative reduction of 30%1. Similarly, the impact of excluding stillbirths before 28 weeks gestation is a relative reduction in preterm births of 4.7%, while the impact of excluding those stillbirths from preterm deaths is a relative reduction of 36%1. Two recent analyses of individual-level data from countries collaborating in European and international comparison projects confirm the major contribution of these differences to international differences in perinatal mortality2,3.

Impact of Ultrasound Assessment of Gestational Age

Along the whole spectrum of preterm births, not just at the registration boundary, one influential factor is the shift from ‘rounding-off’ gestational ages to calculating the number of completed weeks as in the WHO definition. This was shown a long time ago to change the relative proportions of preterm and growth-restricted infants, increasing the former and decreasing the latter4. The other influential factor is the shift from using the date of the last menstrual period (LMP) to measure gestational age to using early ultrasound assessment of fetal size instead, which also increases the proportion of infants defined as preterm. The reasons for this are that small downward reclassifications exceed upward reclassifications of similar magnitude, consistent with reports that ovulation delayed beyond the 14th day of the menstrual cycle is more frequent than early ovulation5.

Is Preterm Birth Declining or Increasing?

Until the last few years the focus of discussion has been the stability of preterm birth proportions in most countries since population data became available. A few countries (Finland, France), and regions (Aberdeen, Scotland) reported reductions in preterm birth from the 1960s to the 1980s6–8. In Aberdeen and Finland, the changes were attributed to marked improvements in socio-economic factors. In France, they were attributed to a primary prevention programme7. Given similar preterm birth proportions to that of France in other European countries, it has also been suggested that these relatively low levels may be an indirect effect of European antenatal care and social policies9.

Increased Ascertainment

In a number of countries where there have been no changes in birth registration rules in the past decade (e.g. USA, Canada), consistent increases in preterm births have been identified10,11. In Australia, the proportion of births, which are preterm has remained stable12 but state data from Victoria for first births, from 1983 to 1997, show a significant increase in singleton births occurring at 20–27 weeks. One likely reason in all three countries is more complete ascertainment of births close to the registration boundary, probably prompted at least in part, by the improving survival of extremely preterm infants and the concomitant increase in resuscitation, ventilation, and admission to intensive care of infants once perceived to be non-viable. In Australia, all births with a gestation of 20 weeks, and/or those weighing at least 400 g if the gestation is not known, require registration. Despite these registration criteria having been defined in the 1970s, there has been a notable increase in the registration of infants weighing less than 500 g in Victoria from 193/65,000 total births in 1991 to 241/62,000 in 199913.

Increase in Multiple Births

The other major reason in developed countries is an increase in multiple pregnancies, for which the predominant risk factor is fertility treatments, both assisted conception and ovulation induction14,15. In Australia, 1.7% of all births followed assisted conception in 1999–2000. Multiple pregnancies occurred in 20.9% of the viable pregnancies, which resulted from assisted conception15. In the USA, the ratio of triplet and higher-order multiple births had quadrupled between 1980 and 1997, with a 10-fold increase among women aged 35 to 3914. The contribution of assisted conception to triplet births in the USA was calculated to be 43.3% in 1997, with 20% estimated to be spontaneous, and the remainder associated with ovulation induction without assisted conception.

As half of all twin pregnancies, which reach 20 weeks gestation end preterm, with 10% being born before 28 weeks gestation and another 10% between 28 and 31 weeks, and virtually all triplet pregnancies end preterm16, there is a disproportionate effect of assisted conception on preterm births. Singleton births following assisted conception are also more likely to be preterm than are other singleton births15. There is inconclusive evidence17,18 that periconceptional folate supplementation might also increase twinning.

Increases in the use of indicated elective delivery for major maternal complications or poor fetal growth are also contributing to increases in preterm birth, at least in tertiary referral hospitals19. This change will also be redistributing preterm births from mildly preterm to moderate or extremely preterm categories.

Risk Factors for Preterm Birth

  1. Top of page
  2. Abstract
  3. Introduction
  4. Factors Affecting Total Counts of Preterm Births
  5. Risk Factors for Preterm Birth
  6. Possible Classifications for Preterm Birth
  7. A Public Health Perspective on Preterm Birth
  8. References

One consistent major risk factor is a prior preterm birth. Hospital-based studies in the 1970s suggested this risk was high, with a quarter to a third of women having a subsequent preterm birth. Population-based interventions identified much lower recurrence risks, closer to 15%. A probable explanation for this discrepancy comes from Norwegian linked data on first and second singleton births: the risk of recurrence depends on the stage of pregnancy when the first preterm birth occurred, being higher following first births at 28–35 weeks (relative risk (RR) = 5.0), than following first births at 36–38 weeks (RR = 2.1)20. Since Norway registers births from 16 weeks gestation the data show even more striking relative risks of recurrence when the first birth occurred at 16–27 weeks (RR = 20.5). Hospital studies will include more of the very early preterm births.

Another is the presence of a birth defect, even one compatible with normal survival. Other risk factors, such as young maternal age, can be interpreted as biological, social or both. Prior pregnancy losses are usually perceived as biological risk factors when they are spontaneous and as social factors when they are induced. There is increasing evidence that their impact on the risk of preterm birth is virtually identical21.

The association of relative social disadvantage with preterm birth is one of the most consistent findings in the preterm birth literature1. One component of this is maternal smoking. The reduction in overall proportions of women who smoke in many developed countries has resulted in increasing social inequalities with respect to smoking. It is now strongly associated with lower levels of education, lower family income, and not living with a partner. Racial differences in preterm birth are a major focus of analysis22, but it is increasingly recognised that there may be differences on average in maternal body size, customs, behaviours, access to services, age distribution, exposure to racism and discrimination and neighbourhood level factors, which make a major contribution to the preterm birth differences attributed to ‘race’ or ‘ethnicity’. Examples of the complexity are that immigrant and refugee women from North Africa have good pregnancy outcomes in a number of other countries23, therefore it is difficult to explain as just the ‘healthy migrant’ factor. In other countries, immigrants may begin with higher preterm birth proportions, which decline with increasing length of residence24, or begin with low preterm birth proportions, which increase with acculturation25.

As preterm birth in twin births shows only small associations with social and economic differences26, an increase in the proportion of multiple births among preterm births will reduce the contribution of social factors to preterm birth. This reduction will be even greater in countries where access to assisted conception is restricted to those able to make a large financial contribution to the costs of providing the service.

Factors, which might be contributing to changes in preterm birth include:

  • reductions in social inequality, [DOWNWARDS ARROW] preterm birth
  • a fall in the proportion of births to young women, [DOWNWARDS ARROW] preterm birth
  • an increase in the proportion of births to older women, [UPWARDS ARROW] preterm birth
  • a decrease in the proportion of women who are married, [UPWARDS ARROW] preterm birth a reduction in the association between cohabitation and preterm birth, as fewer women marry before giving birth, [DOWNWARDS ARROW] preterm birth
  • population changes due to immigration and refugee programmes, [DOWNWARDS ARROW] or [UPWARDS ARROW] preterm birth
  • increases in maternal height and weight, [DOWNWARDS ARROW] or [UPWARDS ARROW] preterm birth
  • increases in miscarriages and pregnancy terminations prior to 1st births, [UPWARDS ARROW] preterm birth
  • changes in the proportion of women smoking during pregnancy, [DOWNWARDS ARROW] or [UPWARDS ARROW] preterm birth.

It has always been difficult to conceptualise the causal pathways between social risk factors and preterm birth, but there is renewed interest in these links.

Possible Classifications for Preterm Birth

  1. Top of page
  2. Abstract
  3. Introduction
  4. Factors Affecting Total Counts of Preterm Births
  5. Risk Factors for Preterm Birth
  6. Possible Classifications for Preterm Birth
  7. A Public Health Perspective on Preterm Birth
  8. References

I have suggested elsewhere a classification of preterm births by gestational age into three categories, births from 20 to 27 weeks of gestation, from 28 to 31 weeks, and from 32 to 36 weeks1. These categories reflect very marked differences in the probability of survival, in the need for, and costs of, intensive care, and in long term health and disability outcomes. They also allow for group comparisons even when there is some uncertainty about the exact gestational age in completed weeks. The other rationale for this classification is that the relative risks of socio-demographic and reproductive factors differ across these categories.

A common classification is by clinical presentation: spontaneous preterm labour, preterm labour complicated by additional factors, such as antepartum haemorrhage or preterm prelabour rupture of the membranes, and elective preterm delivery, indicated on maternal or fetal grounds27. An alternative version makes a strong distinction between preterm labour and rupture of the membranes. Both of these have been criticised as not necessarily reflecting important differences in the underlying processes28. The latter paper proposes that differences in the availability and accessibility of care might contribute to different clinical presentations and goes on to show that a number of conditions requiring elective delivery (e.g. severe pre-eclampsia) are also characterised by an increase in spontaneous preterm birth.

Classification remains problematic. No classification copes really well with fetal death, which occurs prior to preterm labour without any recognised maternal or fetal problems, or with the much higher early ‘ejection’ from the uterus of the fetus with a birth defect. Multiple pregnancies are associated with a number of complications, which increase the probability of preterm birth but the 10-fold higher probability of preterm birth at 20–27 weeks, and again at 28–31 weeks, in twins compared with singletons is difficult to explain. The contribution of stillbirths, and infants with malformations, to the association between preterm birth and perinatal mortality was described more than 25 years ago27 but is often ignored when discussing preterm birth.

A Public Health Perspective on Preterm Birth

  1. Top of page
  2. Abstract
  3. Introduction
  4. Factors Affecting Total Counts of Preterm Births
  5. Risk Factors for Preterm Birth
  6. Possible Classifications for Preterm Birth
  7. A Public Health Perspective on Preterm Birth
  8. References

When discussing this problem, we need to acknowledge the very great improvements over the past decade in the outcomes of preterm birth for live born infants who do not have a birth defect. This improvement in survival has had a profound effect on perceptions of viability and on decision making about the care of extremely preterm infants. Yet both short term and long term morbidity, and costs of care—in all senses of ‘cost’—to families and society remain major problems. At the same time, we need to remember that most preterm infants (80%) are born at 32–36 weeks, and that their mortality and morbidity are now low, yet because of their sheer numbers this group makes the largest contribution to total perinatal mortality following preterm birth29.

Although we have lacked a conceptual framework for thinking about preterm birth in ways, which link the known socio-economic disparities, and newer social stressors into causal pathways, there are three reasons for cautious optimism, summarised in a recent journal supplement30. One is important developmental work in measuring crucial concepts, such as discrimination. Another is the report of several large prospective studies taking on that challenge of developing and testing specific conceptual frameworks. The third is the combination of improved conceptual frameworks incorporating mechanisms, markers and molecular epidemiology30.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Factors Affecting Total Counts of Preterm Births
  5. Risk Factors for Preterm Birth
  6. Possible Classifications for Preterm Birth
  7. A Public Health Perspective on Preterm Birth
  8. References
  • 1
    Lumley J. The epidemiology of preterm birth. Baillière's Clin Obstet Gynaecol 1993;7: 477498.
  • 2
    Graafmans WC, Richardus JH, Macfarlane A, et al. Comparability of published perinatal mortality rates in Western Europe: the quantitative impact of differences in gestational age and birthweight criteria. Br J Obstet Gynaecol 2001;108: 12371245.
  • 3
    Kramer MS, Platt RW, Yang H, Haglund B, Cnattingnius S, Bergsjo P. Registration artefacts in international comparisons of infant mortality. Paediatr Perinat Epidemiol 2002;16: 1622.
  • 4
    Goldenberg RL, Davis RO, Cutter GR, et al. Prematurity, postdates and growth retardation. The influence of the use of ultrasonography on gestational age. Am J Obstet Gynecol 1989;160: 462470.
  • 5
    Yang H, Kramer MS, Platt RW, et al. How does early ultrasound scan estimation of gestational age lead to higher rates of preterm birth?. Am J Obstet Gynecol 2002;186: 433437.
  • 6
    Olsén P, Läärä E, Rantakallio P, et al. Epidemiology of preterm delivery in two birth cohorts with an interval of 20 years. Am J Epidemiol 1995;142: 11841193.
  • 7
    Papiernik E. Prevention of preterm labour and delivery. Baillière's Clin Obstet Gynaecol 1993;7: 499521.
  • 8
    Hall MH. Incidence and distribution of preterm labour. In: BeardRW, SharpF editors. Preterm Labour and its Consequences. Proceedings of the Thirteenth Study Group of the Royal College of Obstetricians and Gynaecologists. London : 513.
  • 9
    Buekens P. Is there an indirect effect of European prenatal care policies on preterm birth?. Prenat Neonat Med 1998;3: 144145.
  • 10
    Joseph K, Kramer M, Marcoux S, et al. Determinants of preterm birth rates in Canada from 1981–1983 and from 1992–1994. N Engl J Med 1998;339: 14341439.
  • 11
    Demissie K, Rhoads GG, Ananth CV, 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: 307315.
  • 12
    Nassar N, Sullivan EA, Lancaster P, Day P. Australia's mothers and babies 1998. AIHW Cat. No PER 15. Sydney: Australian Institute of Health and Welfare National Perinatal Statistics Unit (Perinatal Statistics Series 10).
  • 13
    Consultative Council on Obstetric and Paediatric Mortality and Morbidity. Annual Reports for the years 1991, 1999. Melbourne: Consultative Council on Obstetric and Paediatric Mortality and Morbidity, 1992 and 2001.
  • 14
    Centers for Disease Control and PreventionUse of assisted reproductive technology—United States 1996 and 1998. MMWR 2002;51: 97101.
  • 15
    Hurst T, Lancaster P. Assisted conception Australia and New Zealand 1999 and 2000. AIHW Cat. No. PER 18. Sydney: Australian Institute of Health and Welfare National Perinatal Statistics Unit (Assisted Conception Series 6), 2001.
  • 16
    Kiely JL. What is the population-based risk of preterm birth among twins and other multiples?. Clin Obstet Gynecol 1998;41: 311.
  • 17
    Ericson A, Källén B, Åberg A. Use of multivitamins and folic acid in early pregnancy and multiple births in Sweden. Twin Research 2001;4: 6366.
  • 18
    Lumley J, Watson L, Watson M, Bower C. Modelling the potential impact of population-wide periconceptional folate/multivitamin supplementation on multiple births. Br J Obstet Gynaecol 2001;108: 937942.
  • 19
    Kramer M, Platt R, Yang H, et al. Secular trends in preterm birth: a hospital based cohort study. JAMA 1998;280: 18491854.
  • 20
    Hoffman HJ, Bakketeig LS. Risk factors associated with the occurence of preterm bitrh. Clinical Obstet Gynecol 1984;27: 539552.
  • 21
    Lumley J. The association between prior spontaneous abortion, prior induced abortion and preterm birth in first singleton births. Prenat Neonat Med 1998;3: 2124.
  • 22
    Branum AM, Schoendorf KC. Changing patterns of low birthweight and preterm birth in the US 1981–98. Paediat Perinat Epidemiol 2002;16: 815.
  • 23
    Guendelman S, Buekens P, Blondel B, Kaminski M, Notzon FC, Masuy-Stroobant G. Birth outcomes of immigrant women in the United States, France and Belgium. Matern Child Health 1999;3: 177187.
  • 24
    Foix-L'Helias L, Blondel B. Changes in risk factors of preterm delivery in France between 1981 and 1995. Paediat Perinat Epidemiol 2000;14: 314323.
  • 25
    Crump C, Lipsky S, Mueller BA. Adverse birth outcomes among Mexican-Americans: are US-born women at greater risk than Mexico-born women?. Ethn Health 1999;4: 2934.
  • 26
    Buekens P, MacGillivray I, Campbell D, et al. Social differences in low birthweight and preterm deliveries in twins. Paediatr Perinat Epidemiol 1991;5: 276285.
  • 27
    Rush RW, Keirse MJNC, Howat P, et al. Contribution of preterm delivery to perinatal mortality. BMJ 1976;2: 965968.
  • 28
    Klebanoff MA, Shiono PH. Top down bottom up and inside out: reflections on preterm birth. Paediat Perinat Epidemiol 1995;9: 125129.
  • 29
    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: 843849.
  • 30
    FioreEL, HogueCJR, MattisonDR, DamusK, JohnstonRB, WilliamsMA editors. New perspectives on the stubborn challenge of preterm birth. Paediatr Perinat Epidemiol 2001: 15(Suppl 2): 1163.