Infant Mortality

Infant  mortality  is  defined  as  the  death  of  an infant prior to its first birthday. This definition clearly excludes death in utero, or stillborn infants. Although there has been a gradual reduction in the worldwide rate of infant mortality over the last 20 years, this remains a significant public health concern, as infant death continues to remain a concern throughout the world. The overall rate of infant mortality is around 36 deaths per 100 live births, which worldwide translates into roughly 5 million deaths per year. The largest declines in rates of infant mortality have been evident in more highly developed countries such as the United States. Interestingly, these rates declined steadily through the 1990s but have recently turned upward again for reasons that will be discussed later. For example, in the United States for the year 2000, the rate of infant mortality reached an all-time low of 6.9 deaths per 1,000 live births. Much of this reduction during the last decade was attributed to an increase in maternal education about the prevention of infant death and illness, the development of new medical strategies for supporting infants born at high risk of infant mortality, and a record number of mothers getting early prenatal care. The rate of infant mortality in the United States remains relatively high compared to that of other industrialized countries. For example, despite the improvements cited above, the nation still ranked 27th in infant mortality among industrialized countries in an analysis of 1997 data. Recent data indicate that Sweden is among the countries with the lowest recorded rates of infant mortality in history, with a rate of 3.6 deaths per 1,000 live births. Importantly, disparities remain among racial and ethnic groups in many measures of maternal and child health, both in the United States and elsewhere. For example, the infant mortality rate among black children in the United States is more than double that for white children. Recent data show that whereas the infant mortality rate for Caucasians and Hispanics averages 3.3%, for African Americans it was over 8%.

Recent Progress

An ongoing emphasis on the importance of health behaviors during pregnancy including stress reduction, avoidance of drugs and alcohol, abstinence from smoking cigarettes, and the importance of maternal nutritional status have also contributed to this positive change. Public service campaigns supporting the placement of infants on their backs to sleep have also helped reduce this rate by reducing the occurrences of sudden infant death syndrome (SIDS). Moreover, although significant ethnic disparities remain in the rates of infant mortality in the United States and elsewhere, in recent years rates of infant mortality in all ethnic groups as well as in other high-risk populations such as adolescent mothers were also lower during the 1990s than in previous recording years. Very recently, however, the rate of infant mortality in the United States increased after a long period of decline. This change has been attributed not to reversal of the abovementioned positive trends, but rather to an increase in the average age of childbearing in the United States. More and more women are choosing to delay pregnancy and childrearing until later in life, a time at which the risk of infant mortality as well as that of pregnancy complications and developmental challenges increases significantly.

An additional, related challenge to that of SIDS is the potential for the practice of sharing a bed with a young infant to increase the likelihood of infant mortality, via SIDS or suffocation. This is a complex issue, and the role of bedsharing in infant mortality remains under debate. Some have suggested that bedsharing might reduce the risk of SIDS as a parent would be present to rouse the infant in the event that the infant stopped breathing, but the American Academy of Pediatrics (AAP) has stated that there are no conclusive scientific studies showing clearly that bedsharing reduces the incidence of SIDS and points out that several studies do strongly suggest that under certain circumstances, bedsharing may, in fact, increase the risk of SIDS. The practice of bedsharing is tied to a variety of infant care issues, including breastfeeding and cultural traditions, and thus, there has been no blanket statement condemning the practice as endangering infants.

Educational campaigns, however, have been effective in educating parents about the potential risks of bedsharing along with guidelines for other safe infant sleeping practices including the “back to sleep” campaign, and other aspects of appropriate choices for infant bedding. One important aspect of this educational campaign has been to describe situations to parents in which bedsharing may be especially problematic, including cases in which the parent is obese and is thus more likely to suffocate the infant or the parent has a drugor alcohol-use problem, and may again be more likely to suffocate the infant and less likely to rouse if the infant is distressed. An additional caution to parents considering bedsharing is that some studies suggest that SIDS, both in bedsharing situations and in crib sleeping situations, is more common among infants whose mothers smoke in the postpartum  period. The  connection  between  smoking  and bedsharing is not easily separated from that of smoking and SIDS, however, but it has been suggested that smokers who bedshare with an infant may put the infant at additional risk merely by virtue of additional smoke exposure from a parent smoking in bed.

Ongoing Challenges

As mentioned above, two additional factors that may contribute to infant mortality is the occurrence of premature labor and delivery and infants born with low birthweight. Premature labor is defined as labor that commences prior to 37 weeks of gestation, and low birthweight (LBW) is defined as an infant born weighing less than 5 pounds. Recently, maternal stress has been identified as a potential contributing factor to these poor outcomes; factor in early development, and there is growing evidence that prenatal psychological and environmental stresses are detrimental to pregnancy outcome. Importantly, there are significant disparities among ethnic groups in poor pregnancy outcomes including LBW and prematurity, and these may be contributing factors to the disparities in infant mortality seen in these groups. Maternal stress experiences during and after pregnancy can range from severe (e.g., trauma) to moderate (e.g., life event changes) to low (e.g., experience of daily hassles). Although some studies have shown minimal effects of prenatal stress on pregnancy, the majority of studies show that stress experienced throughout pregnancy can negatively affect pregnancy and infant outcomes. For example, significant stress around conception (e.g., death of a loved one, divorce) resulted in greater chances of delivering an infant with conotruncal heart defects,  neural  tube  defects,  and  isolated  cleft  lip, and trauma from an earthquake had a greater negative effect on gestational length if experienced earlier in the pregnancy. Stress experiences later in pregnancy are related to lower birth weights, reduced gestational length, preterm labor and delivery, and infant mortality; infants of stressed pregnancies have higher rates of childhood allergies and asthma, and perinatal stress appears to contribute to respiratory illness in infancy. The  implications  of  these  studies  are  important: 70% of infant perinatal deaths in the United States are related to LBW, preterm delivery, and restricted fetal growth. Those infants who survive the perinatal period are at greater risk for physical and cognitive developmental delays in later years.

Ethnic Disparities

Especially distressing when examining the frequency of poor pregnancy outcomes in the United States is the fact that the rate of occurrence of these events is significantly higher in non-white minority women than in Caucasian women. For example, the rate of infant mortality and LBW is nearly 2.5 times higher among African American women than among Caucasians. Although Hispanic women as a whole have been reported to have rates of LBW similar to those of whites, women of Puerto Rican descent and highly acculturated Hispanic women have significantly higher rates of LBW, pregnancy complications, and poor infant outcomes than their Caucasian counterparts. A significant component of considering poor pregnancy outcome in Hispanics is the “Hispanic Paradox” that has been described by several researchers, which states that although Latinas suffer similar socioeconomic challenges as other minority groups, they have lower rates of LBW and infant mortality than other minorities and white women. Recently, however, several published reports have called this paradox into question. These studies show, among other things, that although relatively few LBW infants are documented in Hispanics compared to other minority groups, errors in data recording and interpretation often mistakenly represent the rates of LBW in this population. For example, the reported lower rates of LBW in Hispanic infants in their sample reflected more preterm delivery and lower mean birth weights, and thus fewer infants who were small for their gestational ages, in Hispanic women compared to whites. This apparent paradox has also been attributed to other types of recording, sampling, and data-driven errors, such as not uniformly considering country of origin of those being sampled, and inappropriate selection of data. For example, when examining the Hispanic Paradox of infant mortality, removing infant deaths by SIDS or defining infant mortality as exogenous mortality increases the rate of infant mortality in Hispanics to 26% higher than that reported for whites. Thus, although the Hispanic Paradox of LBW and infant mortality is often cited, it may be, at some level, an artifact of data recording, selection, and interpretation, and should not be used to discount the importance of addressing poor birth outcomes in Latinas.

There are several other aspects of pregnancy success that are compromised in Hispanic women that emphasize the need for considering ethnic differences in addition to racial differences in pregnancy outcomes in minority women. For example, Americanborn Latinas have twice the rate of LBW, preterm delivery,  and  pregnancy  complications  as  Latinas born in Mexico, and Mexican-born Latinas living in America have higher rates of adverse pregnancy outcomes compared to Caucasians. Moreover, inadequate weight gain during pregnancy associated with poor infant outcome is a significant problem in Hispanic women compared to whites, and rates of pregnancy-related mortality are significantly higher in  Hispanic  and African American  women.  Taken together, these studies show a disproportionate frequency of poor pregnancy outcomes including complications, LBW, and preterm labor and delivery among minority women, both based on race and ethnicity, and underscore the need for understanding the complex biological, social, psychological, and cultural factors that contribute to these disparities.


Several investigators have suggested that stress, social structure and support, and other psychosocial factors are critical in the disparities in pregnancy outcome and infant mortality observed between minority and nonminority women. For example, although stress exposure during pregnancy is generally related to poorer outcomes regardless of racial or ethnic group, the negative effects of stress appear to be more severe for African American women than for Caucasians. For example, in a study of LBW in urban populations, Orr and colleagues found that although many prenatal behaviors such as smoking and hypertension in pregnancy were risk factors for LBW for both Caucasian and African American women, exposure to stressors was only predictive of LBW for African American women. Other investigators have shown that non-white women experience more adverse birth outcomes associated with prenatal stress, poor social support, and low self-esteem than white women, further suggesting that the combination of being a member of a minority group and psychosocial challenges may be especially perilous for pregnancy success. Interestingly, several investigators have noted ethnic differences in stress experiences and in the effects of coping with psychosocial stress. In a study of the factors contributing to clinical depression in adult women, Myers and colleagues found not only higher rates of stress-associated depression  among African American  women  compared to whites, but also that the combination of being Latina and high levels of perceived stress were predictive of more severe depression. Moreover, a significant stressor for non-American born Hispanic women is the challenge of assimilating and adjusting to American culture. This type of acculturation stress has been associated with generally poorer health in Latinas, more unhealthy behaviors such as smoking and drinking, and, importantly, with poor birth outcomes including LBW and prematurity. Thus, although women of all backgrounds are affected by stress in meaningful ways, women who are members of minority groups may experience more severe effects of stress on their mental and physical health, and on their unborn children.

An important aspect of considering the role of prenatal stress in pregnancy outcome and infant mortality is determining the biological mechanisms through which the external experience of stress exposure might translate into challenges to pregnancy. Stress is often defined as events, situations, emotions, and interactions that are perceived as negatively affecting the well-being of the individual or that cause responses perceived as harmful. The concept of a psychosocial stressor encompasses life experiences, including changes in personal life, job status, housing, domestic violence, and family makeup, which require adaptive coping behavior on the part of the affected individual. Stressor exposure activates a cascade of physiological events that help the individual to cope with stress-inducing stimuli and affects numerous physiological processes and systems. Perception of a stressor activates the sympathetic nervous system (SNS), a process known as the “fight or flight” response, and increases release of catecholamines including norepinephrine (NE). Activation of the hypothalamic-pituitary-adrenal (HPA) axis is a component of this response, beginning with the release of corticotrophin-releasing hormone (CRH) in the hypothalamus. CRH stimulates the anterior pituitary gland to release adrenocorticotrophin hormone (ACTH), which, in turn, stimulates cortisol release from the adrenal  cortex. These  neurochemical  changes  temporarily direct energy toward dealing with the stressful situation and away from other bodily functions. At the conclusion of short-term stress, the body quickly returns to homeostasis. In contrast, chronic stress prolongs activation of these systems and has detrimental effects on health and immunity. Existing evidence shows that stress experienced by mothers can have deleterious effects on pregnancy and infant development, and can adversely affect immunity and health. Further, there is evidence that there may be differences in stress responsiveness among racial and ethnic groups, suggesting that psychosocial stress may affect minority women differently. For example, African Americans have more extreme vasoconstrictive responses to laboratory stress and more pronounced neuroendocrine responses to work stress than Caucasians. Moreover, the actual experience of stress is likely to differ across ethnic groups, as minority groups in this country may experience more chronic stress as a result of deeper poverty, acculturation, decreased access to health care, and experience of institutional racism in the workplace.

Activation of the stress response affects human pregnancy,  and  women  experiencing  preterm  labor and delivery have significantly higher levels of plasma cortisol and CRH prior to onset of labor than women who deliver normally. Considerable interest has focused on the role of stress-related CRH production in modulation of labor and delivery because CRH may act as a signal for normal labor, and stress-related elevations of CRH may induce preterm labor. Others suggest that stress-related changes in immune function may contribute to these effects. A large body of work shows that psychological stress modulates physiology independently of pregnancy. Although these connections have an adaptive role in channeling energy to deal with the immediate stress, this relationship may result in changes in endocrine and immune function that put pregnancies at risk when the stress is of long duration as many psychological or social stressors may be. For example, when maternal CRH levels increase, they engage adrenal functioning in the fetus, producing higher fetal cortisol levels. Recently, animal and human studies support the role of maternal fetal-placental CRH in modulating infant development and behavior, and suggest that prenatal stress may exert  its  effect  on  infant  development  by  altering not only infant stress responsiveness, but also infant neural development. If, as some have suggested, infants with high basal cortisol levels are more prone to stress and distress, then a potential connection exists between high stress during pregnancy, infant neuroendocrine function, and neurobehavioral development. These studies show that prenatal stress has negative consequences for pregnancies and infants, suggest that factors such as ethnic background and stress  responsivity  may  mediate  these  interactions, and underscore the importance of understanding both the significance and mechanism of these effects to improve maternal and child health. Thus, not only may prenatal stress increase the risk of pregnancy complications  and  infant  mortality,  but  it  may potentially alter the course of infant development.


An important aspect of managing infant mortality and its root causes is developing public information programs to assist mothers and prenatal care providers in doing what is possible to prevent suboptimal pregnancy outcomes and ultimately, infant mortality. In the United States a variety of programs have been designed to both understand and prevent infant mortality. These efforts include programs to improve access to prenatal and newborn care, including Healthy Start, Medicaid, and the State Children’s Health Insurance Program (SCHIP). In addition, public health campaigns to promote healthy habits among parents expecting a child or caring for an infant to prevent child malnutrition as well as medical research to better understand and prevent birth defects, premature birth, and sudden infant death syndrome (SIDS) and to promote healthy infant development are focused on preventing and understanding infant mortality.

Improving maternal physical and mental health during pregnancy continues to be a priority for reducing infant mortality. Clear connections exist between some types  of  maternal  illnesses  during  pregnancy  and infant mortality, and there is compelling support for the hypothesis that stress may exacerbate the occurrence of these effects by reducing maternal immune function. Ongoing work in several laboratories is addressing this connection in the hope of further supporting healthy pregnancies. Importantly, infant mortality is not the only negative by-product of prenatal maternal stress. Maternal physical and mental stress is also associated with a variety of developmental problems and delays including chronic illnesses. Perinatal stress has already been linked to the development of childhood illnesses including asthma, allergy, and respiratory syncytial virus, and addressing the role of prenatal stress and maternal health and immunity in pregnancy and in infant immune system development will provide new data for the treatment and prevention of these effects.

The need for thorough, biopsychosocial approaches to supporting maternal prenatal experiences and outcome issues is clear. As already noted, the fact that a range of stressors from low to severe can alter developmental experiences of the embryo and fetus, and thus of the developing infant and child, has major relevance for child health and welfare. Low birth weight, one of the greatest predictors of at-risk status for infants in the first year of life, is an outcome implicated in a number of studies of psychosocial stress during  pregnancy.  Existing  data  clearly  show  that maternal stressors of various types need to be taken seriously during pregnancy, and thorough screening and intervention measures need to be investigated.

The development of prenatal support strategies groups to assist in buffering women and their infants against the impact of prenatal stress is critical, but interventions also must continue through the first years of childhood, as social-emotional development of infants with less optimal emotion regulation has been  linked  to  maladjustment  in  children  between ages 2 and 12, and across variables of social class and ethnicity. Infant–mother relationships can be enhanced through intervention, in situations of at-risk infants and even in those families with not-at-risk status. For example, research has shown that children and parents who were part of a counseling group, regardless of risk status, were able to improve goodness of fit between infant/child temperament and parent interaction, and had fewer psychiatric problems than controls at ages 5, 10, and 14 years. Provision of programs such as these for women with at-risk pregnancies may help to support not only healthy infants, but also family and child development by providing comprehensive biopsychosocial support at critical points in development.

Finally, early and continuous prenatal care helps identify conditions and behavior that can result in low birth-weight babies, such as smoking, drug and alcohol abuse, inadequate weight gain during pregnancy, and repeat pregnancy in 6 months or less. Additional preventive strategies include public education about folic acid consumption during pregnancy, placing infants on their backs to sleep to prevent SIDS, reducing teen pregnancy, and reducing transmission of HIV from mothers to infants. All these measures, combined with ongoing research, will support further understanding and prevention of infant mortality.


  1. Association of SIDS and Infant Mortality Programs,
  1. Beal, A. , Co, J. P., Dougherty, D., Jorsling, T., Kam, J., Perrin, J., et al. (2004). Quality measures for children’s health care. Pediatrics, 113(1, Pt. 2), 199–209.
  2. Bryce, , el Arifeen, S., Pariyo, G., Lanata, C., Gwatkin, D., Habicht, J. P., et al. (2003). Reducing child mortality: can public health deliver? Lancet, 12, 362(9378), 159–164.
  3. Doggrell, A. (2003). Recurrent hope for the treatment of preterm delivery.  Expert  Opinions  in  Pharmacotherapy,4(12), 2363–2366.
  4. Jones, , Steketee, R. W., Black, R. E., Bhutta, Z. A., Morris, S. S., & the Bellagio Child Survival Study Group. (2003). How many child deaths can we prevent this year? Lancet,362(9377), 65–71.
  5. Morris,  S.,  Black,  R.  E.,  &  Tomaskovic,  L.  (2003).Predicting the distribution of under-five deaths by cause in countries  without  adequate  vital  registration  systems. International Journal of Epidemiology, 32(6), 1041–1051.
  6. Noble, L. (2003). Developments in neonatal technology continue to improve infant outcomes. Pediatric Annals, 32(9), 595–603.
  7. Randall, , & Wilson, A. S. D. (2003). The 2002 annual report of the Regional Infant and Child Mortality Committee.Journal of Medicine, 56(12), 505–509.
  8. Roque, , Gillen-Goldstein, J., Funai, E., Young, B. K., & Lockwood, C. J. (2003). Perinatal outcomes in monoamniotic gestations. Journal of Maternal Fetal Neonatal Medicine,13(6), 414–421.
  9. Shrimpton, R. (2003). Preventing low birth-weight and reduction of child mortality. Transactions of the Royal Society of Tropical Medicine and Hygiene, 97(1), 39–42.
  10. Yu, V. Y. (2003). Global, regional and national perinatal and neonatal mortality. Journal of Perinatal Medicine, 31(5),376–379.
  11. Zupan, Z. (2003). Perinatal mortality and morbidity in developing countries. A global view. Medecine Tropicale (Marseilles), 63(4–5), 366–368.