Because of increased survival rates, neurodevelopmental issues, chronic medical problems, and sometimes complex family issues involved with prematurity, mental health clinicians commonly assess preterm clients and manage their behavioral and mental health problems. Understanding prematurity survival and neurodevelopmental outcomes is important for contextualizing the mental health problems seen in this high-risk population. This article provides a brief overview of prematurity outcomes in the domains of prematurity relevant to practicing child psychiatrists. Prematurity is also examined as it relates to parental mental health challenges, infant mental health outcomes, high frequency attention problems, and psychiatric disorders. The complex interactions between prematurity and family well-being are also highlighted. Finally, evidence-based treatment modalities involved in prevention and management are explored.
In the United States, more than 540,000 infants are born preterm each year, accounting for more than 12.8% of all births. Worldwide, the World Health Organization (WHO) reports preterm birth prevalence rates of 9.6%, with a total of 12.9 million preterm births a year. The cost of caring for these children in the United States is estimated to be $26 billion a year, partly because infants born with decreasing gestational age survive at increased rates and have increased needs.
A recent review of the causes of prematurity from the National Academies of Science highlights the co-occurrence of numerous individual and community-level biopsychosocial risk factors that contribute to this increase in preterm births. For example, a mother’s history of traumatic experiences is associated with increased risk for delivering prematurely and impoverished women are more likely than those who have financial resources to deliver a preterm infant at greater risk for morbidity and mortality.
Multiple medical conditions, including maternal infections and obstetric complications, are also associated with elevated risk for preterm delivery. The interactions between biologic and social risks create vulnerability to having a child born preterm, and these biologic and social factors continue to play active roles in the life of the developing infant and family. The poorer neurodevelopmental and physical and mental health outcomes that preterm infants experience then may interact with contextual factors to perpetuate an intergenerational risk for prematurity, as seen in the elevated risk for low birth weight infants among mothers born at low birth weights.
Because of increased survival rates, neurodevelopmental issues, chronic medical problems, and sometimes complex family issues involved with prematurity, mental health clinicians commonly assess preterm clients and manage their behavioral and mental health problems. Understanding prematurity survival and neurodevelopmental outcomes is important for contextualizing the mental health problems seen in this high-risk population.
This article provides a brief overview of prematurity outcomes in the domains of prematurity relevant to practicing child psychiatrists. Prematurity is also examined as it relates to parental mental health challenges, infant mental health outcomes, high frequency attention problems, and psychiatric disorders. The complex interactions between prematurity and family well-being are also highlighted. Finally, evidence-based treatment modalities involved in prevention and management are explored.
Introduction to prematurity
Definitions
Table 1 presents the language used to describe degrees of prematurity and birth weight categories. Although birth weight has been the primary independent variable, degree of prematurity is now preferred because several other factors beyond prematurity, including intrauterine growth restriction, influence birth weight and are likely associated with additional risks to the infant.
| Term | Abbreviation | Definition |
|---|---|---|
| Preterm | PT | Born before 37 wk gestational age |
| Very preterm | VPT | Born before 32 wk gestational age |
| Extremely preterm | EPT | Born before 28 wk gestational age |
| Low birth weight | LBW | Birth weight less than 2500 g (5.5 lb) |
| Very low birth weight | VLBW | Birth weight less than 1500 g (3.3 lb) |
| Extremely low birth weight | ELBW | Birth weight less than 1000 g (2.2 lb) |
| Micropremie | Birth weight less than 750 g (1.7 lb) |
Outcomes Related to Prematurity
Major advances have been made in preterm survival, both at a given birth weight level and at the boundary of viability, which is currently estimated at 22 to 24 weeks gestation. In 1987, only 34% of infants born at 24 weeks survived, compared with 59% in 2000.
Similarly, survival increases as gestational age progresses. In the United States, only 30% of infants born at 23 weeks survived in 2000, compared with 59% of those born at 24 weeks and 70% at 25 weeks. Variability is seen between groups, centers within networks, and internationally and could be based on research method and practice differences.
The advances in preterm infant survival can be attributed to antenatal and neonatal intensive care, such as surfactant use, aggressive resuscitation, mechanical ventilators, antenatal corticosteroids, and parenteral nutritional support. Despite these improvements in survival, corresponding improvements in developmental outcomes have not been reported.
Risk Factors for Adverse Neurodevelopmental Outcomes
Early gestational age and low birth weight are powerful determinants for the development of neurologic/behavioral problems and survival among preterm births. Other biologic factors, such as neurologic insults of intraventricular hemorrhage (IVH), hydrocephalus, periventricular leukomalacia, bronchopulmonary dysplasia, and retinopathy of prematurity, dramatically increase the rates of severe neurodevelopmental impairments. Growth and feeding problems, necrotizing enterocolitis, infections, and cardiac lesions may also result in neurologic insults and lead to worse developmental outcomes. For example, the more severe forms of IVH (grades 3 and 4) increase the risk for neurodevelopmental disability among preterm infants from 35% to 90%. Finally, multiple gestations are at risk for prematurity, death, and neurodevelopmental impairment. Prematurity may influence neurodevelopmental outcomes through disruption of the normal brain maturation that usually occurs in the protected intrauterine environment; direct insults to the central nervous system from medical complications during delivery and the neonatal intensive care unit (NICU) course; or impacts on the family from the additional stressors of parenting a high-risk child. Beyond the stresses related to concern about survival or increased care burden of a medically fragile infant, typical NICUs tend to focus on the medical optimization of the infant at the cost of prolonged separation from the parents and intrusive light and sound environmental stimuli.
As has long been recognized, neurodevelopmental outcomes are also strongly associated with community, family, and caregiving environments. Many risk factors, such as poverty, social support, parental discord, and parental mental health, may predate the pregnancy and may in fact increase in the perinatal period. Those who come from disadvantaged backgrounds are at a higher risk of preterm birth. The combination of a resource-deprived environment with biologic risk for prematurity creates a double jeopardy for poorer outcomes. For example, parental trauma-related symptoms are also well-recognized correlates to adverse infant neurodevelopmental and cognitive outcomes.
Research has also identified protective factors for neurodevelopmental resilience. Protective factors for extremely low birth weight (ELBW) infants include antenatal steroids, cesarean delivery, lack of social disadvantage, singleton birth, and higher gestational age. A consistent pattern of improved survival is seen in girls compared with boys.
General Neurodevelopmental Outcomes
Neurodevelopmental outcomes are becoming the benchmark of successful interventions. Although a comprehensive review of neurodevelopmental outcome literature is beyond the scope of this article, important neurodevelopmental domains that may relate to child psychiatry practice are briefly reviewed. Typical domains of neurodevelopmental outcomes include neurosensory; cognitive and motor; language and social developmental; learning; and adaptive outcomes. Generally, the risk for adverse neurodevelopmental outcomes increases with earlier birth and lower birth weight.
Table 2 presents common prevalence of neurodevelopmental outcomes seen in the preterm population. Beyond delays and disorders, variations of development can also be seen. For example, very low birth weight (VLBW) infants have a greater risk for having a nonverbal learning problem profile with more advanced verbal cognition, and challenges in visual spatial skills.
| Type of Outcome | Outcome | General Population | Low Birth Weight (<2500 g) | Very Low Birth Weight (<1500 g) | Extremely Low Birth Weight (<1000 g) |
|---|---|---|---|---|---|
| Survival | >99% | — | 86% | 43%–70% | |
| Neurosensory | Vision impairment a | <1% | 2% | 4%–24% | 9%–25% |
| Hearing loss | <1% | — | 1%–3% | 1%–28% | |
| Developmental | Cerebral Palsy | <1% | — | 6%–20% | 15%–23% |
| Speech and language delay | 6% | 3%–5% | 8%–45% | — | |
| Neurodevelopmental impairment b | <1% | 9%–61% | 22%–50% | ||
| Learning/ academic | Learning disabilities | 5%–20% | 17% | — | 34%–45% |
| Special education in school | 8% | — | 60%–70% | — | |
| Cognitive impairment | 1% | 7%–17% | 34%–37% | ||
| Behavioral | Attention deficit hyperactivity disorder | 5%–7% | 7%–30% | 9%–30% | 15%–40% |
| Autism | 1% | 2%—3% | — | — |
a Vision impairment is defined as having retinopathy of prematurity, blindness, myopia, or strabismus.
b Neurodevelopmental impairment is defined as <2 standard deviations below the mean on IQ or developmental tests, cerebral palsy, blindness, or significant hearing impairment.
Prematurity and mental health
The following sections review the impact of prematurity on parental mental health, infant mental health, and early parent–child relationships, and in preschool, school-aged, and adolescent/young adult survivors of prematurity.
Parental Posttraumatic Stress Disorder
Preterm birth can be considered to be associated with multiple traumatic experiences. Parents can experience an initial traumatic reaction at the preterm birth based on a stressful birth experience, again with the loss of the healthy imagined child, and sometimes with early separation from the infant and exposure to a highly technological, medical world that is not usually part of a birth experience. Parents may experience trauma-related symptoms, including anxiety, sleep problems, nightmares, and avoidance of medical visits in the context of caring for a seriously ill child who has been cared for in the highly technological environment of the NICU. Based on semi-structured interviews, the Impact of Events Scale, or the Perinatal Posttraumatic Stress Disorder Questionnaire, posttraumatic stress disorder (PTSD) symptoms of increased arousal, re-experiencing, and avoidance can occur acutely in the few days after the birth in mothers, and later in the weeks and up to 14 months postpartum in both mothers and fathers. Research suggests that, compared with mothers of healthy newborns, mothers whose newborns are VLBW and admitted to the NICU report more symptoms of anxiety and depression that may lead to ongoing psychological distress and parenting stress. One explanation for this pattern is the loss of control in not being able to protect their infant in the NICU. Poverty exacerbates the vulnerability for PTSD symptoms related to preterm births.
In general, PTSD symptoms have been associated with poorer physical health outcomes. Parental PTSD symptoms may jeopardize the physical health, developmental, and behavioral outcomes of high-risk infants. Early parental PTSD symptoms are associated with less sensitive and more controlling maternal behaviors. Perhaps based on this interaction pattern, parental PTSD can also predict future sleep and eating problems in infants by 18 months of age. Quinnell reported that maternal posttraumatic stress related to prematurity accounted for almost 6% of the variance in cognitive performance at 30 months among high-risk infants. Parental symptoms of PTSD, such as dissociation, hypervigilance, and numbing, may decrease parental ability to respond sensitively to infant cues, a deficit known to interfere with parent–child interactions, cognitive functioning, and child regulatory patterns.
Maternal Depression
The prevalence of major depressive disorder (MDD) in adults peaks during childbearing years and is higher in women than men. Thus, based on typical prevalence of MDD, it may be present in 10% to 25% of pregnant women. The hormonal changes after childbirth and the major life changes associated with a new birth may make mothers especially vulnerable to postpartum depression (PPD). Given the stresses involved with preterm births, researchers have shown associations between concurrent parental distress and severity of child illness in the NICU. Clinically relevant PPD symptoms after birth of a VLBW infant have been found to range from 12% to 60%, rates that do not necessarily distinguish themselves from those for full-term births. Rates of depressive symptoms seem to be highest at the child’s discharge from the hospital and decrease in most studies by 6 months of age. One study reported that higher levels of maternal socioeconomic factors were associated with a slower decrease in depressive symptoms after discharge and that maternal factors seemed to be stronger predictor of this slowed decline in symptoms than the infant’s medical status.
Although maternal depressive symptoms directly affect the quality of the infant–maternal interaction in full-term dyads and lead to greater rates of insecure attachment, more behavioral problems, and lower cognitive scores, their impact may be different in preterm families. For example, subclinical depressive symptoms do not increase the risk for insecure attachment relationships in term infants, but are associated with an increased risk for insecurity in preterm infants. This disruption leads to lower infant social engagement, unregulated fear regulation, and increased stress reactivity. Whether impacted by the challenges of parenting a preterm child, depression, or traumatic stress, negative maternal recollections of birth experience predicted greater report of internalizing and total problems at 5 years of age on the Child Behavior Checklist (CBCL) for preterm but not full-term children.
Prematurity and Infant–Parent Relationships
An extensive literature highlights the importance of early parent–child relationships as providing the foundation for later development. The context of the parent–child relationship can offer a buffer against biologic and environmental risk factors, including prematurity, and suboptimal relationships can increase the risk for adverse mental health outcomes. These relationships begin before the child is born, when parents develop an internal representation of the child. Infants also develop an internal representation of their parents based on the patterns of their interactions. These internal representations are shaped by ongoing interactions and experiences.
The internal representation about a preterm infant may be influenced by several external influences and fears. In an elegant experimental design, Stern and colleagues recruited infants born full-term and preterm at corrected age of 6 months. They randomly labeled some infants “preterm” and others “full-term” and assigned mothers to interact and describe one of these babies. Even mothers who had preterm infants viewed preterm infants less positively than those who were term. More importantly, infants in this study randomly labelled as full-term interacted more positively with mothers than those labeled as preterm, regardless of actual birth status, highlighting the power of negative attributions on infant behaviors. Early studies of prematurity supported these stereotypes, reporting that mothers of preterm infants worked harder than term mothers to engage their infant; they initiated interactions and talked more. These early reports have not been supported by more recent research, and factors in addition to prematurity have emerged. A more recent study suggested that prematurity itself has a direct influence on how dyads interact at 6 months. Although most term dyads were characterized as having “sensitive maternal style and cooperative infant” interaction styles, only approximately a quarter of preterm dyads interact this way, irrespective of medical status. Preterm dyads with less optimal interaction styles were at higher risk for problems with sleeping, eating, and emotional problems at 18 months.
In the general population, a mother’s state of mind regarding intimate relationships shapes interactions with her infant. Negative expectations and attributions (and positive expectations) in the neonatal period explain a substantial portion of variance of the infant’s behavior in interaction with the mother at 8 weeks and even at 24 months. Similarly, maternal early caregiving experiences that shape her own internal representation of herself in intimate relationships are associated with parent–child interaction style and infant development. Mothers who had a secure view of themselves within intimate relationships were more sensitive with their 3-month-old infants than those who were insecure. In fact, secure mothers who had preterm infants showed higher sensitivity to their infants than secure mothers with full-term infants. However, mothers with insecure representations of themselves showed less sensitivity with preterm infants compared with similar mothers of full-term infants. Thus, maternal sense of self can be either protective or increase an infant’s risk status.
Preterm infants are active participants in interactions with their parents, although their social development may be different from term infants. In a controlled study, preterm infants showed equal positive affect as their full-term counterparts in a parent–child interaction procedure; they also showed more negative affect and had a smaller latency to negative affect, and were less facially responsive than full-term infants. Even after controlling for medical risks, maternal anxiety showed a moderate association with infant facial responsivity, again suggesting an important complex interaction between maternal internal experiences and infant behaviors.
In typical development, infants develop focused attachment relationships at 7 to 9 months of age and begin to show a discriminated preference for specific caregivers, whom they seek out for comfort and nurturing. The quality of the relationship is associated with the infant caregiving experiences, and a secure attachment relationship predicts a range of positive social emotional outcomes. Some have theorized that preterm infants might show different patterns of attachment than full-term infants because of the medically necessary differences in the early caregiving environment. However, in general, studies have not found significant differences in rates of secure attachment behaviors between preterm VLBW infants up to 20 months and term infants, and stability of the attachment classification was similar to full-term infants. This pattern was not supported in a study of ELBW children, who showed higher rates of insecure attachment patterns, raising the question of whether the EBLW and associated medical problems contributed to the difference. These findings suggest that prematurity alone does not have a consistent effect on security of infant attachment in VLBW and low birth weight (LBW) infants.
Only one study has examined preschool attachment in ELBW infants and found substantial deviations from usual distributions, with more children (41%) classified as atypical than expected. This study raises important questions about differences in attachment in ELBW infants or the trajectory of attachment relationships beyond infancy in preterm children.
Findings from studies of infant mental health in preterm infants highlight the important contributions of parental experiences, their internal representations of their infants and the ways they interact with their infants. Medical issues and experiences may influence these internal representations. Data suggest that preterm infants may be particularly susceptible to either the protective or risk qualities in the parent–child relationship. Findings of limited differences in formal attachment quality may suggest that either the early challenges have abated by 7 to 9 months or that prematurity itself is not a direct contributor to the attachment relationship classification although further research is warranted.
Mental Health in Toddlers, Preschoolers, School-aged Children, and Adolescents Born Preterm
Mental health problems have been the focus of research on preterm infants through early adulthood. Many of the studies examining mental health issues in preterm survivors used parent report checklists to describe the mental health outcomes. Parent report measures, such as the commonly used CBCL, have some logistical and psychometric strengths. Parent report measures reflect parent perceptions, which may be influenced by factors other than the child’s behaviors. In fact, in a study of mothers of LBW children, mothers were twice as likely as teachers to identify their child as having behavioral problems. The strongest predictors of parent CBCL responses were caregiving quality and maternal depression, not child factors. This finding provides an important reminder about caution in interpreting results of studies that measure mental health outcomes solely based on parental report.
Most studies that focus on mental health in the first few months of life have examined maternal report of temperament. Early studies suggested that infants born prematurely had more difficult temperaments than those who were full-term. However, subsequent literature has yielded inconsistent results, with most studies showing little difference in the major temperamental categories of easy, difficult, and slow to warm up in preterm infants followed up as early as 6 weeks up to 4 years. Differences between term and preterm groups are generally only reported in preterm children who have significant central nervous system insults or developmental delays. Longitudinal studies suggest that any early differences in temperament decrease over time.
In one of the few measures of VLBW toddler mental health, Trevaud and colleagues reported that mental health scores for VLBW children were indistinguishable from reported norms on the Infant Toddler Social Emotional Assessment. Another study of preschoolers reported a 2- to 3-point difference in T-score for social and attention problems, a finding that is statistically significant but perhaps not clinically relevant. In that study of children born extremely preterm, prematurity and maternal history of adverse caregiving as a child independently predicted their social problems. Degree of prematurity seems to matter in predicting early childhood mental health problems, because preschoolers born extremely preterm were rated as having substantially higher rates of hyperactivity, inattention, peer problems, and emotional problems than those born very preterm. Early childhood literature suggests that some preterm infants may have some differences in temperamental development and early childhood mental health, but that other factors, including medical issues and parental factors, likely explain some of the differences.
Several studies have examined mental health outcomes in school-aged children born preterm, with inconsistent results. This section focuses on internalizing and externalizing symptoms. Attentional problems are discussed in the next section. In 5-year-olds, rates of total mental health and internalizing and externalizing problems are doubled in very preterm and extremely preterm children. Prematurity also showed a small to moderate effect size on these outcomes, and medical problems conferred additional risks. At 7 years of age, extremely preterm/VLBW children showed significant elevations in teacher-reported peer problems and in parent-reported hyperactivity.
More studies have examined the mental health status of premature infants when they are older. In a review of controlled studies published between 1980 and 2001 focusing on very preterm children 5 to 15 years of age, Bhutta and colleagues reported that 81% of the included studies showed elevated parent-reported internalizing and externalizing symptoms, with a relative risk for both domains greater than two. A more recent meta-analysis included nine well-controlled studies published between 1998 and 2008. Most of these studies examined children between 10 and 14 years old and included preterm children born near 27 weeks gestational age. The meta-analysis indicated that teachers reported more internalizing symptoms compared with control children, with an effect size of 0.28. No similar pattern was seen for parent’s report of child internalizing or externalizing symptoms.
Direct comparison of the two major reviews is impossible because of the different methodologies used. However, their divergent findings warrant discussion. Although the overall gestational ages were similar, the dates of the studies evaluated are different. Advances in caring for very preterm children may contribute to the more recent studies finding small to minimal effect size of prematurity on internalizing and externalizing symptoms. As with temperament and early childhood mental health, extreme prematurity (<25 weeks gestational age) confers additional risks and is associated with a sixfold risk for mental health problems and functional impairment at home and school.
Other groups have also examined the mechanism of adverse mental health outcomes in school-aged children. Examining preterm-born children, Whiteside-Mansell and colleagues reported that level of conflict at home and child temperament moderated the association between preterm birth and internalizing and externalizing symptoms at 8 years of age. These moderators interacted in such a way that high household conflict and difficult temperament were synergistic predictors of externalizing symptoms in preterm infants. Internalizing symptoms were affected uniquely by level of household conflict, which did not interact with temperament. Multiple studies have also reported that cognitive status may explain some externalizing symptoms in preterm children. Internalizing symptoms are less well accounted for by other developmental domains.
Studies of preterm infants through adolescence and young adulthood show provide inconsistent results. Two large non-United States studies of very preterm children have found no difference in internalizing or externalizing symptoms according to self-report at 19 and 31 years of age. In contrast, a large study of all children born from 1960 to 1968 in an urban hospital setting found that preterm children had twice the risk for self-reported depression and suicidal ideation compared with the more mature births. However, the higher social risk status of this population may have increased susceptibility to depression in the preterm children, who experienced “two hits.”
Other studies using self-report of adults born before 30 weeks gestational age also suggest an increased risk for psychopathology. More than one study reported that adult women born preterm endorse higher rates of internalizing symptoms compared with term infants. Sociodemographic risk factors, asthma, and exposure to violence predict these symptoms. Perinatal status predicted parental report of the adults’ internalizing symptoms, but not the self-report.
The mechanism of the increased rate of internalizing symptoms in young adults born preterm has not been explored extensively. In one study, LBW was only associated with increased risk for adult depression if at least one parent had depression, in which case 81% met criteria for depression. To the authors’ knowledge, specific genetic studies of this population’s psychiatric risks have not been performed. Abnormalities of hypothalamic-pituitary axis (HPA), known to be present at higher rates in preterm infants, have been implicated as one possible mechanism because of the increased perinatal exposure to stress and known associations between the HPA axis and mood and anxiety disorders, especially in adversity. However, these have not been examined in preterm children.
A decline in cerebellar volume in late adolescence in the very preterm population has been associated with a worsening of global mental health status and feelings of competence at that time. Family mental health and stress may also influence the development of internalizing symptoms. Overall, however, data increasingly suggest that psychiatric outcome of preterm children and adults is influenced by complex genetic, interactional, contextual, and biologic interactions.
In young adulthood, preterm survivors also show evidence of resilience, including lower rates of alcohol use and illicit drug use and male conduct problems compared with full-term adults. Generally, young adults born preterm, very preterm, and extremely preterm also report similar levels of overall self-esteem to their peers’, and their overall measures of quality of life are indistinguishable from adults born at term.
Despite the overall mental health resilience of young adults born preterm according to their self-report, parent reports reflect additional concerns. In two longitudinal studies that include parent’s report, parents identify higher rates of mental health problems, especially internalizing symptoms, in their preterm young adults than parents of term young adults. Correlations between self- and parent report in these studies are lower than in term dyads. It seems plausible that parents may be highly vigilant of their child’s potential vulnerability because of perinatal experiences and/or ongoing medical issue, and therefore are more sensitive to potential symptoms. Alternatively, however, surviving prematurity may confer a sense of resilience and reduce the preterm young adults’ perception of symptomatology.
Overall, measures of mental health in school-aged children and adults suggest that prematurity confers some risk for mental health problems. Across reporters, internalizing symptoms in girls and young women are the most consistently reported symptoms. Because of the notable discrepancy between parent and child reports when both are available, all parent report studies must be interpreted with some caution and attention to parental factors.
Prematurity and mental health
The following sections review the impact of prematurity on parental mental health, infant mental health, and early parent–child relationships, and in preschool, school-aged, and adolescent/young adult survivors of prematurity.
Parental Posttraumatic Stress Disorder
Preterm birth can be considered to be associated with multiple traumatic experiences. Parents can experience an initial traumatic reaction at the preterm birth based on a stressful birth experience, again with the loss of the healthy imagined child, and sometimes with early separation from the infant and exposure to a highly technological, medical world that is not usually part of a birth experience. Parents may experience trauma-related symptoms, including anxiety, sleep problems, nightmares, and avoidance of medical visits in the context of caring for a seriously ill child who has been cared for in the highly technological environment of the NICU. Based on semi-structured interviews, the Impact of Events Scale, or the Perinatal Posttraumatic Stress Disorder Questionnaire, posttraumatic stress disorder (PTSD) symptoms of increased arousal, re-experiencing, and avoidance can occur acutely in the few days after the birth in mothers, and later in the weeks and up to 14 months postpartum in both mothers and fathers. Research suggests that, compared with mothers of healthy newborns, mothers whose newborns are VLBW and admitted to the NICU report more symptoms of anxiety and depression that may lead to ongoing psychological distress and parenting stress. One explanation for this pattern is the loss of control in not being able to protect their infant in the NICU. Poverty exacerbates the vulnerability for PTSD symptoms related to preterm births.
In general, PTSD symptoms have been associated with poorer physical health outcomes. Parental PTSD symptoms may jeopardize the physical health, developmental, and behavioral outcomes of high-risk infants. Early parental PTSD symptoms are associated with less sensitive and more controlling maternal behaviors. Perhaps based on this interaction pattern, parental PTSD can also predict future sleep and eating problems in infants by 18 months of age. Quinnell reported that maternal posttraumatic stress related to prematurity accounted for almost 6% of the variance in cognitive performance at 30 months among high-risk infants. Parental symptoms of PTSD, such as dissociation, hypervigilance, and numbing, may decrease parental ability to respond sensitively to infant cues, a deficit known to interfere with parent–child interactions, cognitive functioning, and child regulatory patterns.
Maternal Depression
The prevalence of major depressive disorder (MDD) in adults peaks during childbearing years and is higher in women than men. Thus, based on typical prevalence of MDD, it may be present in 10% to 25% of pregnant women. The hormonal changes after childbirth and the major life changes associated with a new birth may make mothers especially vulnerable to postpartum depression (PPD). Given the stresses involved with preterm births, researchers have shown associations between concurrent parental distress and severity of child illness in the NICU. Clinically relevant PPD symptoms after birth of a VLBW infant have been found to range from 12% to 60%, rates that do not necessarily distinguish themselves from those for full-term births. Rates of depressive symptoms seem to be highest at the child’s discharge from the hospital and decrease in most studies by 6 months of age. One study reported that higher levels of maternal socioeconomic factors were associated with a slower decrease in depressive symptoms after discharge and that maternal factors seemed to be stronger predictor of this slowed decline in symptoms than the infant’s medical status.
Although maternal depressive symptoms directly affect the quality of the infant–maternal interaction in full-term dyads and lead to greater rates of insecure attachment, more behavioral problems, and lower cognitive scores, their impact may be different in preterm families. For example, subclinical depressive symptoms do not increase the risk for insecure attachment relationships in term infants, but are associated with an increased risk for insecurity in preterm infants. This disruption leads to lower infant social engagement, unregulated fear regulation, and increased stress reactivity. Whether impacted by the challenges of parenting a preterm child, depression, or traumatic stress, negative maternal recollections of birth experience predicted greater report of internalizing and total problems at 5 years of age on the Child Behavior Checklist (CBCL) for preterm but not full-term children.
Prematurity and Infant–Parent Relationships
An extensive literature highlights the importance of early parent–child relationships as providing the foundation for later development. The context of the parent–child relationship can offer a buffer against biologic and environmental risk factors, including prematurity, and suboptimal relationships can increase the risk for adverse mental health outcomes. These relationships begin before the child is born, when parents develop an internal representation of the child. Infants also develop an internal representation of their parents based on the patterns of their interactions. These internal representations are shaped by ongoing interactions and experiences.
The internal representation about a preterm infant may be influenced by several external influences and fears. In an elegant experimental design, Stern and colleagues recruited infants born full-term and preterm at corrected age of 6 months. They randomly labeled some infants “preterm” and others “full-term” and assigned mothers to interact and describe one of these babies. Even mothers who had preterm infants viewed preterm infants less positively than those who were term. More importantly, infants in this study randomly labelled as full-term interacted more positively with mothers than those labeled as preterm, regardless of actual birth status, highlighting the power of negative attributions on infant behaviors. Early studies of prematurity supported these stereotypes, reporting that mothers of preterm infants worked harder than term mothers to engage their infant; they initiated interactions and talked more. These early reports have not been supported by more recent research, and factors in addition to prematurity have emerged. A more recent study suggested that prematurity itself has a direct influence on how dyads interact at 6 months. Although most term dyads were characterized as having “sensitive maternal style and cooperative infant” interaction styles, only approximately a quarter of preterm dyads interact this way, irrespective of medical status. Preterm dyads with less optimal interaction styles were at higher risk for problems with sleeping, eating, and emotional problems at 18 months.
In the general population, a mother’s state of mind regarding intimate relationships shapes interactions with her infant. Negative expectations and attributions (and positive expectations) in the neonatal period explain a substantial portion of variance of the infant’s behavior in interaction with the mother at 8 weeks and even at 24 months. Similarly, maternal early caregiving experiences that shape her own internal representation of herself in intimate relationships are associated with parent–child interaction style and infant development. Mothers who had a secure view of themselves within intimate relationships were more sensitive with their 3-month-old infants than those who were insecure. In fact, secure mothers who had preterm infants showed higher sensitivity to their infants than secure mothers with full-term infants. However, mothers with insecure representations of themselves showed less sensitivity with preterm infants compared with similar mothers of full-term infants. Thus, maternal sense of self can be either protective or increase an infant’s risk status.
Preterm infants are active participants in interactions with their parents, although their social development may be different from term infants. In a controlled study, preterm infants showed equal positive affect as their full-term counterparts in a parent–child interaction procedure; they also showed more negative affect and had a smaller latency to negative affect, and were less facially responsive than full-term infants. Even after controlling for medical risks, maternal anxiety showed a moderate association with infant facial responsivity, again suggesting an important complex interaction between maternal internal experiences and infant behaviors.
In typical development, infants develop focused attachment relationships at 7 to 9 months of age and begin to show a discriminated preference for specific caregivers, whom they seek out for comfort and nurturing. The quality of the relationship is associated with the infant caregiving experiences, and a secure attachment relationship predicts a range of positive social emotional outcomes. Some have theorized that preterm infants might show different patterns of attachment than full-term infants because of the medically necessary differences in the early caregiving environment. However, in general, studies have not found significant differences in rates of secure attachment behaviors between preterm VLBW infants up to 20 months and term infants, and stability of the attachment classification was similar to full-term infants. This pattern was not supported in a study of ELBW children, who showed higher rates of insecure attachment patterns, raising the question of whether the EBLW and associated medical problems contributed to the difference. These findings suggest that prematurity alone does not have a consistent effect on security of infant attachment in VLBW and low birth weight (LBW) infants.
Only one study has examined preschool attachment in ELBW infants and found substantial deviations from usual distributions, with more children (41%) classified as atypical than expected. This study raises important questions about differences in attachment in ELBW infants or the trajectory of attachment relationships beyond infancy in preterm children.
Findings from studies of infant mental health in preterm infants highlight the important contributions of parental experiences, their internal representations of their infants and the ways they interact with their infants. Medical issues and experiences may influence these internal representations. Data suggest that preterm infants may be particularly susceptible to either the protective or risk qualities in the parent–child relationship. Findings of limited differences in formal attachment quality may suggest that either the early challenges have abated by 7 to 9 months or that prematurity itself is not a direct contributor to the attachment relationship classification although further research is warranted.
Mental Health in Toddlers, Preschoolers, School-aged Children, and Adolescents Born Preterm
Mental health problems have been the focus of research on preterm infants through early adulthood. Many of the studies examining mental health issues in preterm survivors used parent report checklists to describe the mental health outcomes. Parent report measures, such as the commonly used CBCL, have some logistical and psychometric strengths. Parent report measures reflect parent perceptions, which may be influenced by factors other than the child’s behaviors. In fact, in a study of mothers of LBW children, mothers were twice as likely as teachers to identify their child as having behavioral problems. The strongest predictors of parent CBCL responses were caregiving quality and maternal depression, not child factors. This finding provides an important reminder about caution in interpreting results of studies that measure mental health outcomes solely based on parental report.
Most studies that focus on mental health in the first few months of life have examined maternal report of temperament. Early studies suggested that infants born prematurely had more difficult temperaments than those who were full-term. However, subsequent literature has yielded inconsistent results, with most studies showing little difference in the major temperamental categories of easy, difficult, and slow to warm up in preterm infants followed up as early as 6 weeks up to 4 years. Differences between term and preterm groups are generally only reported in preterm children who have significant central nervous system insults or developmental delays. Longitudinal studies suggest that any early differences in temperament decrease over time.
In one of the few measures of VLBW toddler mental health, Trevaud and colleagues reported that mental health scores for VLBW children were indistinguishable from reported norms on the Infant Toddler Social Emotional Assessment. Another study of preschoolers reported a 2- to 3-point difference in T-score for social and attention problems, a finding that is statistically significant but perhaps not clinically relevant. In that study of children born extremely preterm, prematurity and maternal history of adverse caregiving as a child independently predicted their social problems. Degree of prematurity seems to matter in predicting early childhood mental health problems, because preschoolers born extremely preterm were rated as having substantially higher rates of hyperactivity, inattention, peer problems, and emotional problems than those born very preterm. Early childhood literature suggests that some preterm infants may have some differences in temperamental development and early childhood mental health, but that other factors, including medical issues and parental factors, likely explain some of the differences.
Several studies have examined mental health outcomes in school-aged children born preterm, with inconsistent results. This section focuses on internalizing and externalizing symptoms. Attentional problems are discussed in the next section. In 5-year-olds, rates of total mental health and internalizing and externalizing problems are doubled in very preterm and extremely preterm children. Prematurity also showed a small to moderate effect size on these outcomes, and medical problems conferred additional risks. At 7 years of age, extremely preterm/VLBW children showed significant elevations in teacher-reported peer problems and in parent-reported hyperactivity.
More studies have examined the mental health status of premature infants when they are older. In a review of controlled studies published between 1980 and 2001 focusing on very preterm children 5 to 15 years of age, Bhutta and colleagues reported that 81% of the included studies showed elevated parent-reported internalizing and externalizing symptoms, with a relative risk for both domains greater than two. A more recent meta-analysis included nine well-controlled studies published between 1998 and 2008. Most of these studies examined children between 10 and 14 years old and included preterm children born near 27 weeks gestational age. The meta-analysis indicated that teachers reported more internalizing symptoms compared with control children, with an effect size of 0.28. No similar pattern was seen for parent’s report of child internalizing or externalizing symptoms.
Direct comparison of the two major reviews is impossible because of the different methodologies used. However, their divergent findings warrant discussion. Although the overall gestational ages were similar, the dates of the studies evaluated are different. Advances in caring for very preterm children may contribute to the more recent studies finding small to minimal effect size of prematurity on internalizing and externalizing symptoms. As with temperament and early childhood mental health, extreme prematurity (<25 weeks gestational age) confers additional risks and is associated with a sixfold risk for mental health problems and functional impairment at home and school.
Other groups have also examined the mechanism of adverse mental health outcomes in school-aged children. Examining preterm-born children, Whiteside-Mansell and colleagues reported that level of conflict at home and child temperament moderated the association between preterm birth and internalizing and externalizing symptoms at 8 years of age. These moderators interacted in such a way that high household conflict and difficult temperament were synergistic predictors of externalizing symptoms in preterm infants. Internalizing symptoms were affected uniquely by level of household conflict, which did not interact with temperament. Multiple studies have also reported that cognitive status may explain some externalizing symptoms in preterm children. Internalizing symptoms are less well accounted for by other developmental domains.
Studies of preterm infants through adolescence and young adulthood show provide inconsistent results. Two large non-United States studies of very preterm children have found no difference in internalizing or externalizing symptoms according to self-report at 19 and 31 years of age. In contrast, a large study of all children born from 1960 to 1968 in an urban hospital setting found that preterm children had twice the risk for self-reported depression and suicidal ideation compared with the more mature births. However, the higher social risk status of this population may have increased susceptibility to depression in the preterm children, who experienced “two hits.”
Other studies using self-report of adults born before 30 weeks gestational age also suggest an increased risk for psychopathology. More than one study reported that adult women born preterm endorse higher rates of internalizing symptoms compared with term infants. Sociodemographic risk factors, asthma, and exposure to violence predict these symptoms. Perinatal status predicted parental report of the adults’ internalizing symptoms, but not the self-report.
The mechanism of the increased rate of internalizing symptoms in young adults born preterm has not been explored extensively. In one study, LBW was only associated with increased risk for adult depression if at least one parent had depression, in which case 81% met criteria for depression. To the authors’ knowledge, specific genetic studies of this population’s psychiatric risks have not been performed. Abnormalities of hypothalamic-pituitary axis (HPA), known to be present at higher rates in preterm infants, have been implicated as one possible mechanism because of the increased perinatal exposure to stress and known associations between the HPA axis and mood and anxiety disorders, especially in adversity. However, these have not been examined in preterm children.
A decline in cerebellar volume in late adolescence in the very preterm population has been associated with a worsening of global mental health status and feelings of competence at that time. Family mental health and stress may also influence the development of internalizing symptoms. Overall, however, data increasingly suggest that psychiatric outcome of preterm children and adults is influenced by complex genetic, interactional, contextual, and biologic interactions.
In young adulthood, preterm survivors also show evidence of resilience, including lower rates of alcohol use and illicit drug use and male conduct problems compared with full-term adults. Generally, young adults born preterm, very preterm, and extremely preterm also report similar levels of overall self-esteem to their peers’, and their overall measures of quality of life are indistinguishable from adults born at term.
Despite the overall mental health resilience of young adults born preterm according to their self-report, parent reports reflect additional concerns. In two longitudinal studies that include parent’s report, parents identify higher rates of mental health problems, especially internalizing symptoms, in their preterm young adults than parents of term young adults. Correlations between self- and parent report in these studies are lower than in term dyads. It seems plausible that parents may be highly vigilant of their child’s potential vulnerability because of perinatal experiences and/or ongoing medical issue, and therefore are more sensitive to potential symptoms. Alternatively, however, surviving prematurity may confer a sense of resilience and reduce the preterm young adults’ perception of symptomatology.
Overall, measures of mental health in school-aged children and adults suggest that prematurity confers some risk for mental health problems. Across reporters, internalizing symptoms in girls and young women are the most consistently reported symptoms. Because of the notable discrepancy between parent and child reports when both are available, all parent report studies must be interpreted with some caution and attention to parental factors.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
