A short-term longitudinal study of 83 families compared patterns of development between full-term small for gestational age (SGA) and normal birth weight (NBW) infants. Data were collected on infant temperament and maternal interaction at 3 and 6 months, and infant developmental outcomes at 6 months in order to investigate relationships between infant and maternal behavior, and developmental outcomes as a function of birth weight. Findings revealed few differences between SGA and NBW groups. However, the relations between infant temperament and maternal behavior varied as a function of birth weight and home environment. Specifically, more positive home environments were associated with higher ratings of maternal behavior and lower levels of infant negative reactivity for SGA but not for NBW infants. In addition, higher negative reactivity was related to lower performance on both the mental and psychomotor scales of the Bayley Scales of Infant Development (BSID), with stronger associations reported for SGA infants than for NBW infants.
The birth of an infant who is small for gestational age (SGA) has been associated with increased risk of infant mortality and morbidity,1 as well as a wide range of other health and behavioral problems.2-4 Although a large body of literature attests to differences between groups that vary on dimensions of birth weight, gestational age, and degree of risk, the purpose of this paper is to more fully explore the mechanisms by which birth weight is associated with developmental outcomes, independent of risks associated with gestational age and neonatal complications.
The etiology of SGA is varied and frequently unknown. In developed countries, common sources of growth restriction include placental or fetal anomalies and infection.5 For example, growth restriction occurs when the placenta is unable to provide adequate nutritional resources to the fetus. Furthermore, infections such as toxoplasmosis and cytomegalovirus (CMV) may also contribute to growth restriction.6 Additionally, environmental factors such as maternal short stature, malnutrition, poor prenatal care, smoking, drug use, and hypertension are also known to be related to poor fetal growth. For example, reported differences among mothers of SGA infants and normal birth weight (NBW) infants on measures of height and weight7 may account for birth weight differences among the infants; that is, smaller mothers are more likely to have smaller babies.8,9
Research on the cognitive and behavioral sequelae of SGA infants is inconclusive.3 A number of investigators have reported differences between SGA and NBW infants on motor development,10 cognitive development,9,11-15 and early behavioral characteristics16,17 typically favoring NBW infants. In contrast, others have reported no differences between SGA and NBW infants.12,18-20
The mixed evidence on the developmental outcomes of SGA infants has led to the hypothesis that a number of moderating factors need to be considered in determining the behavioral trajectories of these infants. That is, the long-term development of SGA infants depends not only on the degree to which growth was compromised in utero but also on the characteristics of the infant (i.e., temperament) and her environment (e.g., maternal behavior, socioeconomic status).21 Illustrative is the research on full-term infants that has established that differences in birth weight are associated with differences in early infant behavior and temperament. Growth-restricted infants show signs of early autonomic instability, poor use of environmental stimuli, more insulated cry states,16 and poor motor performance17 on the Neonatal Behavioral Assessment Scale (NBAS) as compared with NBW infants. Similarly, SGA infants tend to be in awake active states and to look at people less frequently than NBW infants.13 In addition, whereas NBW infants have been reported to be more irritable than SGA infants,22 SGA infants are reported to have lower thresholds of response than NBW infants.23
In addition to behavioral differences among infants, characteristics of home environments and parenting behaviors have also been found to differ as a function of birth weight. For example, differences between SGA and NBW parents in terms of education levels7 and home environment24 have been suggested to account for behavioral differences in the infants. More specifically, mothers with fewer years of education and fewer resources may provide fewer opportunities for stimulation. Furthermore, mothers of growth-restricted infants have been noted to express more concern about their infants than mothers of NBW infants.25 In sum, whereas differences in family background characteristics may contribute to the incidence of SGA births (e.g., through parental health behaviors), ongoing differences in family environments and parental behavior continue to exert influence on the developmental trajectories of these infants.
There is fairly wide consensus that the caregiver-infant relationship develops, in part, as a function of both infant and caregiver characteristics. In particular, studies of infant temperament suggest that infants described as more difficult (e.g., more highly reactive, less easily soothed) are perceived as more challenging for parents and elicit different caregiving behaviors than a social and easy infant.26 In the case of SGA infants, the relationship between temperament and maternal behavior has been shown to differ from NBW infants and to differentially predict developmental outcomes.13 Specifically, higher levels of maternal stimulation and coordination with the infant were related to more optimal scores on the Mental Development Index (MDI) of the Bayley Scales of Infant Development (BSID) in NBW infants, whereas maternal calming and infant quietness were more optimal for SGA infants.13 In addition, the association between infant temperament and MDI scores varied as a function of birth weight, with a stronger relationship between difficult temperament and MDI performance among SGA infants.23 Despite few between-group differences and small sample size in both studies, these data provide evidence of the complex interactions between intrauterine growth and developmental processes.
In this study we compare the behavioral development of NBW and healthy full-term SGA infants during the first 6 months of life. We selected this population to better assess the effects of intrauterine growth restriction without the confounds of prematurity and other types of insults. Based on previous research, we predicted that characteristics of infant temperament (i.e., positive and negative reactivity) would vary as a function of birth weight, such that SGA infants would be rated higher on negative reactivity than NBW infants. In addition, we hypothesized that the relation between maternal behavior and infant temperament would be associated with infant birth weight. That is, we expected that the relationship between infant reactivity, particularly negative reactivity, and maternal behavior would be different for SGA infants as compared with NBW infants. Furthermore, we hypothesized that the association between birth weight and developmental outcomes would vary as a function of both infant characteristics and maternal behavior. Therefore, similar to the results reported in previous studies,23 we expected that the relationships between the MDI and the temperament and parenting variables would differ between SGA and NBW infants. Given the inconclusive results of prior research, we made no predictions about the direction of these relationships; that is, we did not expect parents of SGA infants to be rated higher or lower on specific measures than NBW parents, but we did expect different patterns of associations.
Department of Psychology, University of Rhode Island, Kingston, Rhode Island (GORMAN)
Department of Family and Community Medicine, University of California San Francisco-Fresno, Fresno, California (LOURIE)
Center for Molecular and Behavioral Neuroscience, Rutgers University, Newark, New Jersey (CHOUDHURY)
Address for reprints: Kathleen S. Gorman, Ph.D., Director, Feinstein Center for a Hunger Free America, University of Rhode Island, Providence Campus, 80 Washington St., Providence, RI 02903; e-mail: firstname.lastname@example.org; fax: 401-277-5478.
This research was conducted as a dissertation study submitted by Andrea E. Lourie to the University of Vermont in partial fulfillment of the requirements for the doctoral degree in psychology.
Acknowledgments. This study was supported by a grant to Andrea Lourie from Child and Adolescent Psychology Training and Research, Inc., Burlington, Vermont. We would like to thank Susan Crockenberg, Amy Ducker, and our undergraduate research team for their assistance in various aspects of this project.