WAIMH Handbook of Infant Mental Health, Perspectives on Infant Mental Health / Edition 1 available in Hardcover
WAIMH Handbook of Infant Mental Health, Perspectives on Infant Mental Health / Edition 1
- ISBN-10:
- 0471189413
- ISBN-13:
- 9780471189411
- Pub. Date:
- 12/29/1999
- Publisher:
- Wiley
WAIMH Handbook of Infant Mental Health, Perspectives on Infant Mental Health / Edition 1
Hardcover
Buy New
$349.75Overview
Product Details
| ISBN-13: | 9780471189411 |
|---|---|
| Publisher: | Wiley |
| Publication date: | 12/29/1999 |
| Series: | WAIMH Handbook of Infant Mental Health , #1 |
| Edition description: | Volume 1 |
| Pages: | 432 |
| Product dimensions: | 7.36(w) x 10.37(h) x 1.32(d) |
About the Author
Hiram E. Fitzgerald is Associate Provost for University Outreach and Engagement, University Distinguished Professor of Psychology, and Adjunct Professor in Human Development and Family Studies at Michigan State University.
Read an Excerpt
Note: The Figures and/or Tables mentioned in this sample chapter do not appear on the Web.
1
Ecological Perspectives on Developmental RiskArnold Sameroff
Introduction
Infant mental health is a notable goal in its own right, but more importantly, it lays the groundwork for later mental health and success at social and professional life goals. Although claims have been made that each life period has its own determinants with success or failure at previous points in time having little to do with what happens afterward (Lewis, 1997), the evidence for continuity in both individuals and their life circumstances point to the significance of a good start for successful later development. In this chapter we examine the ingredients of a good start and how it relates to what follows.
When we appeal to a good start we are immediately mired in the nature-nurture debate. The literature is filled with advocates for either characteristics of the infant or the caregivers as primary determinants. For the purposes of this chapter there is no debate on the academic side, for most scientists recognize the inextricability of one from the other (Ford & Lerner, 1992; Sameroff, 1983, 1995). Moreover, there is a dynamic transactional relation between nature and nurture in which both are constantly being changed by their experience with each other. However, on the clinical side, there frequently is a need to separate nature from nurture. Although some interventions with infants are directed at the interaction between parent and child (Brazelton & Cramer, 1990; Clark, 1985; Greenspan & Weider, 1999; Lieberman, 1999; McDonough, 1995; Pawl, 1999), many others are aimed at either changing the child or changing aspects of the caregiving environment. As one moves toward more distal influences such as neighborhood characteristics or socioeconomic status, interventions deal with aspects of the environment at a level where the child is not a participant. Transactional models of intervention have been proposed where interventions can be directly targeted at different aspects of the developmental system: the infant alone, the parent alone, or their interaction (Sameroff & Fiese, 1999). For purposes of this discussion, however, the focus will be on the social environment as the source of risk factors, leaving to other chapters the description of infant characteristics that place the child at risk.
Rationale for Risk Assessment
Developmental outcomes for many children have not been successful. The Children's Defense Fund (1995) estimates that between 3 and 10 million children experience domestic violence yearly, with more than a million confirmed child abuse or neglect cases in 1993. Mental health also continues to be a major problem, with approximately 20 percent of children having diagnosable disorders (U.S. Department of Health and Human Services, 1990). Surveys of child health found that 13.4 percent of children in the United States have emotional or behavioral disorders, 6.5 percent have learning disabilities, and 4 percent have developmental delays (Zill & Schoenborn, 1990). Moreover, these disorders have stability over time, with 40 to 60 percent of children with a psychiatric diagnosis at one point in time continuing to have a diagnosis when assessed several years later (Costello & Angold, 1995).
The pursuit of happiness is a fundamental right in our society, yet the goal of achieving a sense of satisfaction with one's abilities and achievements is becoming increasingly elusive for large groups of children. For example, over half of today's 10- to 17-year-olds engage in two or more behaviors that we would regard as indicating a lack of competence in today's society. These include unsafe sex, teenage pregnancy, drug or alcohol abuse, school fail- ure, delinquency, and crime. Moreover, 10 percent of these youth engage in all of these risks (Dryfoos, 1991). Clearly there is a need to understand why so many youth are unsuccessful by society's standards. Decreasing these numbers requires an understanding of the causes of these problems.
One of the clear correlates of increases in child problems is the decline in the quality of children's environments. Concurrent with the high level of problems among children, family resources for coping with these problems have diminished. In 1991, 22 percent of children lived in families with incomes below the poverty line, the highest rate since the early 1960s (Children's Defense Fund, 1992). During the same period the proportion of female-headed single-parent homes increased from 7 percent to more than 21 percent (McLanahan, Astone, & Marks, 1991). Moreover, 75 percent of mothers of school-age children are in the workforce now, compared to about 50 percent in 1970 (U.S. Department of Health and Human Services, 1993).
Declines in family resources for supporting child development have often been offset by social programs that offer compensatory professional or economic means. Unfortunately, we are living in a historical period in which these resources are also in decline. For all these reasons, if we are to utilize beneficially the intervention resources that are still available, it is of the utmost importance that we have a clear understanding of the sources of problems in child development.
Terminology
Where science is seen as the search for causes, a discussion of risk factors may appear to be a substitute for a more basic understanding of why individuals succeed or fail. Where risks represent only probabilities, causes would seem to represent truths. But the history of research into the etiology of all biological disorders has demonstrated that there are no single sufficient causes. The phrase risk factors itself arose from epidemiological research seeking the cause of heart disease (Costello & Angold, 2000). In the most comprehensive of these efforts, the Framingham Study, it was found that no one factor was either necessary or sufficient (Dawber, 1980). Hypertension, obesity, lack of exercise, and smoking all made significant contributions to heart disease at the population level, but for any single affected individual there was a different combination of these factors. We discovered a similar result in our search for the causes of infant mental problems. It is not a single factor that causes such difficulties, but a set of factors that contributes to the outcome.
In seeking the causes of disorders, epidemiological research historically has cen- tered on negative outcomes. Within the field of developmental psychopathology an alternative perspective has been to seek the roots of competence (Masten & Coatsworth, 1995) as well as the sources of problem behavior. Although initial forays into studies of the etiology of serious mental disorders sought to identify risk factors in the family transmission of such diagnoses as schizophrenia, it was discovered that a number of children who had been judged vulnerable because they were raised by schizophrenic parents living in conditions of adversity showed no obvious deleterious effects of their rearing environment. This observation led Garmezy (1974) to shift his focus to the search for protective factors in stress-resistant children, resulting in major efforts to identify the sources of their resiliency. Since that time research on vulnerability and resilience have become inextricably intertwined.
The terms risk, protective, vulnerable, and resilient have been defined in different ways by different writers with different goals. A full exposition of these differences and their implications is beyond the scope of this chapter. However, by the end the reader should have enough of a connota- tive understanding to appreciate the general issues and be aware of appropriate directions for further study (cf. Luthar, 1993; Zimmerman & Arunkumar, 1994).
Although the purpose of this chapter is to identify risk factors, the conclusion will be that there may be no such things as risk factors. Reaching that conclusion will entail exposition intended to illuminate fundamental developmental processes while providing the rationale for such a conclusion. The outline of the argument to be presented is that factors that produce maladaptive variations in the life course are no different in kind, and perhaps even no different in degree, from those that produce adaptive variations. In other words, for the most part risk factors and protective factors are the same and are represented in the lives of the mentally healthy as well as the disturbed. For example, Masten, Best, and Garmezy (1990) define resilience as producing "successful adaptation despite challenging or threatening circumstances." From a developmental perspec- tive all progress is a response to challenging circumstances. Even the more extreme aspects of threatening situations may differ in degree rather than in kind from the situations faced by most children. From this perspective the identification of variables as risk factors is a heuristic to find targets for intervention rather than a theoretical strategy for understanding the processes that produced the condition toward which the intervention is directed.
To clarify the risks that affect the lives of children, I move toward a number of conclusions through this chapter. The first is that the identification of risk factors is an exercise in estimating probabilities, not finding causes. Some factors seem to affect all children in all families; many others affect only certain children in certain families. The second is the recognition that child development has multiple contributors at multiple levels of the child's social ecology.
Developmental research has concerned itself with delineating the processes and products that typify normative human performance. As developmental theories have moved from simplistic constitutional or environmental models to increasingly more complex transactional and systems views (Sameroff, 1995; Thelen & Smith, 1994), our understanding of the determinants of these processes and products has become more complex. Similarly, our interpretation of individual differences has become more complex. To the extent that deterministic causal models can be discovered, variations in outcomes can be directly related to variations in causes. This was the fundamental tenet of germ theories of disease that typified the medical model (Engel, 1977). However, this tenet has increasingly been challenged as deterministic models are converted into probablistic ones.
The multiple meanings and applications of probability concepts are fundamental issues in the study of risk in development. In the germ model of disease, probability of outcome was simply a measure of error variance. There was a true effect of the germ, but that effect might be more or less difficult to assess. However, when it was discovered that the same germ would have different effects on different individuals, probability entered the equation in a new way. Within epidemiology the focus was less on the agent and more on interactions among agent, host, and environment (Cooper & Morgan, 1973). A more organismic understanding of the effects of germs evolved in which the characteristics of the infected individual played as great a role in the expression of a disease as the germs themselves. Probability now is viewed less as a nuisance, related to measurement error in a deterministic system, and more as the essence of all dynamic processes in complex systems, from the most fundamental subatomic particles to the most complex human institutions. Yet another meaning of probability is found in epidemiology, the discipline most concerned with the study of risk factors. Epidemiological research is devoted to understanding how disorders are distributed in populations, and determining the associated probabilities of various subgroups being affected, rather than how they operate within individuals.
A concern with the abnormal has also brought with it a need to attend to the normal. As with any characteristic, a norm is part of a distribution that will range, depending on the characteristic, from a little to a lot or from one extreme to another. In recent years the emergence of developmental psychopathology as a subfield has focused special attention on how we understand abnormality, that is, characteristics that vary far from the norm, both in its properties and the processes that produce it (Sameroff, Lewis, & Miller, 2000). Clinical attention to children troubled by emotional, social, or cognitive problems has raised the question of whether these boys and girls represent extremes of a normal distribution or whether they differ in kind from less troubled youth. If they differ in degree from more average individuals, then their development can be understood within the variability expected in universal developmental processes. But if they differ in kind, then we must ask whether the processes by which they develop also differ in kind.
This distinction has implications for our understanding of risk factors. If there are normal developmental processes that are different from abnormal ones, then we can easily distinguish what is good and bad in development. But if the same processes produce both normal and deviant outcomes, then distinguishing what is good and bad becomes far more difficult. Our project becomes even more difficult when we ask what is a good or bad outcome, especially when the full range of human behavior is taken into account. Has an individual happily communing with nature in rural Alaska in a log cabin on a subsistence diet reached a better outcome than a millionaire industrialist in a luxury townhouse beset by anxiety in the competitive world of urban New York? Or can the creativity resulting from a manic-depressive affective disorder be considered a positive outcome and the unchanging routine of a homemaker providing a supportive frame for a spouse's advancement a negative one?
To frame the question differently, does the absence of risk make for a better life than the presence of risk? An interesting perspective is provided by a study of 200 eminent individuals (Goertzel & Goertzel, 1962). Children who had loving parents, economic advantage, good education, and good health had good outcomes. They became competent professionals, physicians, lawyers, or academics, but they were never eminent. To achieve eminence required what are commonly seen as risk factors: chronic illnesses, emotional turmoil, or economic adversity. Life's challenges are increasingly seen as necessary conditions for life's achievements (Bandura, 1997; Lewis, 1997). The degree to which these challenges enhance or diminish developmental accomplishments is a major concern when attempting to differentiate positive from negative factors in development.
Psychologists' concern with the full range of human performance has led to increasing attention to factors fostering mental health as well as those related to disorder. This new focus has produced a new category of positive influences labeled protective factors. Just as we need to question whether risk factors are different in kind from developmental influences in general, we need to question whether the distinction between risk and protective factors is theoretically meaningful, or merely a convenience for rhetorical purposes.
From the perspective of modern views of development, the desire for a simple model for understanding risk factors is itself a risk factor. Complexity rules the day, whether it be in the connectionist and chaotic models of the cognitive scientists (Thelen & Smith, 1994), the molecular biology of the neuroscientists, or our own understanding of the human psyche. On the one hand, to meet the goals of this chapter we must separate the fundamental from the ephemeral, the theoretically meaningful aspects of risk research from the heuristic. On the other hand, to meet the needs of clinicians and interventionists, we must be able to identify shortcuts where operational and situated definitions of risk can be useful, despite theoretical oversimplification.
Ecological Model
In this chapter I enumerate a variety of factors that affect mental health. Using an ecological model (Bronfenbrenner, 1979), I examine a range of social influences, from the parent practices that have direct influence on the child to community and economic factors that can impinge on the child only through the action of others. Different disciplines have proposed different variables to explain the sources of emotional problems. Economists have focused on poverty and deprivation as the roots of social maladjustment; sociologists have implicated problems in the community and family structure as the variables that promote deviancy; educators blame the school system; and psychologists have focused on processes within the family and among its members as the environmental influences that most profoundly affect successful development. We can accept all of these proposals, but rather than viewing them as competing, we can see them as additive contributors to a positive or negative trajectory through life. It is not any one of these factors that is damaging or facilitating for children, but their accumulation in the life of any one child. Children reared in families with a large number of negative influences will do worse than children in families with few risk factors. Such a view militates against any simplistic proposal that by changing one thing in society, we will change the fate of our children. Competence is the result of a complex interplay between children with a range of personalities, the variations in their families, and their economic, social, and community resources. Only by attending to such complexity will the development of competence be understood and perhaps altered for the better.
The ecological model emphasizes the contributions of multiple environmental variables at multiple levels of social organization to multiple domains of child development. To review even a fraction of the literature on parents, parenting, families, neighborhoods, cultures, and socioeconomic influences would be beyond the scope of this chapter. My strategy is to present some illustrative research on risk factors that identifies the questions to be asked and the complexity of the answers to be expected.
Rochester Longitudinal Study
Despite the nominal interest of developmentalists in the effects of the environment, the analysis and assessment of con- text has fallen more into the domain of sociology than of developmental psychology (Elder, 1984; Kohn & Schooler, 1983; Mayer & Jencks, 1989). The magnitude of a social ecological analysis involving multiple settings and multiple systems (Bronfenbrenner, 1979) has daunted researchers trained primarily to focus on individual behavioral processes. A further daunting factor has been the increasing necessity to use multicausal models to explain developmental phenomena (Sameroff, 1983).
To examine the effects of the social environment on infant mental health, we began an investigation of the development of a group of children from the prenatal period through adolescence-- the Rochester Longitudinal Study (RLS). The infants lived in a socially heterogeneous set of family circumstances, and about half of their mothers had had a mental illness. During the early childhood phase of the RLS (Sameroff, Seifer, & Zax, 1982) we assessed children and their families at birth, and then at 4, 12, 30, and 48 months of age both in the home and in the laboratory. At each preschool age we evaluated two major indicators of infant mental health: the child's cognitive competence and social-emotional behavior. Later we were able to check the progress of these children when they were 13 and 18 years of age.
In our search for family risk factors that would adversely affect the children's development, we considered three major hypotheses: (1) that problem behavior would be related to risks associated with a specific mental illness in a parent, such as schizophrenia; (2) that problem behavior would be attributable to risks associated with parental mental illness in general, especially severity and chronicity of disorder, but no diagnosis in particular; and (3) that problem behavior in the children would be associated with other aspects of the family's condition, especially socioeconomic status (SES). Because many of the families had single parents, we focused our assessments on characteristics of the mother. This approach was taken not because we believed that fathers were unimportant, but because there were too few available for consistent participation in our study.
When we examined the early childhood data, we found little support for the first hypothesis. There was no specific effect of parents' psychiatric diagnosis on the behavior of their offspring during early childhood. The second hypothesis, that mental illness in general would produce substantial effects, was supported more strongly. Global effects of the severity and chronicity of parental psychopathology were ubiquitous throughout the study. Our third hypothesis, that differences in family socio-economic circumstance would produce differences in child behavior, was also strongly supported. The social status effects were apparent throughout the first 4 years of life. Children from the poorest families in our sample exhibited the poorest development. They had poorer obstetrical status, more difficult temperaments at 4 months, less responsivity during home and laboratory observations at 12 months, and less adaptive social and emotional behavior in the home and laboratory at 30 and 48 months of age (Sameroff, Seifer, & Barocas, 1983; Sameroff, Seifer, & Zax, 1982). When the number of differences in child behavior was examined for the diagnostic, mental illness, and SES comparisons, social status differences were the most frequent (Sameroff & Seifer, 1983). At that point in our study we had discovered, on the one hand, that if the only developmental risk for a child was a mother with a mental illness, that child was doing fine. On the other hand, if the child had a mother who was mentally ill and was also poor, uneducated, without social supports, and with many stressful life events, that child was doing poorly. However, we also found that children whose mothers were poor, uneducated, without social supports, and with many stressful life events had worse outcomes, even if the mother did not have a psychiatric diagnosis. In the RLS social circumstance was a more powerful risk factor than any of the parental mental illness measures. Thus we learned the over- riding importance of context in understanding children's development. However, to better understand the role of SES, more differentiated views of environmental influences were needed. We needed to transform measures of parent's educational and occupational achievement that constituted SES scores into variables that would have a conceptually more direct influence on the child. We had to discover what was different about the experience of children raised in different socioeconomic environments.
Single versus Multiple Environmental Risks
Socioeconomic status was our best single variable for predicting children's cognitive competence, and an important, if less powerful, predictor of social-emotional functioning. However, we realized that circumstances of families within the same social class differ quite markedly. We decided to add more psychological content to this primarily economic variable because SES affects many levels of the ecology of children. It affects parenting, parental attitudes and beliefs, family interactions, and many institutions in the surrounding community. From the data available in the RLS we chose a set of variables that were related to economic circumstance but were not the same as SES. The factors we chose ranged from distal variables such as the financial resources of the family, to intermediate variables like the mother's mental health, to proximal variables like the mother's behavioral interactions with the child (Sameroff, Seifer, Barocas, Zax, & Greenspan, 1987). We then tested whether poor cognitive and social-emotional development in our preschool children was related to the risk factors associated with low socioeconomic circumstances. The definitions of the 10 environmental risk variables can be seen in Table 1.1: (1) a history of maternal mental illness, (2) high maternal anxiety, (3) parental perspectives that reflected rigidity in the attitudes, beliefs, and values that mothers had in regard to their child's development, (4) few positive maternal interactions with the child observed during infancy, (5) minimal maternal education, (6) head of household in unskilled occupations, (7) disadvantaged minority status, (8) single parenthood, (9) stressful life events, and (10) large family size. Each of these proposed risk factors has a large literature documenting the potential for deleterious developmental effects (Cicchetti & Cohen, 1995; Sameroff, Lewis, & Miller, 2000; Zeanah, 2000). The 4-year mental health outcome was the Rochester Adaptive Behavior Inventory (Seifer, Sameroff, & Jones, 1981), a parental interview for rating preschool mental health, and the cognitive outcome was the Weschler Preschool and Primary Scale of Intelligence verbal intelligence score.
We found, indeed, that each of these variables was a risk factor. We compared the high risk and low risk groups for each variable separately. For both the cognitive and mental health outcomes, the low risk group had better scores than the high risk group. Although most of the differences were of medium effect size, enough to demonstrate the differences in group comparisons, they were not large enough to indicate which specific individuals with the risk factor would have an adverse outcome. If risk factor analysis is to provide a method for identifying individuals at high risk for mental health problems, a different strategy must be used.
As a way of improving predictive power, Rutter (1979) argued that it was not any particular risk factor but the number of risk factors in a child's background that led to psychiatric disorder. Psychiatric risk for a sample of 10-year-olds he studied rose from 2 percent in families with zero or one risk factors to 20 percent in families with four or more. The six risk factors considered included severe marital distress, low socioeconomic status, large family size or overcrowding, paternal criminality, maternal psychiatric disorder, and admission of the child to foster care. Similarly, Williams, Anderson, McGee, and Silva (1990) related behavioral disorders in 11-year-olds to a cumulative disadvantage score based on number of residence and school changes, single parenthood, low SES, marital separation, young motherhood, low maternal cognitive ability, poor family relations, seeking marriage guidance, and maternal mental health symptoms. Among children with fewer than two disadvantages only 7 percent had behavior problems, whereas among children with eight or more disadvantages, the rate was 40 percent. Even more risk factors were used by Fergusson, Horwood, and Lynsky (1994) in a study of the effects of 39 measures of family problems on the adolescent mental health of a sample of New Zealand children. Again, these researchers found that more risk factors corresponded to more behavioral problems.
Although there were statistically significant effects for the single risk factors in the RLS at the population level, most children with only a single risk factor did not have a major developmental problem. But when we created a multiple risk score that incorporated the total number of risks for each individual family, major differences were found on mental health and intelligence measures between those children with few risks and those with many, with differences of one standard deviation for the mental health measure and two standard deviations for the IQ measure. On the intelligence test, children with no environmental risks scored more than 30 points higher than children with eight or nine risk factors. On average, each risk factor reduced the child's IQ score by 4 points. Four-year- olds in the high risk group (five or more risk factors) were 12.3 times as likely to be rated as having clinical mental health symptoms, and while no preschoolers in the zero-risk group had IQs below 85, 26 percent of those in the high risk group did (see Figure 1.1).
It is clear that the effect of combining the 10 risk variables was to strongly accentuate the differences noted for the individual factors described above. As the number of risk factors increased, competence decreased for preschool children (Sameroff, Seifer, Zax, & Barocas, 1987). Moreover, it appeared that the number of risk factors rather than their kind was most predictive of outcome.
These analyses of the RLS data were attempts to elaborate environmental risk factors by differentiating global measures such as SES into component social and behavioral variables. Although each of our risk factors was found predominantly in lower SES groups, the factors affected child outcomes in all social classes. Moreover, no single variable was either a necessary or a sufficient determinant of a good or bad outcome. Only in families with multiple risk factors was the child's competence placed in jeopardy.
The multiple pressures of environmental context in terms of amount of stress from the environment, the family's resources for coping with that stress, the number of children that must share those resources, and the parents' flexibility in understanding and dealing with their children all play a role in the contemporary development of child intelligence test performance and mental health. However, the sample of infants studied in the RLS was biased toward families in which a parent had a psychiatric diagnosis. Do multiple risks have the same effect on mental health in samples more representative of the general community?
Community Studies of Risk
In order to establish a normative base for the prevalence of risk factors and their association with mental health outcomes, two conditions must be met. There must be study with a large representative sample, and there must be a clearly conceptualized model of risk. Unfortunately, as yet there has not been a large epidemiological study of children's mental health, much less one for infants. Moreover, most studies of the effects of risk on development have not applied an ecological perspective in their conceptualization. As a consequence, ecological analyses are post hoc rather than a priori.
An example of such a study is an analysis of the progress of several thousand young children from kindergarten to third grade using community samples from 30 sites (Peck, Sameroff, Ramey, & Ramey, 1999). From the data collected, 14 risk factors were chosen that tapped ecological levels from parent behavior to neighborhood characteristics. The number of risk factors were summed and a linear relation was found between the multiple environmental risk score and school outcomes of academic achievement and social competence, supporting the findings from the RLS. Although this study used a large sample in multiple sites, the children were not a representative sample of the community, and the risk factors were selected from available data rather than planned in advance.
Another set of data on the effects of multiple environmental risks on child development was provided by a study of adolescents in a group of Philadelphia families (Furstenberg, Cook, Eccles, Elder, & Sameroff, 1999). Mothers, fathers, and offspring were interviewed in close to 500 families in which there was a youth between the ages of 11 and 14. Although not a representative sample, the families varied widely in socioeconomic status, from middle class to families living in poverty, and racial composition included African Americans and non-Hispanic and Hispanic Whites.
An advantage of the Philadelphia project was that a more conceptual approach was taken in the design of the project so that environmental measures were available at a number of ecological levels. For the analyses of environmental risk, variables were grouped and examined within subsystems that affected the adolescent, from those microsystems (Bronfenbrenner, 1979) in which the child was an active participant to those systems more distal to the child where any effect had to be mediated by more proximal variables. A distinction was made between the characteristics of systems that were theoretically independent of the child and those in which the child was an active participant. For example, the family system was subdivided into management processes, such as behavioral control, where it is difficult to determine whether the variable is influenced more by the parent or the child, and structural variables, such as marital status and household density, that were relatively independent of the child.
To approximate an ecological model, six groupings reflecting different relations to the adolescent were built into the design (see Table 1.2). Twenty variables were selected to serve as risk factors, twice as many as in the Rochester study. The intention was to provide multiple factors in each of the six ecological levels. Family process was the first grouping and included variables in the family microsystem that were directly experienced by the child and would fit into a category of parent-child interaction. These included support for autonomy, behavior control, parental involvement, and family climate. The second grouping was parent characteristics, which included the mother's mental health, sense of efficacy, resourcefulness, and level of education. This group included variables that influenced the child but, generally speaking, were less influenced by the child. The third grouping was family structure, which included the parents' marital status, and such socioeconomic indicators as household crowding and receiving welfare payments. The fourth grouping was family management of the community and consisted of variables that characterized the family's management of its relation to the larger community as reflected in the variables institutional involvement, informal networks, social resources, and adjustments to economic pressure. The fifth grouping, peers, included indicators of another microsystem of the child, the extent to which the youth was associated with prosocial and antisocial peers. Community was the sixth grouping, representing the ecological level most distal to the youth and the family. It included a census tract variable reflecting the average income and educational level of the neighborhood the family lived in, a parental report of the number of problems in the neighborhood, and the climate of the adolescent's school.
In the Philadelphia study, in addition to the larger number of ecological variables, we had a wider array of assessments available for interpreting developmental competence than would be available to researchers with infants. The five outcomes used to characterize successful adolescence were summary variables: psychological adjustment, based on parent reports of adolescent mental health; self-competence, based on youth reports of mental health; few problem behaviors, based on youth reports of experiences with drugs, delinquency, and early sexual behavior; activity involvement, based on combined youth and parent reports of participation in sports, religious, extracurricular, and community projects; and academic performance as reflected in grade reports reported by the parent and the adolescent.
Identifying Risks
For risk research the first step is to assess whether each of these variables was indeed a risk factor. In this study two criteria were used to identify each risk factor. The first was that the raw variable was correlated with one of the five outcome variables, and the second was that adolescents in families that had the risk factor did significantly worse on at least one of the outcomes than adolescents in families without that environmental risk. For those variables that met the correlational criteria, we chose a cutoff score to optimize the difference between the outcomes for children with the risk factor and those without. In general, the cutoff separated about 25 percent of the sample as a high risk group from the remaining 75 percent, which were defined as low risk.
The results were that risks existed at every ecological level associated with child outcomes. It was not only the parent or the family that had an influence on child competence, but also the peer group, neighborhood, and community together with their interactions with the family. Some of the variables were risks for each of the five outcomes. These included lack of support for autonomy, a negative family climate, and few prosocial peers. At the other extreme were variables that affected only a few outcomes, such as having parents who lacked education and resourcefulness, single marital status, much adjustment to economic pressure, a lack of informal networks, and low neighborhood census tract socioeconomic status.
Many risk factors have been identified in previous research that used only a single adolescent outcome, such as delinquency (Stouthamer-Loeber et al., 1993). In order to examine the generality of risk factors, there must be multiple outcomes in the study. In the Philadelphia study the pattern of relations between ecological variables chosen as risk factors and adolescent behavior was different for each outcome. On the one hand, academic performance, psycho- logical adjustment, and problem behavior were related to risks at every ecological level. On the other hand, the correlates of self-competence and activity involvement presented two more limited and contrasting pictures. Activity involvement was strongly related to family management of the community and community characteristics, whereas self-competence was unrelated to either. In contrast, family structure played a significant role in adolescent self-competence but not in activity involvement.
The emerging field of prevention science (Coie et al., 1993; Mrazek & Haggerty, 1994) is much concerned with the universality of risk factors. Common findings are that the same risk factors affect multiple outcomes, such as depression, conduct disorder, or substance abuse, and that each disorder has multiple risk factors (Coie, Miller-Johnson, & Bagwell, 2000). In studies of single risks and single outcomes, this fact would be missed. The comprehensiveness and the unity of the developmental process require studies of multiple risks and multiple outcomes to avoid a distorted view of the importance of any single risk.
Multiple Risk Scores
As in the Rochester study, when the differences between high and low risk groups in the Philadelphia study were examined for each individual risk factor, the effect sizes were small or moderate. As in Rochester, this was not the case when multiple risk scores were used. Multiple environmental risk scores were calculated for each adolescent. The resulting range was from a minimum of 0 to a maximum of 13 of a possible 20 risk factors. When the five normalized adolescent outcome scores were plotted against the number of risk factors, a very large decline in outcome was found with increasing risk (see Figure 1.2).
Whether cumulative risk scores meaningfully increase predictive efficiency can be demonstrated by an odds-ratio analysis, comparisons of the odds of having a bad outcome in a high risk versus a low risk environment. For the typical analysis of relative and attributable risk the outcome variable is usually discrete, succumbing to a disease or disorder. For samples of children, there are few discrete negative outcomes. The participants are generally too young to have many pregnancies or arrests, and the rate of academic failure is not particularly high. In the Philadelphia study bad outcomes were artificially created by making cut scores for each of the five outcomes by dichotomizing the sample at the 25th percentile. The 25 percent of adolescents who were doing the worst in terms of mental health, self-competence, problem behavior, activity involvement, or academic performance were considered to have a bad outcome.
The relative risk in the high risk group (eight or more risks) for each of the bad outcomes was substantially higher than in the low risk group (three or fewer risks). The strongest effects were for academic performance, where the relative risk for a bad outcome increased from 7 percent in the low risk group to 45 percent in the high risk group, an odds ratio of 6.7 to 1. The odds ratios for psychological adjustment, problem behavior, self-competence, and activity involvement were 5.7, 4.5, 3.4, and 2.7, respectively. For the important cognitive and social-emotional outcomes of youth there seem to be powerful negative effects of the accumulation of environmental risk factors.
Promotive Factors
As was indicated earlier in this chapter, there is disagreement concerning the definition of (and distinction between) risk and protective factors and vulnerability and resiliency. The concern with preventing developmental failures has often led researchers to overlook the fact that the majority of children in every social class and ethnic group are not failures. They get jobs, have successful social relationships, and raise a new generation of children. The concern with the source of such success has fostered an increasing concern with the development of competence and the identification of protective factors (Garmezy, Masten, & Tellegen, 1984). However, the differentiation between risk and protective factors is far from clear (Seifer & Sameroff, 1987), and there continue to be many theoretical and methodological limitations in their identification (Cicchetti & Garmezy, 1993; Luthar & Zigler, 1991).
Some have argued that protective factors can have meaning only in the face of adversity (Rutter, 1987). To demonstrate that it has protective qualities, a factor should show a greater effect under conditions of higher adversity. An example would be cases of perinatal birth complications, in which high SES can be considered a protective factor. Infants with similar complications such as low birthweight raised in high SES homes have better developmental outcomes than such infants raised in lower SES homes (Sameroff & Chandler, 1975; Wilson, 1985). However, children from higher SES families also have better outcomes in general than children from less advantaged families.
In most studies protective factors are equated with the positive pole of risk factors (Stouthamer-Loeber et al., 1993). Perhaps a better term for the positive end of the risk dimension would be promotive rather than protective factors. This term indicates that having the factor benefits all children, independent of risk status. To test this simplification, we created a set of promotive factors by cutting each of our risk dimensions at the top quartile (Sameroff, Bartko, et al., 1998). So, for example, where a negative family climate had been a risk factor, a positive family climate now became a promotive factor, or where a mother's poor mental health was a risk factor, her good mental health became promotive. We then summed these promotive factors and examined their relation to our five outcomes. The results mirrored our analysis of the effects of multiple risks. There was a similar range of promotive factors, from families with none to families with 15 out of a possible 20, and a similar relation to outcomes, families with many pro- motive factors doing substantially better than those from contexts with few promotive factors. For the youth in the Philadelphia sample there does not seem to be much difference between the influence of risk and promotive variables. The more risk factors, the worse the outcomes; the more promotive factors, the better the outcomes. In short, when taken as part of a constellation of environmental influences on child development, most contextual variables in the parents, the family, the neighborhood, and the culture at large seem to be dimensional, aiding in general child development at one end and inhibiting it at the other.
Some analyses of risk and protective factors have labeled distal variables such as poverty as risk factors and more proximal variables such as family process as protective factors (Furstenberg et al., 1999). The usual finding is that the proximal protective factors can reduce the effect of the distal risk factors. However, these results are simply the effects of relabeling variables and do not contradict the more general finding that the more risk factors, defined as those associated with scores at the bottom of the distribution, and the fewer promotive (protective) factors, those associated with scores at the top of the distribution, the worse the outcomes for children.
Seeking Vulnerability and Resilience
Although most of the family and social factors we have studied seem to have linear effects on child competence (i.e., the more risk, the worse the outcome; the more promotion, the better the outcome), we thought it worthwhile to determine whether some factors would show an interactive protective effect. One approach is to determine whether some environmental factor would buffer the effects of multiple risk. Another is to search for factors in the child that would serve such functions.
On the environmental side, economists have argued that income supplementation programs would solve many family problems, and sociologists have argued that if children were raised in two-parent homes, their developmental success would be much improved. If, indeed, higher income or two-parent families would serve a protective function, we would expect to find that income and family structure variables would be related to developmental success. Such was the case only if one disregarded the effects of all other environmental risks.
Although one would think that these factors should have powerful effects on the fate of children, we did not find such differences when these single variables were put into a broader ecological framework in the Philadelphia study. Differences in effects on child competence disappeared when we controlled for the number of other environmental risk factors in each family. To simplify the analysis, the five youth outcomes were combined into one overall adolescent competence score reflecting general adaptation across personal, academic, and social domains. When the families in the study were divided into high, middle, and low income levels, there was no difference in child competence when the level of multiple risk was controlled (see Figure 1.3). Similarly, when the sample was split into groups of children living in two-parent or single-parent families, no differences were found when the level of multiple risk was controlled (see Figure 1.4). In each case there were no differences in the relation to child competence when we compared groups of children with the same number of risk factors raised in rich or poor families or families with one or two parents.
The results of these analyses support a multiple risk model in which no single environmental factor makes a deterministic contribution to child mental health at the population level. The reason that income and marital status seem to make major differences in child development is not because they are overarching variables in themselves, but because they are strongly associated with a combination of other risk factors. For example, in the Philadelphia sample 39 percent of poor children lived in high risk families with more than seven risk factors, but only 7 percent of affluent children did. Similarly, 29 percent of single-parent families lived in high risk social conditions, but only 15 percent of two-parent families did.
Income or marital status taken alone may have statistically significant effects on adolescent behavior, but these differences are small or nonexistent in comparison with the effects of the accumulation of multiple negative influences that characterize our high risk groups. The overlap in outcomes for youth in high and low income families, and in single-and two-parent families, is substantial for any and all psychological outcomes. There are many successful adults who were raised in poverty and unsuccessful ones who were raised in affluence. There are many healthy and happy adults who come from broken homes, and there are many unhappy ones who were raised by two parents.
Personal Factors
It cannot be denied that personal characteristics are important ingredients in each infant's development. The nature of children, including their temperament and intellect, may contribute to their development; however, it cannot explain their later success or failure, because nature is only one ingredient in the dynamic system that characterizes human development (Lewis, 1997; Sameroff & Chandler, 1975). To give some perspective on the individual contribution to the risk equation, some child characteristics were included in the Philadelphia study. Personal variables can be divided into demographic and behavioral domains. For these analyses, sex and race were the demographic variables and efficacy was the behavioral one. The relation between risk scores and outcomes for separate groups of boys and girls and African Americans and Whites were examined and no differences were found. When the relation between our summary competence measure and risk factors was compared for sex and racial groups, the curves were essentially overlapping.
Like the SES variable on the environmental side, race and sex are not behavioral variables. Therefore it would be of greater interest to investigate the influence of variables with psychological content. A personality variable that is given great importance in discussions of successful development is resourcefulness or efficacy (Bandura, 1997). Is it possible that despite social adversity those children with high levels of "human capital" (Coleman, 1988) are able to overcome minimal resources at home and in the community to reach levels of achievement comparable to those of children from more highly advantaged social strata?
In the Philadelphia study the construct of resourcefulness was measured with a set of questions asked of the parent and child about the youth's capacity to solve problems, overcome difficulties, and bounce back from setbacks. The summary score for adolescent competence was compared for groups of youth divided into high and low efficacy groups and, high efficacious youth on average were more competent than those with low resourcefulness, indicating that personal resourcefulness did seem to pay off.
But what happens to this effect when environmental adversity is taken into account? When high and low efficacy children were matched for the number of environmental risk factors, the group difference did not disappear. However, the difference in general competence between youth in the high and low environmental risk conditions was far greater than that between high resourceful and low resourceful groups of youth (see Figure 1.5). High efficacious adolescents in high risk conditions did worse than low efficacious youth in low risk conditions. It may not be a surprise to learn that the less effective offspring of advantaged families may have a much easier life course than more resourceful offspring of multirisk families.
Resilience during Infancy
The foregoing description of the Philadelphia study was devoted to laying out the issues for an appreciation of some of the intricacies of risk research, both in terms of methodology and interpretation. However, the purpose of this volume is to illuminate infant mental health, not adolescent mental health. Unfortunately, few studies of infants have directly addressed the complexities of risk research from an ecological perspective, especially when the issue of resiliency is raised. Infancy is a starting point, and much of our concern is with the consequences of what happens during that period of life. The study of consequences requires a longitudinal design wherein one can examine the relation of infant life to what follows.
Although the Philadelphia study was helpful in laying out issues, it fell short in reaching many conclusions about risk because it is not a longitudinal study. Interpreting intercorrelations of risk and outcome measures at a single point of time is difficult because the causal direction is unclear. From an environmental perspective it is easy to declare that parenting behavior causes child competence, but from a personality perspective, it may be that competent children elicit warmth and permissive- ness from their parents while emotionally disturbed children elicit negativity and authoritarian rearing styles. To disentangle the direction of effects, one needs to examine the relation between child and context across time. Although the Rochester study had a smaller and more select sample than Philadelphia, it did have a series of developmental assessments beginning in infancy that permitted a longitudinal view of the contribution of individual factors to developmental success.
To test the relative contribution of infant behavioral variables to the child's later performance, in the RLS we created another multiple risk score using 13 infant items related to competent behavior. These were based on subscores of the infant's perinatal physical condition from the Research Obstetrical Scale (Sameroff, Seifer, & Zax, 1982), cluster scores for newborn interactive behaviors, motor behaviors, state control, and stress from the Brazelton Neonatal Behavioral Assessment Scales (Brazelton, 1973), the Mental Development Index at 4 and 12 months and the Psychomotor Development Index at 4 and 12 months from the Bayley Scales of Infant Development, and level of activity and crying at 4 and 12 months of age from our home observations of infant behavior.
Based on the infant multiple risk score, we divided the sample into low and high competence groups of infants and examined as outcomes their 4-year IQ and mental health functioning scores (see Figure 1.6). There was no relation between infant competence scores and 4-year IQ. Similarly, there was no relation between infant multiple risk scores and the global rating of 4-year mental health. High competent infants in high risk environments did worse as 4-year-olds than low competent infants in low risk environments. As in Philadelphia, individual characteristics were not able to overcome the effects of environmental adversity. If one wants to predict the developmental course for infants, attention to the accumulation of environmental risk factors would be the best strategy.
What these analyses of personal factors tell us is that income level and marital status on the family side, and sex, race, and efficacy on the personal side, taken alone may have statistically significant effects on child behavior, but these differences pale in comparison with the accumulation of multiple negative influences that characterize high environmental risk groups. The overlap in outcomes for low income versus high income families, families with one or two parents, boys versus girls, Blacks versus Whites, and high resourceful and low resourceful children is substantial for most psychological outcomes when only single risks are considered, but the difference is far greater in comparisons of groups of children reared in conditions of high versus low multiple risk where the combinations of risk factors are assessed. The important implication is that a focus on single characteristics of individuals or families can explain only small proportions of variance in behavioral development. To truly appreciate the determinants of mental health requires attention being paid to the broad constellation of ecological factors in which these individuals and families are embedded.
Continuity of Environmental Risk
The Rochester and Philadelphia studies have revealed major consequences for children living in multiproblem families. What are the long-term consequences of these early adverse circumstances? Will later conditions alter the course for such children, or will early experiences lock children into pathways of deviance? To answer this question we must return to a consideration of data from the adolescent phase of the Rochester Longitudinal Study, in which we completed further assessments of the sample (Baldwin et al. 1993; Sameroff, Seifer, Baldwin, & Baldwin, 1993).
New multiple environmental risk scores were calculated for each family based on their situation 9 years later. To our surprise, very few families showed major shifts in the number of risk factors across the intervening periods. Between ages 4 and 13 years the factor that showed the most improvement was maternal education: The number of mothers without a high school diploma or equivalent decreased from 33 to 22 percent of the sample. The risk factor that increased the most was single parenthood, with the number of children being raised by their mothers alone increasing from 24 to 41 percent. In the main, however, there was little change in the environ- ments of the children in our sample.
The typical statistic reported in longitudinal research is the correlation between early and later performance of the children. In the RLS intelligence at 4 years correlated .72 with intelligence at 13 years and mental health correlated .41 between the two ages. The usual interpretation of such statistically significant relations is that there are continuities of competence or incompetence in the child. Such a conclusion cannot be challenged if the only assessments in the study are of the children. In the RLS we examined and were able to correlate environmental characteristics as well as child ones across time. The high correlation of .77 that we found between environmental risk scores at the two ages was as great as or greater than any continuity within the child. Whatever the child's ability for achieving higher levels of competence, it was severely undermined by the continuing paucity of environmental support in high risk contexts. Conversely, it was fostered in low risk contexts. Whatever the capabilities provided to the child by personal factors, the presence or lack of risk factors in the environment acted to limit or expand further opportunities for development.
Risk Factors and Development
In this chapter I have attempted to explore issues relevant to risk research with infants. A central theme was the unity of development. Attempts to categorize risk, protection, or resiliency as a characteristic of only some children were contrasted with a more general model of development in which children and families differ in degree rather than kind. Most developmental influences are bipolar, enhancing development on one end and constraining it on the other. Emotional warmth is good and hostility is bad for all children. Growing up is an exercise in resilience for all children. Beginning with the challenges provided by the infant's own body, the efforts necessary to roll over and to walk, and continuing through the environmental challenges reflected in maintaining social relations or the challenges in the educational curricula of the school, children constantly are faced with problems to be solved.
Given this universality, attention to risk factors generally means making arbitrary cut points on continuous distributions, selecting the top or bottom quartile. Even such a seemingly distinct category as child abuse can be considered an extreme of a control dimension of parent behavior. Single parenthood does not necessarily mean that one adult is raising the child, given the prevalence of extended families and child care facilities. There is a rationale for making the universal unique by identifying categorical risk factors in order to select a population for special treatment. Given the restricted resources for intervening to support infant mental health, there is a need to be able to identify those families most in need, that is, those with children who are most likely to have mental health problems. Here risk factor research provides the entry point. But such heuristic strategies for identifying clients in need of service should not be confused with scientific understanding of the developmental process. The scientific questions are captured in the issues raised at the beginning of this chapter.
The first issue was that the identification of risk factors is an exercise in estimating probabilities, not finding causes. The myriad factors involved in the developmental process militate against any single variable being either necessary or sufficient for producing an outcome of mental health or disturbance. At the population level each social factor contributes to the mental health of developing children, but for each child there will be a unique subset of factors working their influence. Herein lies the requirement for the skill of the clinician to determine the unique forces in every unique situation. The second issue was the recognition that child development has multiple contributors at multiple levels of the child's ecology. Although it would appear that the source of continuity in child mental health should be in the child, continuities in the child's environment may be of even greater importance. The focus in research on infant mental health is generally on the parent-child relationship, but parents frequently are affected by family dynamics independent of the child, as well as the stresses and supports provided by the broader community, and the amount of resources associated with economic status. All of these will affe ct their ability to parent. The ecological model offers an ex- panded view of the range of factors that will have to be considered. Separating the universal from the unique and what is good for all babies from what is necessary for only some babies will continue to be the goal of our scientific and clinical interests.
References
Baldwin, A. L., Baldwin, C. P., Kasser, T., Zax, M., Sameroff, A., & Seifer, R. (1993). Contextual risk and resiliency during late adolescence. Development and Psychopathology, 5, 741- 761.
Bandura, A. (1997). Self-efficacy: The exercise of control. New York: W. H. Freeman. Brazelton, T. B. (1973). Neonatal Behavioral Assessment Scale. London: Heinemann. Brazelton, T. B., & Cramer, B. G. (1990). The earliest relationship. New York: Addison-Wesley.
Bronfenbrenner, U. (1979). The ecology of human development. Cambridge, MA: Harvard University Press.
Children's Defense Fund (1992). Child poverty hits 25-year high, growing by nearly 1 million children in 1991. CDF Reports, 13(12), 2.
Children's Defense Fund (1995). The state of America's children: 1995. Washington, DC: Children's Defense Fund.
Cicchetti, D., & Cohen, D. (Eds.). (1995). Developmental Psychopathology: Vol. 2. Risk, disorder, and adaptation. New York: Wiley.
Cicchetti, D., & Garmezy, N. (1993). Prospects and promise in the study of resilience. Development and Psychopathology, 5, 497- 502.
Clark, R. (1985). The Parent-Child Early Relational Assessment. Manual and instrument. Madison, WI: University of Wiscon- sin Department of Psychiatry.
Coie, J. D., Miller-Johnson, S., & Bagwell, C. (2000). Prevention science. In A. Sam- eroff, M. Lewis, & S. Miller (Eds.), Handbook of developmental psychopathology. New York: Plenum.
Coie, J. D., Watt, N. F., West, S. G., Hawkins, J. D., Asarnow, J. R., Markman, J. H., Ramey, S. L., Shure, M. B., & Long, B. (1993). The science of prevention: A conceptual framework and some directions for a national research program. American Psychologist, 48, 1013-1022.
Coleman, J. (1988). Social capital in the creation of human capital. American Journal of Sociology, 94, S95- S120.
Cooper, B., & Morgan, H. G. (1973). Epidemiological psychiatry. Springfield, IL: C. C. Thomas.
Costello, E. J., & Angold, A. (1995). Developmental epidemiology. In D. Cicchetti & D. Cohen (Eds.), Developmental psychopathology: Vol. 1. Theory and methods (pp. 23- 56). New York: Wiley.
Costello, E. J., & Angold, A. (2000). Developmental epidemiology: A framework for developmental psychopathology. In A. Sameroff, M. Lewis, & S. Miller (Eds.), Handbook of developmental psychopathology. New York: Plenum.
Damon, W., & Eisenberg, N. (Eds.). (1998). Handbook of child psychology: Vol. 3. Social, emotional, and personality development (5th ed.). New York: Wiley.
Dawber, T. R. (1980). The Framingham Study: The epidemiology of coronary heart disease. Cambridge, MA: Harvard University Press.
Dryfoos, J. G. (1991). Adolescents at risk: A summation of work in the field: Programs and policies. Journal of Adolescent Health, 12, 630- 637.
Elder, G. H., Jr. (1984). Families, kin, and the life course: A sociological perspective. In R. D. Parke (Ed.), Review of Child Development Research: The Family (Vol. 7) (pp. 80- 136). Chicago: University of Chicago Press.
Engel, G. L. (1977). The need for a new medical model: A challenge for biomedicine. Science, 196, 129- 136.
Fergusson, D. M., Horwood, L. J., & Lynsky, M. T. (1994). The childhoods of multiple problem adolescents: A 15-year longitudinal study. Journal of Child Psychology and Psychiatry, 35, 1123- 1140.
Ford, D. H., & Lerner, R. M. (1992). Developmental systems theory: An integrative approach. Newbury Park, CA: Sage.
Furstenberg, F. F., Jr., Cook, T. D., Eccles, J., Elder, G. H., Jr., & Sameroff, A. (1999). Managing to make it: Urban families and adolescent success. Chicago: University of Chicago Press.
Garmezy, N., Masten, A. S., & Tellegan, A. (1984). The study of stress and competence in children: A building block of developmental psychopathology. Child Development, 55, 97- 111.
Garmezy, N. (1974). The study of competence in children at risk for severe psychopathology. In A. Koupernik (Ed.), The child in his family-- Children at psychiatric risk (Vol. 3, pp. 77- 97). New York: Wiley.
Goertzel, V., & Goertzel, M. G. (1962). Cradles of eminence. Boston: Little, Brown.
Greenspan, S., & Weider, S. (2000). The assessment and diagnosis of infant disorders: Developmental level, individual differences, and relationship based interactions. In J. D. Osofsky & H. E. Fitzgerald (Eds.), The WAIMH Handbook of Infant Mental Health: Vol. 2. New York: Wiley.
Kohn, M., & Schooler, C. (1983). Work and personality: An inquiry into the impact of social stratification. Norwood, NJ: Ablex.
Lewis, M. (1997). Altering fate: Why the past does not predict the future. New York: Guilford.
Lieberman, A. (2000). Parenting toddlers: Developmental and clinical considerations. In J. D. Osofsky & H. E. Fitzgerald (Eds.), The WAIMH Handbook of Infant Mental Health: Vol. 3. New York: Wiley.
Luthar, S. S. (1993). Annotation: Methodological and conceptual issues in research on childhood resilience. Journal of Child Psychology and Psychiatry, 34, 441- 453.
Luthar, S. S., & Zigler, E. (1991). Vulnerability and competence: A review of research on resilience in childhood. American Journal of Orthopsychiatry, 61, 6- 22.
Masten, A. S., & Coatsworth, J. D. (1995). Competence, resilience, and psychopathology. In D. Cicchetti & D. Cohen (Eds.), Developmental psychopathology: Vol. 2: Risk, disorder, and adaptation (pp. 715- 752). New York: Wiley.
Masten, A. S., Best, K. M., & Garmezy, N. (1990). Resilience and development: Contributions from the study of children who overcome adversity. Development and Psychopathology, 2, 425- 444.
Mayer, S. E., & Jencks, C. (1989). Growing up in poor neighborhoods: How much does it matter? Science, 243, 1441- 1445.
McDonough, S. C. (1995). Promoting positive early parent-infant relationships through interaction guidance. Child and Adolescent Clinics of North America, 4(3), 661- 672.
McLanahan, S. S., Astone, N. M., & Marks, N. F. (1991). The role of mother-only families in reproducing poverty. In A. C. Huston (Ed.), Children in Poverty (pp. 51- 78). Cambridge, UK: Cambridge University Press.
Mrazek, P. G., & Haggerty, R. J. (Eds.). (1994). Reducing risks for mental disorders: Frontiers for preventive intervention programs. Washington, DC: National Academy Press.
Pawl, J. (2000). Training mental health and other professionals in infant mental health: Conversations with trainees. In J. D. Osofsky & H. E. Fitzgerald (Eds.), The WAIMH Handbook of Infant Mental Health: Vol. 2. New York: Wiley.
Peck, S., Sameroff, A., Ramey, S., & Ramey, C. (1999, April). Transition into school: Ecological risks for adaptation and achievement in a national sample. Paper presented at the biennial meeting of the Society for Research and Development, Albuquerque, NM.
Rutter, M. (1979). Protective factors in children's responses to stress and disadvantage. In M. W. Kent & J. E. Rolf (Eds.), Primary prevention of psychopathology: Vol. 3. Social competence in children (pp. 49- 74). Hanover, NH: University Press of New England.
Rutter, M. (1987). Psychosocial resilience and protective mechanisms. American Journal of Orthopsychiatry, 57, 316- 331.
Sameroff, A., Lewis, M., & Miller, S. (Eds.). (2000). Handbook of developmental psychopathology. New York: Plenum.
Sameroff, A. J. (1983). Developmental systems: Contexts and evolution. In P. H. Mussen (Series Ed.) & W. Kessen (Vol. Ed.), Handbook of child psychology: Vol. 1. History, theories, and methods (pp. 237- 294). New York: Wiley.
Sameroff, A. J. (1993). Models of development and developmental risk. In C. H. Zeanah (Ed.), Handbook of Infant Mental Health (pp. 3- 13). New York: Guilford.
Sameroff, A. J. (1995). General systems theories and developmental psychopathology. In D. Cicchetti & D. Cohen (Eds.), Developmental Psychopathology: Vol. 1. Theory and Methods (pp. 659-695). New York: Wiley.
Sameroff, A. J., Bartko, W. T., Baldwin, A., Baldwin, C., & Seifer, R. (1998). Family and social influences on the development of child competence. In M. Lewis & C. Feiring (Eds.), Families, risk, and competence (pp. 161- 186). Mahwah, NJ: Erlbaum.
Sameroff, A. J., & Chandler, M. J. (1975). Reproductive risk and the continuum of caretaking casualty. In F. D. Horowitz, M. Hetherington, S. Scarr-Salapatek, & G. Siegel (Eds.), Review of child development research (Vol. 4, pp. 187- 244). Chicago: University of Chicago.
Sameroff, A. J., & Fiese, B. H. (1999). Transactional regulation: The developmental ecology of early intervention. In S. J. Meisels & J. P. Shonkoff (Eds.), Early intervention: A handbook of theory, practice, and analysis. New York: Cambridge University Press.
Sameroff, A. J., & Seifer, R. (1983). Familial risk and child competence. Child Development, 54, 1254- 1268.
Sameroff, A. J., Seifer, R., Baldwin, A., & Baldwin, C. (1993). Stability of intelligence from preschool to adolescence: The influence of social and family risk factors. Child Development, 64, 80- 97.
Sameroff, A. J., Seifer, R., & Barocas, R. (1983). Developmental impact of parental psychopathology: Diagnosis, severity or social status effects? Infant Mental Health Journal, 4, 236- 249.
Sameroff, A. J., Seifer, R., Barocas, R., Zax, M., & Greenspan, S. (1987). Intelligence quotient scores of 4-year-old children: Social environmental risk factors. Pediatrics, 79, 343- 350.
Sameroff, A. J., Seifer, R., & Zax, M. (1982). Early development of children at risk for emotional disorder. Monographs of the Society for Research in Child Development, 47 (Serial No. 199).
Sameroff, A. J., Seifer, R., Zax, M. , & Barocas, R. (1987). Early indicators of developmental risk: The Rochester Longitudinal Study. Schizophrenia Bulletin, 13, 383-393.
Sameroff, A. J., Seifer, R., Zax, M., Barocas, R., & Greenspan, S. (1992). Socialenvironmental risk and early competence. In C. Chiland & J. G. Young (Eds.), New approaches to mental health from birth to adolescence (pp. 269- 280). New Haven: Yale University Press.
Seifer, R., & Sameroff, A. J. (1987). Multiple determinants of risk and vulnerability. In E. J. Anthony & B. J. Cohler (Eds.), The invulnerable child (pp. 51- 69). New York: Guilford.
Seifer, R., Sameroff, A. J., Baldwin, C. P., & Baldwin, A. (1992). Child and family factors that ameliorate risk between 4 and 13 years of age. Journal of the American Academy of Child Psychiatry, 31, 893- 903.
Seifer, R., Sameroff, A. J., & Jones, F. (1981). Adaptive behavior in young children of emotionally disturbed women. Journal of Applied Developmental Psychology, 1, 251- 276.
Stouthamer-Loeber, M., Loeber, R., Farrington, D. P., Zhang, Q., van Kam- men, W., & Maguin, E. (1993). The double edge of protective and risk factors for delinquency: Interrelations and developmental patterns. Development and Psychopathology, 5, 683- 701.
Thelen, E., & Smith, L. B. (1994). A dynamic systems approach to the development of cognition and action. Cambridge, MA: MIT Press.
U.S. Department of Health and Human Services. (1990). Child health USA '90. Washington, DC: U.S. Government Printing Office.
U.S. Department of Health and Human Services. (1993). Child health USA '92. Washington, DC: U.S. Government Printing Office.
Williams, S., Anderson, J., McGee, R., & Silva, P. A. (1990). Risk factors for behavioral and emotional disorder in preadolescent children. Journal of the American Academy of Child and Adolescent Psychiatry, 29, 413- 419.
Wilson, R. S. (1985). Risk and resilience in early mental development. Developmental Psychology, 21, 795- 805.
Zeanah, C. H. (2000). Handbook of infant mental health. New York: Guilford.
Zill, N., & Schoenborn, C. A. (1990). Developmental, learning, and emotional problems: Health of our nation's children, United States, 1988. Hyattsville, Md.: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics National Center for Health Statistics.
Zimmerman, M. A., & Arunkumar, R. (1994). Resiliency research: Implications for schools and policy. Social Policy Report: Society for Research in Child Development, 8(4), 1-18.
Table of Contents
Infant Mental Health: Origins and Emergence of an InterdisciplinaryField (H. Fitzgerald & L. Barton).Cross-Cultural Perspectives on Infant Mental Health: Japan (K.Okonogi & Y. Hamada).
Longitudinal Aspects of Early Parent-Infant Interactions andContacts with Mental Health Agencies (P. de Chateau).
Cross-Cultural Perspectives on Infant Mental Health: Development ofInfant Mental Health in Australia (B. Warren).
Cross-Cultural Perspectives on Infant Mental Health: A French Viewon the History of Infant and Child Psychiatry (S. Lebovici).
The Sociocultural Context of American Indian Infant Mental Health(M. Heart & P. Spicer).
The Sociocultural Context of Infant Mental Health in the People'sRepublic of China (R. Fong).
Infant Mental Health in Brazil (S. Celia).
Infant Mental Health in Scandinavia (P. Mothander).
New Attitudes: Infant Care Facilities in Saint Petersburg, Russia(R. Muhamedrahimov).
Cross-Cultural Perspectives on Infant Mental Health: Germany,Austria, and Switzerland (P. Scheer & M. Dunitz-Scheer).
Work Projects Toward Infant Mental Health in a Child and AdolescentPsychiatric Clinic (M. Aguerre).
Indexes.