Risk in infancy: Origins and implications

Risk in Infancy Origins and Implications This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Claire B. Kopp and Sandra R. Kaler University of California, Los Angeles ABSTRACT." This article focuses on biological risks that can adversely influence development during infancy and later. In the first part of the article, the origins of risks and their potential consequences are discussed relative to prepregnancy, prenatal, perinatal, and postnatal periods. Epidemiological data are presented. The second part of the article addresses issues pertaining to assessment of infant development and interventions provided for infants whose development may be in jeopardy. Finally, goals for prevention are highlighted and the need for a nationally derived data base on risk and outcomes is emphasized. Infancy, the foundation period of our species, is no more and no less important than other phases of life. Still, infancy marks the beginning of our contacts with the outside world, and for this reason it has been accorded special status by some philosophers, scientists, and lay people. There are others, we among them, who prefer to think of infancy as a way-point in childhood's journey. Important, developmentally related events occur before birth, and many more will occur long after. When genetic heritage and prenatal life are favorable, the infant's roots are securely anchored and sound development should occur. If physical or psychological adversities arise later on, there is potential for resiliency given that rearing conditions are reasonably supportive. Alternatively, unfavorable genetic or prenatal factors set the stage for vulnerabilities. Sometimes an adverse condition is so profound that development is irreversibly impaired. In other instances, risk factors may disrupt processes of growth but not conelusively alter them. In this case development is less predictable and depends on both the nature of early risks and the child's ensuing life experiences. Often supportive caregiving can buffer the child and facilitate development, whereas rearing that is neglectful, uncaring, or abusive compounds earlier vulnerabilities. Exposed to multiple and continuing liabilities, a few children manage to adapt, many more marginally accommodate to educational and social demands, and still others ultimately fail and deteriorate mentally, behaviorally, and socially. Risks that compromise development are diverse and complex. They include those that are (a) biological in origin; (b) from the environment, primarily adverse rearing conditions; and (c) from a combination of the two. Our focus is on biological risk, and where appropriate, combined risks are mentioned as well. Topics discussed 224 include the origins of risks and their effects on development, epidemiology, assessment of status in infancy, and interventions for infants whose development may be problematic. Origins and Implications of Risk Risk Factors, Development, and Timing The phrase biological risk encompasses different kinds of adverse conditions; these are discussed relative to prenatal, perinatal, and postnatal life along with implications for development. Biological risks stem from genetic conditions (e.g., mutant recessive genetic material) as well as from exposure to harmful nonsocial environmental factors. Viral infections or use of drugs during pregnancy are examples of the latter. Biological risks pose serious threats to the developing organism and can result in death or outcomes that include physical malformations, growth retardation, neurological (e.g., cerebral palsy) and physiological problems, mental retardation, sensory disorders (e.g., blindness or deafness), learning disabilities and other educational difficulties, psychiatric disorders, and social deviancy. Some professionals also consider preterm birth and low birth weight as negative outcomes. Biological risks may show developmental effects early in life or later, may lead to severe impairments or subtle dysfunctions, and may influence all aspects of human behavior or only one or two. Because development is a lifelong process, problems observed in infancy are n o t necessarily those manifested in subsequent years. On the other hand, particularly during early life, development, whether proceeding as expected or not, is associated with increasing maturation of brain structures and functions. The immature brain is adaptive to new experiences and, equally important, often has potential for recovery should certain kinds of brain insult occur. Developmental risk, a term frequently used by professionals, is reserved for immature organisms and refers to a statistical probability that ongoing development •will be compromised in some way. Statistical probabilities are expectations about the percentage of children that will have developmental problems. Timing refers to the particular period in early development when a risk event occurs, that is, before pregnancy, at conception, during prenatal life, at the perinatal period, or after birth (postnatal life). Consideration of February 1989 • American Psychologist Copyright 1989 by the American Psychological Association, Inc. 0003-066X/89/$00.75 Vet. 44, No. 2, 224-230 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. timing leads to inferences about the effects a risk event can have when the developing organism is at age x as opposed to one that is at age y (Freeman, 1985; Hagberg, 1978; Kopp, 1983). In the remainder of this section, risks, developmental outcomes, and timing are interrelated, and illustrative examples are provided. We use chronological order beginning with risks that occur before pregnancy and end with those that arise after birth. Prepregnancy risks. Data unequivocally show that deleterious factors occurring years before a pregnancy may have developmental implications for the conceptus, throughout infancy and beyond. Prepregnancy risks inelude maternal chronic illness, history of drug use, inadequate nutrition during childhood and adolescence, genetic vulnerability within the family, previous numerous closely spaced pregnancies, and female age (extremes) at time of conception. Among the economically disadvantaged, prepregnancy risks may run through generations. Poor nutrition and lack of health care during childhood are often compounded by adverse pregnancy conditions (e.g., lack of prenatal care, drug use, or exposure to infections). Combinations such as this portend constitutionally vulnerable infants who become vulnerable children and then highrisk childbearing adults. All too often, health and developmental disorders such as poor school achievement and social deviancy arise. Unfortunately, the extent of these problems is unknown. Of course, conditions other than poverty give rise to prepregnancy risks. In our culture, social and economic forces have moved increasing numbers of women into the job market. Concomitantly, these women are postponing first pregnancies beyond the optimal childbearing period (20 to 30 years). The largest growth in fertility rates has occurred among women in their early 30s, with first births among 30- to 34-year-olds almost doubling between 1970 and 1979 (Dorfman, 1986). A substantial number of these births were to educated women who were not in economic distress. This trend in later pregnancies will likely continue. Concerns have heightened about child outcomes. In addition to the well-established association between maternal age and chromosomal abnormalities (e.g., the risk for Down syndrome for women over 40 years is manyfold greater than for women in their 20s), there has been worry about other forms of nonoptimal development. Earlier studies suggested that maternal age was a risk factor in and of itself; however, conditions that coexist with increasing age such as hypertension or diabetes may explain previous findings. Recently completed, well-designed studies are optimistic. In a comparison of 511 pregnancies Support for writing this article was provided in part by National Science Foundation Grant BNS 87-10028 and Department of Education Grant G0086 35232 to Claire B. Kopp. Correspondence concerning this article should be addressed to Claire B. Kopp, Department of Psychology, Franz Hall, University of California, Los Angeles, CA 90024. February 1989 • American Psychologist of women over 40 with 26,000+ pregnancies of women between 20-30 years, controlling for maternal weight, cigarette smoking, and parity, there were no higher risks for older women (Spellacy, Miller, & Winegar, 1986). Undeniably, entering pregnancy with good overall health status and having access to careful prenatal monitoring are critically important. Prenatal period. In the aggregate, risks that arise around the time of conception or during the first weeks of pregnancy (when structures are being formed) are associated with more serious developmental ramifications than those that arise earlier or later. Impairments tend to be both major and irreversible and include malformations of head, face, limbs, and organs; sensory and neurological disorders; and severe mental retardation (IQ of 50 or less). Brain disorders are often obvious at birth; later, developmental delays and intellectual problems become apparent. The seriousness of early prenatal risks is underscored by the fact that few affected individuals escape long-term difficulties, and many require lifelong assistance. Adverse conditions and events related to prenatal risks include mutant recessive genes and chromosomal disorders, environmental agents (e.g., harmful agents such as drugs, chemicals, and toxins), maternal viral infections such as rubella, and chronic and severe maternal health conditions that continue into pregnancy (e.g., major kidney disease and cardiac anomalies). Multiple biological factors and adverse environmental events are also likely to be involved. The developmental picture for later pregnancy risks is less clear. It is known that some children develop normally, a few will have major handicapping conditions, and others will show various kinds of less severe intellectual and social impairments. For some indeterminate number, there will be a variety of dysfunctions involving physiological and psychological systems (Vorhees, 1986). The relatively new research discipline called behavioral teratology focuses on behavioral effects subsequent to stress or damage to the developing brain. Intellectual and behavioral dysfunctions that occur in the absence of malformations can be as important as those that occur with them. Researchers are examining behavioral effects of alcohol, chemicals, pollutants, and other agents using a variety of animal models and human studies. There is evidence that methylmercury, alcohol, lead, and high doses of ionizing radiation are disruptive to children's development, and there is suspicion that nicotine, narcotics, some pollutants, hormones, and lower doses of ionizing radiation may affect development (Vorhees & Mollinow, 1987). The challenge for behavioral teratology is substantial because long-term effects as well as transient ones must be demonstrated while controlling for variables other than the risk agents, and it is important that the reason underlying variability in outcome effects be ascertained. A case in point is alcohol: Not all children born to alcoholic mothers show fetal alcohol syndrome or even less severe symptomatology. Other research approaches yield promising findings. 225 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Drawing on suggestive data linking pregnancy illnesses and schizophrenia, Mednick, Machon, Huttunen, and Bonett (1988) examined the incidence of schizophrenia in a 1957 Finnish birth cohort in which pregnant women were exposed to a widespread severe type A2 influenza epidemic. Using a variety of procedures, the authors were able to ascertain that second-trimester viral exposure was significantly linked to a greater percentage of offspring who were later admitted to psychiatric hospitals with a diagnosis of Schizophrenia. Expressing caution about their findings, the authors nonetheless argue for an association between mid-pregnancy risk and psychiatric vulnerability. Moreover, they suggest that second-trimester risk may hold implications for brain functioning in that risks may interfere with the development of late appearing cortical neurons, destruction of existing cells, or disruption of cell migration. The implications of this research are profound for defining long-term effects, particularly those that pertain to emotional dysfunction rather than intellectual impairment. Perinatal period. This third time frame overlaps prenatal and postnatal life and extends from the seventh prenatal month to the end of the first postnatal month. Illustrating the importance ascribed to this period, perinatal mortality statistics are used as one measure of the health and well-being of nations. Perinatal trends in the United States show a 1985 mortality rate of 10.6 deaths per 1,000 live births, contrasted with the 7 to 9 per 1,000 rates found in Scandinavian and some European countries (Miller, 1987; National Center for Health Statistics [NCHS], 1987). Part of the United States's high rate is due to higher mortality for certain minority groups (for Blacks the 1984 rate was 18.4, whereas for Whites it was 9.4 per 1,000 [NCHS, 1987]). Poverty, teenage pregnancies, and lack of prenatal care are implicated (Hughes, Johnson, Rosenbaum, Butler, & Simons, 1988). What are perinatal risks? Historically, they have been defined as any symptom or condition that may stress the fetus or newborn. Thus preterm birth, congenital malformations, undergrowth in the newborn period whether observed in full-term or preterm babies, respiratory distress, newborn seizures, asphyxia, birth trauma, and newborn infections have been labeled perinatal risks. Although categorizing these risks as perinatal has been effective for deriving general mortality statistics, the category is less useful for estimating developmental outcomes. More recently, attempts are being made to separate conditions that had their origins in prenatal life from those that originate solely within the perinatal period. Profound growth retardation observed in either preterm or full-term newborns is an example of a risk that began in the prenatal period, whereas an infection acquired from the mother during the birth process exemplifies a de novo perinatal risk. These distinctions encourage more precise diagnosis and treatment. Moreover, evidence also suggests that early or late pregnancy risks predispose the fetus to additional perinatal stresses (Freeman, 1985), and a combination of prenatal-perinatal risks has more adverse 226 developmental consequences than de novo perinatal risks (Rosen, 1985). Even if definitional problems did not exist, perinatal risks pose difficult challenges for understanding and determining outcomes. For one, nationally derived statistics for nondeath outcomes are virtually nonexistent. Developmental estimations are generated by different research and clinical teams. Moreover, documented outcomes for infancy and childhood vary widely and are a function of quality of medical care in the newborn period, the risk condition that is studied, criteria selected as measures of outcome, and social class and family rearing conditions. Second, despite evidence that the environment in which the infant lives is a continuing influence on development, the interaction of biological risk and rearing milieu is largely unspecified. It is clear that perinatal risks combined with nonoptimal rearing conditions portend poor developmental outcomes, whereas good rearing ameliorates the risk (Sameroff& Chandler, 1975; Werner & Smith, 1982). Because nonoptimal rearing is relatively more common among socially and economically distressed individuals than among affluent individuals, social economic status (SES) is a frequently used index of rearing conditions. However, SES is not a pure measure. Nominally it is a descriptor of education and occupational status, but SES actually subsumes considerable heterogeneity within middle and lower class categories in terms of caregiver knowledge and attributes, family atmosphere, and home density. Furthermore, accumulating evidence suggests more extensive variation exists in today's society among lower class families than in our recent past. This appears to result from increased levels of joblessness, poverty, homelessness, drug use, and caregiver emotional disturbance. Until more systematic documentation is made of child characteristics, specific rearing conditions, and home atmosphere factors it will be difficult to further specify the influences of combined biological and social risks on development.I No discussion ofperinatal risks is complete without calling attention to low birth weight; this is one of the most highly visible perinatal risks. Low birth weight is the designation given to infants born weighing less than 5 lb; most often, low birth weight goes along with preterrn birth, that is, birth at or before the 37th week of pregnancy. Low-birth-weight infants are further differentiated by their size in relation to gestational age; if disproportional, prenatal stresses were encountered. Multiple factors contribute to low birth weight, including social and economic conditions, maternal size, maternal health, smoking during pregnancy, lack of prenatal care, geographic locale, ethnicity, toxic exposure during pregnancy, poor pregnancy histories, and more (e.g., Kramer, 1987). However, there is great discrepancy among our population. Healthy, economically comfortable, White women between 20 and 30 years of age can Werner and Smith (1982) provided an interesting picture of biological risk factors, child characteristics, and home conditions that distinguish children and adolescents who succeed versus those who fail. February 1989 • American Psychologist This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. have a 3% low birth weight rate, whereas economically digressed ethnic teenagers often have rates that are three times higher (NCHS, 1987). During the recent past, major improvements have occurred in long-term outcomes for larger low-birthweight infants as a result of m o d e m medical technology and skill found in newborn intensive-care nurseries. 2 Sophisticated care is also permitting increased rates o f survival for the tiniest low-birth-weight infants; however, developmental appraisals suggest a greater proportion of developmental, neurological, and health problems among this group than with their heavier preterm counterparts, In terms of numbers, approximately 6% to 8% of the larger low-birth-weight infants, and perhaps twice that number for the smallest, will show major developmental sequelae although there is considerable variability across samples (e.g., Hoy, Bill, & Sykes, 1988). Given this extraordinary complexity, what generalizations can be made about perinatal outcomes? As, suming differentiation in diagnosis, with de novo perinatal risks most children develop normally, some have mild to moderate dysfunctions, and a few are seriously handicapped. It is considerably more difficult to repeat even this most general statement with respect to outcomes for children exposed to dual prenatal-perinatal risks. More research with carefully defined groups is needed. Postnatal risks. The postnatal period extends from the end of the first month of life to the end o f the first year. Postnatal risks include respiratory disorders, infections, accidents, exposure to environmental contaminants, and nutritional deficiencies. In some cases, such as meningitis, a short, acute episode may be as devastating as a chronic, debilitating condition such as anemia. Postnatal risk events contribute only a small amount to major handicapping conditions among infants and children (Hagberg, 1978). Their contribution to less severe developmental problems is unknown. (Discussion of postnatal risks of nutritional factors and accidents is provided by Lozoff(pp. 231-236) and Christopherson (pp. 237-241), respectively, in this issue. Concluding comment. There is little doubt that risk events hold serious consequences for development. Steps can be taken to prevent risks from occurring or, in some cases, to ameliorate their effects. Preventions and interventions are discussed later in this article. Epidemiology In this section, we discuss the number of infants born each year who have discernible problems and mention the types and percentages of developmental disordersthat are noted afterinfancy. With the exception of birth and mortality statistics,all figuresrepresent approximations as a resultof differencesin reporting methods, sampling criteria,selectionof outcome variables,and definitionof mental retardationand handicap.3 During 1985, more than 3,700,000 births were recorded in the United States.Of these,3% showed evidence of major malformations (Kalter & Warkany, 1983). In February 1989 * American Psychologist real numbers, this means more than 100,000 infants were born with problems that will interfere with their own lives and those o f their families. A precise cause for malformations can be identified only 40% of the time (Vorbees, 1986). Examples include Fetal Alcohol Syndrome, with an overall incidence of I to 2 cases per 1,000 births (Abel, 1984) and Down syndrome, with a similar overall incidence rate. 4 Six to seven percent of all births in this country are of low birth weight (250,000 infants in 1985). Included in this figure are the 1% of infants who are born with a birth weight of 3.5 lb or less (Hughes et al., 1988). Excluding those with major malformations, from 1985 births 15,000 of the larger infants and 3,500 smaller ones have major handicapping conditions. Unfortunately, these numbers do not convey the magnitude of the problem. Only the severest and most obvious difficulties are diagnosed in the first two years of life. After infancy, 3 individuals per 1,000 members of our population show evidence of major developmental disorders. It is thought that most of these stem from prenatal risks. Roughly 35% are due to chromosomal dis. orders including Down syndrome, 20% to multiple anomalies, 8% to inadequate fetal supply, 5% to adverse drug and chemical effects, around 4% to single gene defects (e.g., biochemical disorders that result in Tay Sachs disease, phenylketonuria, or other conditions associated with mental retardation), and about 3% to 4% for early prenatal infections; unknown causes, presumably prenatal in origin, are felt to be instrumental factors in 10% or more. Perinatal risks account for approximately 10% and postnatal risks less than 5% of severe developmental problems (Hagberg, 1975, 1978; Susser, Hauser, Kiely, Paneth, & Stein, 1985). Less severe developmental problems (e.g., IQs in the 50-70 range or minor neurological and sensory disorders) are found in approximately 10 individuals per 1,000 (Haskins, 1986). In contrast to the major developmental disorders, it is believed that only about 20% are due to biological causes; the majority are linked to deleterious social and economic circumstances or to a combination of biological and social vulnerabilities (Haskins, 1986). Many of the less severe developmental difficulties are primarily intellectual and social and are not identified until the child enters educational settings. Sometimes intelligence is in the normal range, but school achievement is lower than expected; this occurs, for example, among 2 The newborn intensive-carenursery is a relativelyrecent development. Physicianswho specializein the care and treatment of sick and tiny infants, called neonatologists,have additional training in newborn function and care. 3There are more similaritiesthan differencesin the rates that are estimatedto occur in major industrializednations. However,these rates generallyare inapplicableto developingnations. 4 The averagerate givenfor Fetal AlcoholSyndromeobscuresthe considerabledisparityfound in the United States. The rate is manyfold higheramongEskimosand NativeAmericans.The averagerate for Down syndrome also obscures actual rates by age of pregnant women. The incidence for a woman in her 20s is often givenas I in 1,000 births, whereas for a woman over40 the rate is I in 50 births. 227 some low.birth-weight children and most frequently among boys (Kopp, 1983). Problematic intellectual functioning and difficulty in school are serious matters for society. Increasingly, data indicate that intellectual problems, irrespective of cause, can be associated with a variety of social and emotional ills (e.g., Olweus, Block, & Radke-Yarrow, 1986). Assessments and Interventions This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Assessment of Infant Developmental Status Infants are typically categorized as those with an uneventful early course and whose development is normal, those exposed to risk events and whose development is either suspect or requires observation, and those exposed to risks that are invariably associated with mental retardation and whose development is delayed. Assessment of developmental status, and the implications thereof, for each of these groups has been fraught with challenges. Nonetheless, assessment takes on heightened meaning because of recently enacted Education of the Handicapped Act Amendments of 1986 (PL 99-457), which provide financial assistance to states for implementing intervention services for infants and families (birth to two years) where a known handicapping condition exists or where development is perceived to be at risk. Developmental assessment is mandated prior to provision of preventive or remedial services. The following discussion focuses primarily on developmental testing for infants whose development may be suspect or at risk. Assessment by a good clinician provides a reasonable evaluation of infant capability; this is an ongoing process in which infant and toddler (up to age three years) performance is observed and a judgment is made about developmental status at the time of testing (e.g., normal, suspect, delayed, or abnormal). Typically, assessment involves clinical evaluation and use of a developmental test, for example the Gesell Schedules (Gesell & Amatruda, 1941) or the Bayley Scales (Bayley, 1969). The tests are based on normative trends, that is, the kinds of mental, motor, language, and social abilities most infants demonstrate at one or another month. A few items ask for primitive problem solving, reasoning, and memory. Developmental tests are not comparable to the intelligence and achievement tests given to school-aged children. The latter ask for practical and abstract intelligence, factual knowledge, arithmetic skills, and sophisticated use of social language. The differences between developmental assessments and intelligence tests highlight the differences between infant capability and the capabilities demonstrated and needed by more mature children and adults. Developmental tests have limitations. They do not evaluate the efficiency by which mental processes are used; how effectively particular behaviors are produced, controlled, or terminated; the characteristics defining growth of adaptive beha~ors; or even why a minor deviation from an expected response might be developmentally significant. 228 The most vocal challenge to developmental tests has surfaced around the issue of prediction from infancy to later childhood capabilities. Repeatedly, research findings from these tests indicate that predictive accuracy is dubious unless the infant has a known handicap (e.g., Down syndrome) invariably associated with later mental retardation or repeated assessments yield moderate to severe developmental delays (KOPp & McCall, 1982). Attempts have been made to enhance predictions by combining infant test scores with social class indexes and family conditions or by cumulating or weighting medical, behavioral, and performance variables. Increases in levels of success have been modest (Goodman, in press). Recently, a new type of infant test has garnered considerable publicity because of claims for prediction. The procedure involves measurement of infant informationprocessing abilities such as visual attention to and memory for patterns and photographs. Because infants have some ability to process incoming information from the earliest months of life, it is suggested that continuity may exist between the processing abilities of infants and children. Further, it is assumed that individual differences in infant processing capability will be related to differences in child information-processing performance. In this regard, several researchers report moderate associations between infancy and preschool language/intelligence test performance. Despite these promising data, caution is warranted before the more traditional developmental tests are discarded. Some methodological and design issues have yet to be addressed. However, developmental tests could also be improved. In the interim, sufficiently valuable judgments of ongoing infant capability can be made with traditional developmental tests if well-trained, developmentally knowledgeable individuals do the assessments. For the long term, it is unlikely that any test will provide absolute definitions of future capabilities for all infants. This would require skills that science does not have---determining the degree of learning a brain is capable of and the ability of children to produce adaptive behaviors and predicting whether ameliorative or adverse factors will surface in a child's life. Interventionsfor Infants Programs of intervention are based on the assumption that there are recognizable conditions in infancy that, if not dealt with appropriately, will result in poor developmental prospects. Interventions may involve services geared exclusively to infants, to their caregivers, or to both. Examples include the carefully controlled introduction of social experiences for preterm infants still in isolettes, provision of counseling and support services for parents who have an infant with a diagnosed handical> ping condition, and interventions aimed at teaching an adolescent mother how to provide play opportunities for her toddler whose development is moving along slowly because of a de novo perinatal risk. All interventions have increased in the past decade. February 1989 • American Psychologist This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Although few would quarrel with the need for program services, concerns have risen about effective implementation and evaluation (e.g., Shonkoff & Hauser-Cram, 1987). In fact, interventions per se are being examined for directions and effectiveness (e.g., Casto & Mastropieri, 1986; Dunst & Snyder, 1986; Guralnick & Bennett, 1987; Marfo & Kysela, 1985). The issues, particularly for infant interventions, do not lend themselves to easy resolution. Aside from challenges that arise because of differences in program goals and orientation, characteristics of the recipients of intervention, and methods of evaluation, infancy presents its own contradictions. At the age when the brain is considered to be most open to learning and experiences, many infant behaviors are remarkably resistant to modification. Driven by strong biological givens, they have their own timetable for growth. Thus teasing out effects requires analysis of the content of intervention in terms of processes targeted for change along with an understanding of processes that are most susceptible to change (Horowitz, 1987; Keogh & Kopp, 1978). Because Public Law 99-457 provides financial assistance for infant interventions when development is or might be compromised, there is opportunity to service more needy infants and families than ever before. Yet policies to determine who should be served vary widely. In some instances, decisions are based on assessment data and include inferences about when to intervene, the nature of the interventions that should be provided, and the kinds of outcomes that can be anticipated. Other times, services are provided on the basis of infant diagnosis (e.g., known handicap or risk condition), because of group membership (e.g., economic distress), or even intervenor philosophy (e.g., that all preterm infants require stimulation). This unsystematic approach has ramifications for training and certification of personnel, quality programming~ and availability of resources at the state and community levels (Burke, McLaughlin, & Valdivieso, 1988, provide a timely discussion of personnel issues). The content and techniques of intervention should differ relative to age and condition of the infant at risk, the circumstances of risks, caregiver information and emotional needs, program models, and personnel and program resources. This diversity leads to questions about how to measure program outcomes. Several have argued for more focus on adaptive characteristics of families, analysis of infant behavioral processes, and measurement of infant social performance rather than extensive reliance on developmental or intelligence test performance. We would also add measures that capture growth of infant adaptive skills, including those related to self-regulatory abilities, emotional control, and anticipatory planning. These are skills often found wanting among children with developmental, educational, and social difficulties. Use of well-defined sample groups for study of outcomes and long-term follow-ups is also advised (Dunst & Snyder, 1986; Shonkoff & Hauser-Cram, 1987). Although this brief discussion has highlighted stumbling blocks for infant intervention, it should not obscure February 1989 • American Psychologist the real and important progress made in the field. Infant intervention is a fact, multiple services are available, and researchers and clinicians are working together toward clarifying the meaning of infant growth and development in the face of risk. All of this did not exist a generation ago. Future Directions This article has focused on the origins of biological risk, influences on infancy and later, assessment of infant developmental status, and interventions for infants. We have called attention to progress that has been made as well as to challenges that remain. Our discussion of risks and developmental outcomes underscores the importance of establishing policies so that this nation's infants and children can optimally function in tomorrow's society. Although it is unlikely that biological risks will be completely eradicated, we can set reasonable goals for prevention and thereby reduce the incidence of adverse outcomes. Accordingly, prevention of avoidable prenatal, perinatal, and postnatal risks should be a high priority. It is our belief that prevention will be served best by formulating information campaigns directed toward young children, adolescents, high-risk family groups, pregnant women, and new parents with a goal of informing without alarming. Innovative, dynamic media messages about nutrition, personal health care, drug and alcohol abuse, known and possible environmental pollutants, contraceptive options, pregnancy diagnosis, prenatal care, and infant development could be systematically disseminated. Effective information models currently exist (e.g., drugs and children, screening efforts for Tay Sachs disease, and avoidance of rubella) and could be the basis for additional efforts involving national agencies, community groups, school systems, health clinics, and recreational associations. Finally, we suggest that the United States require a nationally derived data base that would permit definition of the scope of developmental problems, establishment of research and service priorities based on that knowledge, and determination of resource allocation for preventions and interventions. With reference to infants at risk, a national data base would provide better understanding of infants who have been exposed to risk conditions, the nature of the risks, the percentage of infants and children who have known handicapping conditions, the percentage of infants whose development is considered to be at risk, and the percentage of infants who subsequently develop educational and social problems in childhood. The multidisciplinary expertise required to mount such a coordinated effort is available. REFERENCES Abel,E. L. (1984).Fetal alcohol syndrome andfetal alcohol effects. New York: Plenum. Bayley,N. ( 1969). The Bayley Scales of Mental Development. NewYork: PsychologicalCorporation. Burke, P. J., McLaughlin, M. J., & Valdivieso,C. H. (1988). Preparing professionalsto educatehandicappedinfantsand youngchildren:Some 229 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. policy considerations. Topics in Early Childhood Special Education, 8, 73-80. Casto, G., & Mastropieri, M. A. (1986). The efficacy ofeady intervention programs: A meta-analysis. Exceptional Children, 52, 417--424. Christophersen, E. R. (1989). Injury control. American Psychologist, 44, 237-241. Dorfman, S. E (1986). Age as a factor in pregnancy. Contemporary OB/ GYN. 24, 64-77. Dunst, C. J., & Snyder, S. W. (1986). A critique of the Utah State University early intervention meta-analysis research. Exceptional Children, 52, 269-276. The Education of the Handicapped Act Amendments of 1986, Pub. L. No. 99-457. Freeman, J. M. (1985). Summary. In J. M. Freeman (Ed.), Prenataland perinatal factors associated with brain disorders. (NIH Publication No. 85-1149, pp. 3-32). Washington, DC: U.S. Department of Health and Human Services. Gesell, A., & Amatruda, C. S. ( 1941 ). Developmental diagnosis: Normal and abnormal child development. New York: Hoeber. Goodman, J. (in press). Infant intelligence: Do we, can we, should we assess it? In C. R. Reynolds & R. Kamphaus (Eds.), Handbook of psychological and educational assessment. New York: Guilford. Guralnick, M. J., & Bennett, E C. (1987). The effectiveness of early interventionfor at-risk and handicapped children. Orlando: Academic Press. Hagberg, B. (1975). Ire-, peri- and postnatal prevention of major neuropediatric handicaps. Neuropaediatrie, 6. 331-338. Hagberg, B. (1978). Severe mental retardation in Swedish children born 1959-1970: Epidemiolngical panorama and causative factors. In Major mental handicap." Methods and costs of preventiou (Ciba Foundation Symposium No. 59, pp. 29-51). Amsterdam: Elsevier. Haskins, R. (1986). Social and cultural factors in risk assessment and mild mental retardation. In D. C. Farran & J. D. McKinney (Eds.), Risk in intellectualand psychosocialdevelopment (pp. 29-60). Orlando: Academic Press. Horowitz, E D. (1987). Tar~:ting infant stimulation efforts: Theoretical challenges for research and intervention. In N. Gunzenhauser (Ed.), Infant stimulation: For whom, what kind, when, and how much? (pp. 97-108). Skillman, N J: Johnson & Johnson. Hoy, E. A., Bill, J. M., & Sykes, D. H. (1988). Very low birthweight: A long-term developmental impairment? International Journal of Behavioral Development, 11, 37-67. Hughes, D., Johnson, K., Rosenbaum, S., Butler, E., & Simons, J. (1988). The health of America's children: Maternal and child health data book. Washington, DC: Children's Defense Fund. Kalter, H., & Warkany, J. (1983). Congenital malformations: Etiologic factors and their role in prevention. New England Journal of Medicine, 308, 424-431,491--497. Kcogh, B. K., & Kopp, C. B. (1978). From assessment to intervention: an elusive bridge. In E D. Minifie & L. L. Lloyd (Eds.), Communicative and cognitive abilities: Early behavioral assessment. Baltimore: University Park Press. KOPp, C. B. (1983). Risk factors in development. In M. M. Haith & 230 J. J. Campos (Eds.), Handbook of child psychology: Vol. 2. Infancy and developmentalpsychobiology (pp. 1081-1188). New York: Wiley. Kol~, C. B., & McCall, R. B. (1982). Predicting later mental performance for normal, at-risk, and handicapped infants. In P. B. Baltes & O. G. Brim (Eds.), Life-span development and behavior (Vol. 4, pp. 33-60). New York: Academic Press. Kramer, M. S. (1987). Intrauterine growth retardation and gestational duration determinants. Pediatrics, 80, 502-51 I. Lozoff, B. (1989). Nutrition and behavior. American Psychologist, 44, 231-236. Marfo, K., & Kysela, G. M. (1985). Early intervention with mentally handicapped children: a critical appraisal of applied research. Journal of Pediatric Psychology, 10, 305-324. Mednick, S. A., Machon, R. A., Huttenen, M. O., & Bonett, D. (1988). Adult schizophrenia following prenatal exposure to an influenza epidemic. Archives of General Psychiat~ 45, 189-192. Miller, A. C. (1987). Maternal health and infant survival. Washington, DC: National Center for Clinical Infant Programs. National Center for Health Statistics. (1987). Advance report of final natality statistics, 1985 (Monthly vital statistics report, Vol. 36). Washington, DC: Author. Olweus, D., Block, J., & Radke-Yarrow, M. (1986). Development of antisocial and prosocial behavior:Research. theo~ and issues. Orlando: Academic Press. Rosen, M. G. (1985). Factors during labor and delivery that influence brain disorders. In J. M. Freeman (Ed.), Prenatal andperinatalfactors associated with brain disorders (NIH Publication No. 85-1149, pp. 239-262). Washington, DC: U.S. Department of Health and Human Services. Sameroff, A. J., & Chandler, M. J. (1975). Reproductive risk and the continuum of earetaking casualty. In E D. Horowitz (Ed.), Review of child development research(Vol. 4, pp. 187-244). Chicago: University of Chicago Press. Shonkoff, J. P., & Hanscr-Cram, P. (1987). Early intervention for disabled infants and their families: A quantitative analysis. Pediatrics, 80, 650657. Spellacy, W. N., Miller,S. J.,& Winegar, A. (1986). Pregnancy after40 years of age. Obstetricsand Gynecology,,68, 452--454. Susser, M. B., Hauser, W. A., Kiely,J. L., Paneth, N., & Stein,Z. (1985). Quantitative estimates of prenatal and perinatal risk factorsfor perinatal mortality, cerebral palsy, mental retardation, and epilepsy. In J. M. Freeman (Ed.), Prenatal and perinatal factors associated with brain disorders (NIH Publication No. 85-1149, pp. 359--439). Washington, DC: U.S. Department of Health and Human Services. Vorhees, C. V. (1986). Origins of behavioral teratology. In E. P. Riley & C. V. Vorhees (Eds.), Handbook of behavioral teratology (pp. 3-22). New York: Plenum. Vorhees, C. V., & Moilinow, E. (1987). Behavioral teratogenesis: Longterm influences on behavior from early exposure to environmental agents. In J. D. Osofsky (Ed.), Handbook of infant development (2nd ed., pp. 913-971). New York: Wiley. Werner, E. E., & Smith, R. S. (1982). Vulnerable but invincible. New York: McGraw-Hill. F e b r u a r y 1989 * A m e r i c a n Psychologist