Early Childhood Assessment: Why, What, and How (2008)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

4 Screening Young Children I n this chapter we review procedures for the screening of young children for both health-related and developmental purposes. We discuss the uses of assessment for infants and toddlers and the contexts in which they are assessed. We describe various screens performed on infants and toddlers to detect possible physio­logical, sensorineural, (micro)nutritional, and environ­mental threats to healthy development, and then we review assess­ments of developmental status and their use for screening purposes. We discuss two sorts of challenges to effec- tive screening: the difficulties of assessing very young children and the societal conditions that need to be in place. Finally, we present a set of summary tables of some of the assessment instru- ments available for use with children age 3 years and younger. We realize that this review is not exhaustive. We have focused on screening for conditions that have implications for educational outcomes, in line with our charge, the primary audience for this report, and acknowledging the limitations on the committee’s resources. The issues central to the committee’s charge pertain chiefly to instruments used with children in early childhood care and educational programs, so our major focus was on the pre- school age group, which forms the majority of the children served by these programs. For the same reasons, we have focused our discussion here on screening rather than on in-depth diagnostic 61 

62 EARLY CHILDHOOD ASSESSMENT assessment for infants and toddlers, although we do list widely used diagnostic instruments in our tables. The diagnostic instru- ments are most often used by specialists after screening-based referral. In lieu of more extensive discussion, we provide refer- ences to recent reviews of infant and toddler instruments in which more detailed information can be found. assessing infants and toddlers The traditional model for assessment of infants in the first several months of life was primarily medical. It focused on using assessments for pediatric appraisal of normative ­physical and neuromotor development. Attention to behavioral and psycho­social factors was secondary, although most pediatricians acknowledged the importance of developmental factors over those of a purely physical or biological nature. Over the past half-century, behavioral development has become an integral part of regular pediatric evaluation, and pediatricians routinely provide clinical information on behavioral, cognitive, and psycho­ social factors, thus providing a more comprehensive picture of each child’s overall growth and development. The integration of biophysical examination with cognitive and socioemotional assessment links early infant assessment with the developmental outcomes of interest in this report. Uses of Assessment The uses and purposes of assessment in the infant-toddler period determine to some extent the domains assessed. One source suggests four purposes of infant assessment (Wyly, 1997): 1. to identify infants who may be at risk for developmental delay, 2. to diagnose the presence and extent of developmental problems, 3. to identify an infant’s specific abilities and skills, and 4. to determine appropriate intervention strategies. 

SCREENING YOUNG CHILDREN 63 This list does not include many purposes typical of assess- ment for older preschoolers, such as evaluation of intervention strategies, prediction of future competencies, or assessment of skills that are fundamental for success in a classroom environ- ment, such as ease of gaining the child’s attention and ability to sustain it. The focus is on the identification of possible develop- mental problems at an early age—in part, we argue, because of the relatively undifferentiated nature of developmental organization in early infancy and the associated difficulty of making precise predictions to later abilities. We note also that in spite of wide agreement that screening and monitoring of the development of these youngest children is important, pediatricians still do not fully agree on the most important domains to measure or the best measures to use (McCormick, 2008). Most of the assessment conducted in this age range is actually screening to identify potential problems, to be followed by more definitive diagnostic assessment. The principles of a good screen- ing program are thus relevant (Wilson and Jungner, 1968): • a valid and reliable measure, • acceptability to the population being screened and their parents or guardians, • facilities to conduct the screening, • facilities to ensure follow-up and treatment, and • cost-effectiveness. Contexts and Assessment As noted, assessment of infants and toddlers often takes place in pediatric settings, with screening as a primary goal. Screening may also take place in early childhood education and intervention settings, such as Early Head Start and home visiting programs. Interpreting results from such assessments must take into account the effects of a wide variety of inputs into the child’s development, for example, safety of the residence, care practices of parents and other caregivers, exposure to substances that might hamper normal development, and consistency of care settings, as well as information about the infant’s state of health and alertness during the assessment. 

64 EARLY CHILDHOOD ASSESSMENT There is an explicit assumption that child care practices, caregiver stability, and infant-caregiver attachment provide the basis for optimal social and cognitive development. However, for many children, including those under age 3, substantial vari- ability exists in the types, extent, and number of forms of out-of- home care available (Johnson, 2005); this variability may be even greater for children at risk of developmental delay, who may also be eligible to receive community-based early intervention services ( ­ Widerstrom, 1999). Understanding the quality of these variable settings, as well as the impact of the child’s exposure to different settings, is crucial in interpreting child-based outcomes. Because of the variety of the settings in which infants and t ­ oddlers are cared for, the equivalent of the older child’s classroom as a place for administering developmental assessments is avail- able only for the minority of children now reached by infant and toddler intervention and education programs like Early Head Start. However, because the vast majority of children under age 4 are monitored by pediatricians or family practitioners (Freed, Nahra, and Wheeler, 2004) and regular developmental assessment is rec- ommended for well-child care, the pediatric setting thus becomes the most likely site for infant and toddler screening. This fact has implications for the training of pediatric personnel, for the design of organized data systems useful in ensuring that all children are screened for developmental problems, and for an integrated service delivery system that spans medical and educational settings. Assessing Threats to Normative Development We focus here on threats that are susceptible to prevention or amenable to postnatal intervention. There is a much longer list of factors associated with increased risk to normative development, ranging from child-specific (low birth weight, prematurity) to societal (poverty) factors; the ones discussed here are merely a selection. Genetic/Metabolic Screening Currently, every newborn in the United States is screened at birth for certain genetic conditions and metabolic disorders, 

SCREENING YOUNG CHILDREN 65 although the number of conditions varies by state (Kaye and C ­ ommittee on Genetics, 2006; Lloyd-Puryear et al., 2007). Many of these conditions result in significant nervous system damage, leading to severe developmental delays, which early treatment may prevent or ameliorate (Kaye and Committee on Genetics, 2006). In the past, such screening depended on chemical analyses of a spot of blood taken at the time of discharge from the hospital nursery, lim- iting the number of conditions for which screening could be done. More recently, the use of tandem mass spectrometry (MS/MS) has greatly expanded the number of conditions for which screening is possible (Schulze, 2003). Although this technology is expensive to implement, its use has been argued to be very cost-effective (Carroll and Downs, 2005). Moreover, since neonatal metabolic screening has been so well incorporated into care following birth, it is gener- ally well accepted by both providers and parents. Estimating the effect of newborn genetic/metabolic screen- ing is made difficult by several factors (Botkin, 2004; Kaye and Committee on Genetics, 2006). First, when newborn screening programs were initiated, the assumption was that an affected gene led to disease. Advances in modern genetics have revealed that many mutations may occur in a single gene, not all of them leading to significant disease, and it often is unclear whether treat- ment is needed. Second, the expanded MS/MS techniques reveal biochemical abnormalities that may or may not be associated with specific disease states, so the natural history of some of these abnormalities is unknown. Infant maturation may affect detec- tion; for example, congenital hypothyroidism may be difficult to detect in preterm infants. Moreover, these tests, while having some power of detection, are not a proxy for functional outcomes related to behavior. The prevention of developmental disability requires a system of detection, validation, and treatment, and the treatments may be onerous, thereby affecting compliance. Finally, many more infants test positive on the screening tests than have the disease, and assessing these infants adds to the costs without preventing disability. In addition to the costs, simply identifying the infants who test falsely positive may have unintended consequences on their development (Fisher and Welch, 1999; Newman, Browner, and Hulley, 1990). Despite these concerns, neonatal ­ metabolic screening has proven to be an effective screening process. 

66 EARLY CHILDHOOD ASSESSMENT Newborn Hearing Screening Most states have introduced neonatal screening for con­genital hearing loss (Kaye and Committee on Genetics, 2006). Such screen- ing has been made possible by the development of relatively portable, computerized equipment. One approach, automated auditory brainstem response, a variant on electro­encephalography, detects within 10 milliseconds the speed and amplitude of 5-7 com- ponent waves from the auditory nerve through structures along the auditory pathway in response to specific sounds. The screen- ing test consists of soft clicks delivered through earphones and the detection of an auditory brainstem response through electrodes on the infant’s forehead. Another approach, otoacoustic emissions, involves using a tiny microphone to detect the sounds made by the outer hair cells of the cochlea (National Center for Hearing Assessment and Management, 2007). Proponents argue that children with hearing loss who receive intensive early intervention do better on school performance measures and have improved receptive language and less devel- opmental delay (Blake and Hall, 1990; Moeller, 2000; Yoshinaga- Itano et al., 1998). However, unlike genetic/metabolic screening, in which specimens are sent to a central laboratory, offering greater control over technical quality, newborn hearing screening is conducted in the newborn nursery by a variety of personnel using a variety of protocols (Kaye and Committee on Genetics, 2006). The evidence does not appear to favor one protocol over another, and some hearing-impaired infants are still being missed (Gravel et al., 1999). Thus, a 2001 review of newborn hearing screening by the U.S. Preventive Services Task Force concluded that the evidence at that time was incomplete as to the benefit of newborn hearing screening, but an updated review is in progress (U.S. Preventive Services Task Force, 2001). Vision Screening Vision screening is a recurrent and routine part of the pedi- atric physical examination. Early assessments focus on ensuring that there is a clear pathway from the front of the eye to the retina, where images are received; that the connection between the retina 

SCREENING YOUNG CHILDREN 67 and the relevant part of the brain is intact, indicated by pupil- lary responses to light; and that the eyes move in a coordinated f ­ ashion. Between ages 2 and 4 years, it becomes possible to test for visual acuity—that is, the size of objects that can be seen at certain distances (American Academy of Pediatrics, 1996). The goal of these procedures is to reduce poor vision or risk factors that lead to abnormal visual development. Recent evidence supports the effectiveness of intensive screening for the reduction of amblyopia and improved visual acuity. The U.S. Preventive Services Task Force concluded that the routine screening currently done has not been shown to be effective, although the potential benefit outweighed the minimal risk of the screening (U.S. Preventive Services Task Force, 2004). Iron Deficiency Screening A lengthy literature addresses the effect of nutritional defi- ciency and child development (Grantham-McGregor, 1984). Since poor nutrition and micronutrient deficiency are more likely in the context of poverty and ill health, disentangling the effect of specific nutritional deficiencies on development is sometimes difficult. However, evidence from developing and industrialized countries supports a relation between iron deficiency and poorer socioemotional, sensorimotor, and cognitive development and school performance (Lozoff et al., 2000, 2003). Recommendations for screening for iron deficiency are consistent with this body of research (American Academy of Pediatrics, 2003). However, substantial questions about the specificity of using blood hemo- globin levels to assess the presence of iron deficiency led the U.S. Preventive Services Task Force to conclude that the evidence is insufficient to recommend for or against such screening (U.S. Preventive Services Task Force, 2006). Acuity tests, such as Teller Acuity Cards, are available for infants and toddlers, and they can be useful for at-risk (e.g., premature) infants, but they are not suitable for general screening and good predictive validity has not been demonstrated (National Research Council, 2002). 

68 EARLY CHILDHOOD ASSESSMENT Lead Screening Lead absorbed from the environment has long been recog- nized as a neurotoxicant, and major efforts have been undertaken to reduce environmental lead (Grandjean and Landrigan, 2006). The success of these efforts has led to a sharp decline in the blood lead levels of children in America: as of 2006, only slightly more than 1 percent had blood lead levels above the cutoff of 10 micrograms/deciliter (Centers for Disease Control and Preven- tion, 2007). Nonetheless, certain populations, such as minority children and those living in older housing stock, remain at risk, and thus a targeted screening strategy has been recommended by the American Academy of Pediatrics (2005). Several studies have reported that children with low-level prenatal lead exposure (< 10 mg/dl) have intellectual deficits as measured by standard IQ tests (Banks, Ferrittee, and Shucard, 1997; Lanphear et al., 2000, 2002; Needleman and Gatsonis, 1990) reflected in poorer perfor- mance on specific items on the Neonatal Behavioral Assessment Scale (Brazelton and Nugent, 1995; Emory et al., 1999) and on infant intelligence at age 7 months (Emory et al., 2003; Shepherd and Fagan, 1981). The study by Emory et al. (2003) characterized the effects found as lowered optimal performance rather than an increase in impaired performance across the board. DEVELOPMENTAL ASSESSMENT Newborns Developmental assessments provide useful information about overall physiological status and risk. Neurodevelopmental exami- nations initially focused on neurological reflexes and postural reactions that can be elicited in the newborn, which emerge and disappear within fairly specific time periods, as a means of assess- ing central nervous system integrity, especially early signs of cerebral palsy (Zafeieriou, 2003). Primitive reflexes are mediated by the brainstem and consist of complex, automatic movement patterns that emerge from 25 weeks of gestation and disappear by age 6 months. Postural reactions are infant responses to being held in different standardized positions and probably reflect more 

SCREENING YOUNG CHILDREN 69 complex stimuli, such as those from joints, muscles, and other proprioceptors. Persistence of primitive reflexes and postural reactions or asymmetry in response tend to suggest central ner- vous system problems. In his review, however, Zafeieriou (2003) notes that there is considerable controversy about which reflexes or reactions, or combination thereof, provide the best clinical pre- diction. It should be noted that the major outcome being predicted is cerebral palsy. More recently, neurodevelopmental assessment has also focused on behavioral attributes of the infant, particularly as they reflect organizational state and the ability to interact with the environment. The premise of this approach is that the infant is an active learner from birth and that his or her ability to change states or control his or her state in response to internal and external stimuli facilitates that learning. The Neonatal Behavioral Assess- ment Scale (NBAS) evaluates the self-regulatory capacity of the newborn infant to achieve two sleep states, two awake states, and one distress state. Within states, the infant may modify his or her response to external or internal stimuli by either habituat- ing (i.e., not responding to a stimulus) or processing information from various sensory inputs (Tronick, 1987). The items on the NBAS cluster into seven areas: habituation, motor performance (tone and maturity), range of state, regulation of state, autonomic regulation, and reflexes (Tronick, 1987). The NBAS relies on well- trained examiners, and it is unclear to what extent variations in examiner ability influence its predictive validity. The NBAS has been used to assess the effects of a variety of prenatal exposures, including obstetric medications, recreational drug use, and environmental toxins (Tronick, 1987). It has proven sensitive to normal and abnormal variations in fetal responses to labor (Emory, Walker, and Cruz, 1982), birth weight difference inde- pendent of gestational age (Emory and Walker, 1982), and clinical conditions with known neurological and neurobehavioral conse- quences (Emory, Tynan, and Davé, 1989). The NBAS is also used as a tool to educate parents about the skills of their newborn infants and to improve infant-parent interactions (Beeghly et al., 1995). A meta-analysis by Das Eiden and Reifman (1996) concluded that interventions using this scale during the neonatal period have a small to moderate positive effect on the quality of later parenting. 

70 EARLY CHILDHOOD ASSESSMENT Using the model of the NBAS, Als et al. (2005) have devel- oped the Assessment of Preterm Infants’ Behavior (APIB). The scale assesses what are theorized to be five interacting systems of functioning: autonomic, motor, state organization, attention, and self-regulation. Like the NBAS, the APIB forms the basis of an intervention, the Newborn Individualized Development Care and Assessment Program, intended to improve the develop­ mental outcomes of preterm infants by teaching caregivers in the neonatal intensive care unit how to interact more sensitively with the infant. If the intervention improves performance on the APIB and leads to better long-term outcomes in early childhood, then one might argue that the APIB has predictive validity, and Als et al. (2003) have argued for such an effect. However, a recent meta- analysis of individualized developmental interventions in the neonatal intensive care unit suggests that the data do not support this argument (Jacobs, Sokol, and Ohlsson, 2002). Infants and Toddlers Developmental assessment of infants and toddlers occurs routinely in medical care settings and is carried out by a variety of people; some children receive this service through infant-­toddler care/education/intervention programs. In view of the time pres- sures in primary care settings, the approach has been to rely on brief screening instruments, with more complete assessments of children who do not seem to be developing at the usual pace. Since most young children are monitored by pediatricians or other primary medical care personnel, it seems reasonable to use the clinical guidelines from the American Academy of Pediatrics (American Academy of Pediatrics, Committee on Children with Disabilities, 2001; American Academy of Pediatrics, Council on Children With Disabilities, 2006) as a template for this process. The first step is developmental surveillance performed as part of the regular well-child visit. Surveillance is considered to include “eliciting and attending to the parents’ concerns, document- ing and maintaining a developmental history, making accurate observations of the child, identifying risk and protective factors, and . . . documenting the process and findings” (American Acad- emy of Pediatrics, Council on Children with Disabilities, 2006). If 

SCREENING YOUNG CHILDREN 71 developmental concerns are identified or the visit occurs at 9, 18, or 30 months, then a more structured developmental screen is rec- ommended, and several examples are provided. If the screening results are positive, then the child should be referred for formal developmental assessment and early intervention services. There are three general types of infant and toddler devel- opmental assessments (see Appendix Tables 4-1 through 4-7 for some examples). First, screening tests may take the form of ques- tionnaires for the primary caregiver about the child’s activities, either soliciting the achievement of specific developmental mile- stones (smiling, walking, specific words) (Glascoe, Martin, and Humphrey, 1990) or eliciting more general assessments of child development (Glascoe, 2003). Second, developmental assess- ment may take the form of observations of child activities on a limited number of items. No matter which of these approaches is used initially, any child found to have developmental difficulties requires access to the third type, a more refined assessment with a professionally administered developmental tool. The large number of available assessment instruments has been summarized in several publications (Buros Institute of Mental Measurements, 2007; Child Trends, 2004; Glascoe, 2003, 2005; Glascoe et al., 1990; Mathematica Policy Research, 2003). The website (http://www.dbpeds.org) of the American Academy of Pediatrics, in the section on Developmental and Behavioral Pediatrics, provides information on developmental screening and assessment. The Educational Testing Service also maintains an online catalog of instruments (http://sydneyplus.ets.org/ search.asp?). The National Institute for Early Education Research (NIEER) has a similar database (http://nieer.org/assessment/). A new document from the National Early Childhood Technical Assistance Center covers instruments with a focus on social and emotional measures. Appendix D provides more information on these sources for detailed descriptions of instruments. The instruments tend to cover similar domains of develop- ment, such as general cognitive skills; language, motor, and socioemotional development; and functional abilities appropriate to the age of the child. Appendix Tables 4-1 through 4-7 list some of the available instruments. Each table identifies the domain assessed, the type of instrument (usually screening or diagnostic), 

72 EARLY CHILDHOOD ASSESSMENT and the method by which data are gathered: caregiver report, direct observation of the child, or both methods. The tables are not meant as an endorsement of any instru- ments, but rather as a way to categorize instruments that are fre- quently used and to lead the reader to references, like those listed above, that provide more detailed information on each. Challenges in Effective Infant Screening There are two sets of challenges to be faced in generating an optimal system of infant assessment for screening purposes. The first set has to do with the inherent difficulty of assessing very young children reliably and validly, and the second with the many societal conditions that need to be in place to ensure effective infant assessment and use of infant assessment information The Difficulty of Assessing Young Children Very young children are hard to assess reliably and validly because of the relatively undifferentiated nature of their capa- bilities. Infants are less differentiated than older children—that is, children express their developmental status in increasingly differentiated ways as they mature (National Research Council and Institute of Medicine, 2000). Moreover, the environment in which abilities are expressed changes drastically from infancy to preschool and beyond, thus requiring changes in the child’s adap- tive capacity as well. Young children also show enormous variability within and across individuals, reflecting the emerging differentiation of functional systems. This developmental state gradually gives way in later childhood to narrower windows of performance considered to be “within normal limits.” Embedded in this concept of “normal limits” is an expectation that, as children mature, their behavior will conform to the increasingly stringent standards and expectations associated with social and academic success. In infancy, biological homeostasis, autonomic regulation, and organizational properties of behavioral development are important indicators. These might be informally assessed by 

SCREENING YOUNG CHILDREN 73 observing how long it takes for an infant to calm down after a stressful event, such as an injection; whether an infant turns away from highly stimulating events before becoming overexcited; or whether a 14-month-old turns to a caregiver when confronted by an ­unfamiliar or frightening stimulus. By the time a child reaches age 2 years, autonomic regulation is typically under control, so the developmental challenges associated with gross and fine motor control, receptive and early expressive communication skills, and socioemotional regulation of affective states are now more impor- tant and more susceptible to assessment. By age 5, the child’s major developmental challenges include expressive language and social communication skills, affect regulation in the context of broader social and peer relations, and cognitive maturation com- mensurate with instruction in a formal educational setting. The child’s expanding repertoire of behavioral and social a ­ bilities, including linguistic communication skills, opens up more options for assessment during the toddler years. Assessing infants permits only a relatively global appraisal of level of func- tioning. Infant assessment is therefore focused on optimal perfor- mance and the testing of limits more than on assessing whether the infant can pass a minimum threshold of performance in any particular domain. For infants even more than older children, optimal performance is dependent on state of arousal. For infants and toddlers as for older preschoolers, effective assessment of behavioral functioning presupposes that the child attends to the relevant information. If the child is not attending, assessment results are typically viewed as invalid. The ability to sustain attention for information-processing purposes can itself be assessed from birth through age 5. Prediction of later outcomes would be much easier if devel- opmental assessments used with infants had a one-to-one cor- respondence to measures taken later. Under such circumstances, the timing of early developmental milestones—such as when the child sits unassisted, begins to grasp objects, crawl, babble, and declare wants and intentions—would lead to accurate predictions of later walking, handedness, speech development, and emotion regulation. There is no practical or reliable measure of any specific domain in early infancy that gives a precise prediction about the child’s performance in that domain several years later; in part 

74 EARLY CHILDHOOD ASSESSMENT this fact reflects the enormous plasticity of the developing child and susceptibility to environmental influences. Thus, though screening measures of infant functioning can be very important in identifying the need for further diagnostic assessment to reveal conditions that represent risk for poor performance later on, as well as in allowing early access to prevention or intervention, assessment for purposes of tracking development or predicting later outcomes is less likely to be useful. Conditions Required for Effective Screening A second set of challenges to effective screening arises from the complexity of putting together the societal conditions required to do it well. Several problems limit the potential usefulness of the current system for infant and toddler developmental screen- ing. First, there are concerns about the validity of the instru- ments themselves. The sample sizes on which many tests were validated may be insufficient to provide robust estimates of their s ­ ensitivity—that is, their ability to identify those affected—and specificity—the ability to avoid identifying those not affected (Camp, 2007). Sensitivity may be further affected when the ref- erence test is given to all who score in the abnormal range but to only a sample of those in the normal range. Some screening tests have used reference tests with outdated norms, resulting in inflated scores. In addition, several have procedural problems that could lead to biased results, and often the reported results do not indicate the predictive validity (Camp, 2007). A second issue is that responsibilities for screening are dispersed across individuals and settings, and that a standard procedure for administering screenings has not been established. Thus, the screening assessments may not be administered and, if they are, may not provide comparable information across pro- viders. A recent assessment of the quality of pediatric ambula- tory care revealed that children received fewer than half of the recommended procedures and that screening procedures were particularly unlikely to be performed (Mangione-Smith et al., 2007). These results parallel those for specific screening tests (Biondich et al., 2006; Wasserman, Croft, and Brotherton, 1992). Clearly, if administration procedures are to be standardized and well implemented, medical and education practitioners working 

SCREENING YOUNG CHILDREN 75 with infants and young children need training and support in the appropriate procedures. Finally, the effectiveness of screening may be further limited by the fact that the system of access to screening settings and of response to abnormalities found may be as diffuse and unstan- dardized as the assessment process itself. Unlike the classroom setting, in which more standardized and local approaches to developmental and learning problems may be taken, response to abnormalities of development in infants, toddlers, and older preschoolers not already enrolled in intervention programs typi- cally requires referral to other services for diagnosis and manage- ment. In part, this variability in response reflects the diversity of state and other policies regarding young children. This means that some infants and toddlers are not screened, and that those who are identified as requiring diagnostic assessments and other services may not receive them. As noted above, much of the early screening is accomplished in health care settings, and access to care is heavily dependent on having health insurance. Children without health insurance are more likely to have low family income, to come from minority families, to use medical care less intensely, and to be referred to other settings for services (Simpson et al., 2005). Even with insurance, access to some services is more difficult than others. Although the Individuals with Disabilities Education Act does mandate testing for all children suspected of developmental disability or delay and requires the provision of appropriate services to children so identified, there remains considerable local variation in the capacity to respond to this mandate. A recent chapter by Gilliam, Meisels, and Mayes (2005) proposes a system of screening and surveillance that uses many available community resources to provide a more integrated screening, referral, and assessment system. Finally, even if the current assessment of infant and toddler development were more universally effective, fitting well into a larger system and building continuity with the assessment of slightly older preschoolers would improve its usefulness. The focus of infant-toddler assessment procedures is primarily on monitoring development and risks to development for purposes of ensuring adequate progress and to rule out health-related chal- lenges to normal development. For example, the vision examina- tions conducted by health care providers may focus less on the 

76 EARLY CHILDHOOD ASSESSMENT visual acuity needed for classroom work and more on detecting opacities in the eye (e.g., cataracts) that may hamper visual devel- opment or muscle imbalances that might signal other neurological problems. Likewise, screening for iron deficiency should attend to the cognitive deficits associated with it as much as evaluating the child’s nutritional status and addressing questions about the production and destruction of red blood cells and potential covert blood loss. Conclusion Assessment of important behavioral and physiological out- comes for infants and toddlers is an increasing focus of pedia- tricians, primary medical care providers, and providers of care and education to infants and toddlers. Ideally, these individuals recognize the full array of information—child performance, care- giver report, observation—that can be used and are well trained to collect information systematically. While screening for risk is a key goal of assessment during this developmental period, an equally important goal is tracking well-child developmental indicators and focusing on what children can do as well as what they have problems with. For children with disabilities that have already been identified in this early period, a focus on functional capacities may be more important than a delineation of limita- tions. Although screening for risk and assessment for well-child functioning are widely practiced, the system of infant and toddler assessment needs to be expanded in a number of ways. First, it is important that children living in poverty and chil- dren from cultural and language-minority groups are included in these assessments. Second, the system linking assessment results to other resources—referrals, follow-up, access to services—is at this time far from seamless. Identifying risk or disability in a young child does little good if no provisions have been made to remedy or mediate the problem, to help caregivers understand and address it, or to link the early available information to deci- sions about interventions, schooling, and ongoing attention. We raise again the importance of thinking systematically if the potential of assessment to improve child learning and welfare is to be realized. 

Appendix TableS: Summary of Assessment Instruments for Children 0-3 Years of Age APPENDIX TABLE 4-1  Domain: Cognition Data-Gathering Method Instrument Type Caregiver Report Observation Mixed/Both Screening Ages and Stagesa Developmental Indicators for Battelle Developmental Inventory Infant Development Inventory Assessment of Learning-Revised Screening Testa NCHS/NLSY Questionnaire Slosson Intelligence Test Developmental Profile-II (U.S. Department of Health and Lexington Developmental Scalesa Preschool Screening System Human Services, National Center Bayley Infant Neuro­developmental Denver Developmental Screening for Health Statistics, 1981) Screener (BINS) (Aylward, 1995) Test IIa Parents’ Evaluation of Brigance Screens Developmental Statusa Fagan Test of Infant Intelligence Infant Monitoring Systema Denver Prescreening Developmental Questionnairea 77 continued 

APPENDIX TABLE 4-1  Continued 78 Data-Gathering Method Instrument Type Caregiver Report Observation Mixed/Both Child Development Inventory and Child Development Review-Parent Questionnaire (Ireton, 1992) Parents/Evaluation of Developmental Status (PEDS)a Capute Scales (CAT/CLAMS) (Voigt et al., 2003) Diagnostic Bayley Scales of Infant Development, Third ed. McCarthy Scales of Children’s Ability Mullen Scales of Early Learning aIncludes questions on behavioral issues or personal-social development. 

APPENDIX TABLE 4-2  Domain: Language Data-Gathering Method Instrument Type Caregiver Report Observation Mixed/Both Screening The Quick Test Peabody Picture Vocabulary Test Communication and Symbolic Expressive One-Word Picture Behavior Scales (Wetherby and Vocabulary Test Prizant, 2002) Early Language Milestone Scale (Coplan, 1993) Diagnostic Receptive Expressive Emergent Reynell Developmental Language Sequenced Inventory of Language Scale (REEL)a Scales Communication Development MacArthur-Bates Communicative Preschool Language Scale Development Inventories Test of Early Language Development aRequires trained interviewer/observer. 79 

APPENDIX TABLE 4-3  Domain: Motor 80 Data-Gathering Method Instrument Type Caregiver Report Observation Mixed/Both Screening Early Motor Pattern Profile (EMPP) (Morgan and Aldag, 1996) Motor Quotient (Capute and Shapiro, 1985) Diagnostic Bayley Scales of Infant Development, Third ed. (see above) Movement Assessment of Infants (Chandler, Andrews, and Swanson, 1980) Peabody Developmental Motor Scales (Folio and Fewell, 1983) Alberta Infant Motor Scale (Piper and Darrah, 1994) 

APPENDIX TABLE 4-4  Domain: Social-Emotional Data-Gathering Method Instrument Type Caregiver Report Observation Mixed/Both General Eyberg Child Behavior Inventory Bayley Scales of Infant Vineland Social-Emotional Maturity Infant-Toddler Social Emotional Development, Third ed. Scalea Assessment, ITSEA Brief Infant-Toddler Social Emotional Assessment, BITSEA Achenbach System of Empirically Based Assessment Devereux Early Childhood Assessment Temperament and Atypical Behavior Scale (TABS) (Bagnato et al., 1999) Screens for Modified Checklist of Autism in Specific Toddlers (Dumont-Mathieu and Developmental Fine, 2005) Disabilities Checklist for Autism in Toddlers (CHAT) (Baird et al., 2000) 81 continued 

APPENDIX TABLE 4-4  Continued 82 Data-Gathering Method Instrument Type Caregiver Report Observation Mixed/Both Pervasive Developmental Disorders Screening Test-II (PDDST-II) (Siegel, 2004) Screening Tool for Autism in Two- Year-Olds (STAT) (Stone, Coonrod, and Ousley, 2000) Social Communication Questionnaire (SCQ) (Rutter, Bailey, and Lord, 2003) aRequires trained interviewer/observer. APPENDIX TABLE 4-5  Domain: Function/Activities of Daily Living Data-Gathering Method Instrument Type Caregiver Report Observation Mixed/Both All Vineland Adaptive Behavior Scale-IIa aRequires trained interviewer/observer. 

APPENDIX TABLE 4-6  Domain: Temperament Data-Gathering Method Instrument Type Caregiver Report Observation Mixed/Both All Toddler Behavior Assessment Questionnaire (Carey Scales) Children’s Behavior Questionnaire (Putnam and Rothbart, 2006) Infant Characteristics Questionnaire (Bates, Freeland, and Lounsbury, 1979) Pictorial Assessment of Temperament (PAT) (Clarke-Stewart et al., 2000) 83 

84 APPENDIX TABLE 4-7 Domain: Attachment/Caregiver-Child Interaction Data-Gathering Method Instrument Type Caregiver Report Observation Mixed/Both All Ainsworth Strange Situation Procedure Preschool Assessment of Attachment (Teti and Gelfand, 1997) Nursing Child Assessment Satellite Training