Original Article

Predictive Validity of the Bayley, Third Edition at 2 Years for Intelligence Quotient at 4 Years in Preterm Infants Michelle M. Bode, MD,*† Diane B. D’Eugenio, MA,† Barbara B. Mettelman, PhD,† Steven J. Gross, MD*† ABSTRACT: Objective: To determine the predictive validity of the Bayley Scales of Infant and Toddler Development, Third Edition (Bayley-III) at age 2 years for cognitive abilities in preschool children born at £30 weeks’ gestation. Methods: This prospective regional study included all 187 liveborn infants £30 weeks’ gestation born between July 2005 and June 2006. Of the 172 children who survived to 4 years, 156 (91%) were evaluated at 2 and 4 years. A socioeconomically matched term control group also was recruited to provide normative data. The predictive validity of the Bayley-III cognitive and language scales for the Weschler Preschool and Primary Scale of Intelligence-III (WPPSI-III) was examined through correlation coefficients and sensitivity and specificity of the Bayley-III to predict normal and abnormal cognitive outcomes. Results: Correlations of the WPPSI-III intelligence quotient (IQ) score with the Bayley-III cognitive and language scores were .81 and .78, respectively. The preterm children were classified as normal (Bayley Scales of Infant Development-Third Edition [BSID-III] cognitive score or WPPSI-III IQ score not lower than 1 SD below the control group mean), mild to moderately delayed (scores between 1 and 2 SD deviations below the control group mean), or severely delayed (scores greater than 2 SD below the control group mean). At 2 and 4 years, 126 (81%) preterm children retained the same developmental classification. Conclusions: In contrast with previous editions of the BSID, the Bayley-III has strong predictive validity for WPPSI-III IQ at age 4 years in preterm children. This has important implications for more timely evaluation of perinatal interventions, establishment of guidelines for neonatal care, and counseling parents. (J Dev Behav Pediatr 35:570–575, 2014) Index terms: Bayley-III, preterm infants, IQ, developmental outcomes.

A

dvances in perinatal and neonatal care have resulted in increased survival in preterm infants. As survival of premature infants increases, the question whether there is a concomitant increase in long-term neurologic and developmental morbidity has assumed great importance. Neurodevelopmental follow-up during the first 2 years of life allows timely evaluation of new perinatal treatment strategies and provides useful data for counseling parents both before and after preterm birth. However, preschool and school age follow-up are important to assess impact on cognition and adaptation to society. Ideally, neurodevelopmental assessments in high-risk infants should predict later cognitive outcomes. Unfortunately, this has not been the case.1–8 The mental developmental index (MDI) of the Bayley Scales of Infant Development-Second Edition (BSID-II)9 is the most widely used instrument to assess neurodevelopmental outcomes in the first 2 years of life.10,11 The Bayley was initially published in 1969 and revised in From the *Department of Pediatrics, Upstate Medical Center, SUNY, Syracuse, NY; †Department of Neonatology, Crouse Hospital, Syracuse, NY. Received May 2014; accepted August 2014. Disclosure: The authors declare no conflict of interest. Address for reprints: Michelle M. Bode, MD, Neonatal Associates of CNY, Suite 9100, Crouse Hospital, Syracuse, NY 13210; e-mail: [email protected]. Copyright Ó 2014 Lippincott Williams & Wilkins

570 | www.jdbp.org

1992.9,12 The original Bayley and BSID-II are poor predictors of later cognitive function.1,2,4,6,7,13,14 Both Hack et al1 and Roberts et al2 found that in the majority of cases, the results from the BSID-II at 2 years did not predict cognition at school age. Hack et al1 reported that in a population of extremely low birth weight infants, a subnormal MDI on the BSID-II at 20 months of age had ,50% predictive validity for both moderate and severe cognitive deficits at age 8 years. Roberts et al2 using developmental classifications derived from the BSID-II at age 2 years (normal, mild, moderate or severe delay) showed that only 46% of infants remained at the same classifications at 8 years. The BSID-II was significantly revised in 2006.15 The Bayley-III differs substantially from previous editions both in content and standardization methodology. The Bayley-III includes a separation of the original MDI into independent cognitive and language scales and the psychomotor index into a fine and gross motor scale. The standardization sample of the BSID-II was comprised of healthy full-term infants, whereas that of the Bayley-III included a more heterogeneous sample including children at risk for developmental delay (prematurity, birth asphyxia, etc.).15 The extent to which the Bayley-III predicts subsequent outcomes is unknown. The objective of the current study was to assess the predictive ability of the Bayley-III at age 2 years for subsequent Journal of Developmental & Behavioral Pediatrics

cognitive function at 4 years of age in infants born at #30 weeks’ gestation.

METHODS Study Subjects The study population included all liveborn infants #30 weeks’ gestation admitted to the NICU at Crouse Hospital during a 1-year period, (July 1, 2005–June 30, 2006). This hospital serves as the sole regional perinatal center for 17 counties in central New York; thus, this population represents a geographically based sample. Of note, 174 of the 187 liveborn preterm infants (93%) survived to hospital discharge. Two children subsequently died at 3 and 10 months of age. Of the 172 children alive at 4 years, 160 (93%) were evaluated at both 2 and 4 years of age. Scores were not obtainable for 4 children: 1 child was blind and 3 children were noncooperative for testing; these 4 children did not have neurologic abnormalities. Thus, 156 children (91% of survivors) constitute the study population. Table 1 provides perinatal and neonatal characteristics of the study population. Greater than 80% of infants were inborn secondary to a highly coordinated perinatal service; more than 50% of inborn babies followed maternal transport. The children had a mean birth weight of 1058 6 308 g and a mean gestational age of 27.5 6 1.8 weeks. One quarter of the children was from multiple births with over half of those pregnancies the result of assisted fertility. Cranial ultrasounds were performed on Table 1. Perinatal Characteristics and Neonatal Morbidities of the Study Group

all infants at 4 and 14 days of life to identify intraventricular hemorrhage and at 35 weeks’ postmenstrual age to identify periventricular leukomalacia. Of note, 81% of infants had normal head ultrasound findings. Neurodevelopmental outcomes at 2 years were reported previously for this cohort.16

Control Group To interpret outcome in preterm infants, we recruited a socioeconomically matched control group. Full term, appropriately grown and healthy neonates born at the regional center were identified during the same time period and matched 1-to-1 with preterm infants for the following characteristics: gender, race, maternal age (,18, 18–34, and $35 years), years of formal education (,12, 12–15, and $16), and marital status. Assessments performed on the control group were used to establish norms for the preterm group. Preterm and term groups were similar for gender (56% and 54% males, respectively) and were tightly matched for all maternal characteristics (Table 2).

Follow-up Assessments All infants were followed serially as part of the high-risk follow-up program. Infants were evaluated at 24 months of age and again at age 4 years, using age adjusted for prematurity. At both times, a neurologic examination and developmental testing were performed. At age 2 years, neurologic outcome was assessed using the Infant Neurological International Battery.17 At age 2 and 4 years, Table 2. Maternal Socioeconomic Characteristics of Study Groups

N 5 156, n (%) Inborn

Preterm Group Term Group (N 5 156), n (%) (N 5 155), n (%)

125 (80) Maternal Age, y

Gestational age, wk Mean

27.5

SD

1.8

,27 wk

45 (29)

Mean

28

28

SD

7

6

,18

8 (5)

7 (5)

127 (82)

125 (80)

21 (13)

23 (15)

Mean

12

13

SD

3

3

18–35

Birth weight, g Mean

.35

1058

SD

Maternal Education, y

308

,750 g

28 (18)

Small for gestational age

12 (8)

Male gender

87 (56)

Multiple birth

40 (26)

Mechanical ventilation

122 (78) a

Bronchopulmonary dysplasia

88 (56)

Sepsis

39 (25)

Necrotizing enterocolitis Severe cranial ultrasound abnormalityb aSupplemental

3 (2) 14 (9) bGrade

oxygen requirement at 36 weeks’ postmenstrual age. 3 or 4 intraventricular hemorrhage, PVL, or persistent ventriculomegaly at 35 weeks’ postmenstrual age.

Vol. 35, No. 9, November/December 2014

,12

56 (36)

50 (32)

12–15

73 (47)

78 (50)

.15

27 (17)

27 (17)

69 (44)

73 (47)

White

121 (78)

126 (81)

Black

23 (15)

23 (15)

Hispanic

7 (4)

4 (3)

Other

5 (3)

3 (2)

Married Race

© 2014 Lippincott Williams & Wilkins

571

cerebral palsy (CP) was classified using the Gross Motor Function Classification System for Cerebral Palsy.18 For a child with increased muscle tone, this instrument defines mild CP (level 1 or 2) when the child’s gross motor skills are minimally affected (i.e., child is able to walk without assistance, moves from floor to chair or standing with minimal assistance, or has hands free in sitting to manipulate objects). Severe CP (level 3, 4, or 5) is defined when a child’s self mobility is limited requiring the use of assistive devices both for mobility and sitting. Deafness was defined as the need for bilateral hearing aides. Blindness was defined as no vision in both eyes. Severe neurologic abnormality was defined as severe CP, hypotonia interfering with ambulation, deafness, or blindness. The Bayley-III cognitive, motor, and language scales were administered at age 2 years. The cognitive scale is comprised largely of nonverbal activities evaluating sensorimotor development, object relatedness, memory, manipulation, and problem solving. The language scale is composed of 2 subtests: receptive communication (measuring preverbal behaviors through verbal comprehension) and expressive communication (measuring preverbal communication to verbal communication skills) subtests. The motor scale is composed of fine motor and gross motor subtests. Each composite scale has a mean of 100 and a SD of 15.15 The WPPSI-III was administered at age 4 years. All 14 subtests were administered, 7 of which provide a full scale intelligence quotient (IQ) (mean of 100 and SD of 15).19 Children unable to complete the cognitive and/or language scale of the Bayley-III or the subscales comprising IQ of the WPPSI because of severe delay were assigned scores of 50. The developmental testers were blinded to childrens’ gestational age and all perinatal factors. Testers at 4 years were blinded to results of the 2-year evaluation. Outcomes in preterm children at age 2 years were reported using the Bayley-III cognitive scale and the Bayley-III language scale. Since language ability has traditionally been an important part of cognitive assessment in infancy, we also used a combined Bayley-III score by averaging the cognitive and language scales.20 Outcomes at 4 years were derived from WPPSI-III full scale IQ. Outcomes were classified as normal (score no lower than 1 SD below the mean of the control group); mild to moderately delayed (score between 1 and 2 SD below the mean of the control group), or severely delayed (score greater than 2 SD below the mean of the control group). The ability to predict 4-year scores from 2-year scores was assessed with correlation coefficients and stability of cognitive classification over time. Sensitivity and specificity of the Bayley-III for 4-year IQ were calculated. A secondary analysis compared the ability to predict outcome in the most immature infants (,27 weeks’ gestational age) with those born at later gestational age (27–30 weeks).

Statistical Analysis Data were analyzed using Stata 10 and Statview. Continuous variables are expressed as mean 6 SD. 572 Improved Prediction of IQ by Bayley-III

Differences between preterm and term infants were tested for significance by the unpaired Student’s t test or the Mann-Whitney U test. The x2 test is used to compare categorical variables. Cohen’s kappa was used to examine stability of cognitive categorization over time. Cohen’s kappa values approaching 1 indicate stable categorization over time.21 Pearson’s correlation coefficients were generated to evaluate the linear relationship between the Bayley-III and WPPSI-III. Correlation coefficients greater than .7 are considered highly significant.22 The study was approved by Crouse Hospital’s Human Research Review Committee, and informed consent was obtained from the parent(s) of each child.

RESULTS Neurologic Outcome Ten preterm children (6%) had major neurologic abnormalities at both 2 years and 4 years. Six children had severe cerebral palsy (CP) (including 1 child who was also blind), and 3 children had severe truncal hypotonia interfering with independent sitting and ambulation. Additionally, 1 child was deaf. All children diagnosed at age 2 with a major neurologic abnormality retained the same diagnosis at age 4 years.

Control Norms The control group had a mean Bayley-III cognitive scale score of 97 6 10, therefore, defining normal cognition for preterm infants as a cognitive scale score $87 (no lower than 1SD below the mean of the control group); mild to moderate delay as a cognitive score between 77 and 86, and severe delay as a score ,77. The control group had a mean Bayley-III language scale score of 100 6 13, defining normal, mild to moderate delay, and severe delay in the preterm group as scores $87, 74 to 86, and ,74, respectively. At 4 years, the control group demonstrated a mean WPPSI-III intelligence quotient (IQ) score of 102 6 16.

Cognitive Outcome in Preterm Children At 2 years, the preterm children had mean Bayley-III cognitive (92 6 14) and language (93 6 17) scores that were significantly lower than those of the control group, p , .001. At 4 years, the preterm children continued to score on average lower than the term group (WPPSI-III IQ score of 97 6 20 vs 102 6 16, respectively, p 5 .04). Despite this, .70% of preterm children had cognitive and language scores in the normal range at age 2 and 4 years, and only 11% demonstrated severe cognitive delay (Table 3).

Predictive Validity of 2-year Outcome at 4 Years The correlations of WPPSI-III IQ at age 4 years with the Bayley-III cognitive and language scales at age 2 years are shown in Figure 1. The correlation coefficient for the cognitive scale was .81 and for the language scale Journal of Developmental & Behavioral Pediatrics

Table 3. Classification of Cognitive Outcome in the Preterm Children

Normal

BSID-III Cognitive Scale at 2 yr (N 5 156) n (%)

BSID-III Language Scale at 2 yr (N 5 154) n (%)

WPPSI IQ at 4 yr (N 5 156) n (%)

111 (71)

110 (71)

122 (78)

Mild-tomoderate delay

28 (18)

26 (17)

17 (11)

Severe delay

17 (11)

18 (12)

17 (11)

Normal is a score no lower than 1 SD below the mean of the control group; mild to moderately delayed is a score between 1 and 2 SD below the mean of the control group; severe delay is a score .2 SD below the mean of the control group.

was .78. There was a modest improvement in the correlation coefficient to .83 using the combined Bayley-III score. The relationships between Bayley-III scale scores and the WPPSI-III IQ were greater for infants born at ,27 weeks’ gestation (cognitive score R 5 .84 and language score R 5 .84) than for those born at $27 weeks’ gestation (R 5 .76 and R 5 .73, respectively, p , .05). The correlation between the Bayley-III and WPPSI-III was

weakest in the full-term control group (cognitive score R 5 .63 and language score R 5 .67). Sensitivity and specificity of the Bayley-III cognitive, language, and combined scores for any delay and for severe delay at 4 years were calculated (Table 4). Sensitivity of the Bayley-III cognitive score for any delay (i.e., the ability of a score .1 SD below the mean at 2 years to predict a similar IQ at 4 years) was greater than 80%. Sensitivity for severe delay (ability of a score greater than 2 SD below the mean to predict a similar score at 4 years) was greater than 70%. The addition of the language scale did not affect sensitivity for any delay but did improve sensitivity for severe delay (Table 4). Specificity (i.e., the ability to rule out cognitive delay at age 4 with a normal score at age 2) was extremely high (.80%) when using any of the scale scores. When preterm children were classified as normal, mild to moderately delayed, or severely delayed, 81% of the children remained in the same developmental classification at age 2 and 4 years. One hundred six children (68%) were classified as normal, 8 (5%) as mild to moderately delayed, and 12 (8%) as severely delayed at both evaluations. Of the 30 children who changed classification, the Bayley-III underestimated outcome in the majority of cases. Two thirds of the children who changed classification between 2 and 4 years of age moved into an improved classification. Fifteen children who were classified as mild to moderately delayed and 1 child who was severely delayed at age 2 were classified as normal at age 4 years. The severe delay classification based on the combined score had the strongest kappa value (.82), whereas the classification based on any delay had kappa values ranging from .54 to .65, for language and cognitive scores respectively.

DISCUSSION To our knowledge, this is the first study to report developmental outcome for a regional cohort of preterm infants using the Bayley-III and to compare those outcomes with long-term cognitive functioning. Our results Table 4. Prediction of the BSID-III at Age 2 years on 4-year Intelligence Quotient in Preterm Infants Positive Negative Sensitivity, Specificity, Predictive Predictive % % Value, % Value, % Any delay Cognitive

85

87

64

95

Language

76

84

57

93

Combined

85

87

61

97

Cognitive

71

96

71

96

Language

76

96

72

97

Combined

82

97

76

99

Severe delay

Figure 1. The relationship between BSID-III cognitive scores (A) and language scores (B) at 2 years and intelligence quotient at 4 years of age in preterm children. Vol. 35, No. 9, November/December 2014

Any delay is a score more than 1 SD below the mean of the control group, and severe delay is a score more than 2 SD below the mean of the control group. © 2014 Lippincott Williams & Wilkins

573

show that the Bayley-III administered at 24 months of age to extremely preterm infants is a good predictor of cognitive function at 4 years. There was a high correlation between Bayley-III cognitive and language scale scores and subsequent intelligence quotient (IQ), such that we were able to accurately classify developmental outcome as normal, mild to moderately delayed, or severely delayed at 4 years based on classification at 2 years in more than 80% of the children. Our results are in contrast with previous studies using earlier versions of the Bayley Scales of Infant Development (BSID). In extremely preterm infants, the BSID-II mental developmental index (MDI) overestimated the number of children with moderate or severe cognitive disability at later school age.1,2 In fact, only half of the infants identified with cognitive impairment at 20 months of age were so classified at 8 years.1 The change in developmental classification from 20 months (based on MDI) to school age IQ was so great that results were felt to be unreliable endpoints for follow-up. Our results are more encouraging suggesting that infant testing with the Bayley-III may be of greater value in making long-term predictions. The administration of the Bayley-III at 24 months age adjusted for prematurity, later than the age of 20 months in study by Hack et al,1 might also have influenced the predictive value. A critical period in cognitive development occurs between 18 and 24 months during which skills in symbolic function, language, and early concept formation emerge. Furthermore, the effects of socioeconomic variables become increasingly apparent.23 Therefore, when testing is performed in later infancy, we might expect improved prediction of outcome. Indeed, our own data support this concept. The preterm children in this study were assessed previously at 15 months of age with the Bayley-III.16 At the earlier age, the correlation with 4-year outcome was far poorer than it was at 24 months of age. The correlation coefficients for the cognitive and language scale scores administered at 15 months for 4-year WPPSI-III IQ were .68 and .57, respectively, as compared with .81 and .78 for the same scales administered at 24 months. Furthermore, the improved predictive validity of the Bayley-III for later cognitive function may be due to a number of factors. It is possible that the Bayley-III is a better instrument than its predecessors, more accurately capturing a child’s performance. The latest revision was introduced recently to provide updated normative data and to improve test structure.24 There were also changes in administrative guidelines. In older versions of the Bayley Scales of Infant Development (BSID), different scores could be obtained depending on the selection of starting point of testing. Starting at a lower age level would result in a lower score than starting at a higher level because starting at the lower level may not allow infants to earn points on a higher item set.25 The Bayley-III allows for crediting an item from an earlier administered scale if the child spontaneously performs that item later in testing. 574 Improved Prediction of IQ by Bayley-III

Interpretation of outcomes in high-risk infants using the Bayley-III has generated considerable controversy.26,27 A major issue is higher scores and therefore underestimation of developmental delay when Bayley-III norms are used.11 This is due to inclusion in the standardization of sample infants with known disabilities who were at risk for developmental delay.15 This resulted in inflated norms. For example, Moore et al20 administered the Bayley-II MDI and the Bayley-III cognitive and language scales concurrently in 186 extremely preterm children at 29 to 41 months of age. Bayley-III cognitive scores averaged 10 points higher than the Bayley-II MDI and Bayley-III language scores averaged 3 points higher.20 Bayley-III scores were increasingly higher at lower MDI scores, thus underestimating rates of severe delay.20 Therefore, it is suggested that the cutoff for identifying any developmental delay and severe delay be revised upward from the traditional 1 and 2 SD below the test mean; that is, scores of 85 and 70, respectively.11,20,28,29 Strength of our study is the inclusion of a socioeconomically matched control group. A control group is important to decrease effects of confounding variables to minimize relying on test norms derived from a different population and to provide a contemporary reference group for comparing outcome over time using different testing instruments.30 It is noteworthy that our control group had similar mean cognitive and language scores to those of the Bayley-III standardization sample. However, the variances in scores in our control group were smaller (mean 6 SD for cognitive scale, 97 6 10 and language scale, 100 6 13). As a result, a cutoff cognitive score of 77 (2 SD below control mean) was indicative of severe developmental delay rather than a score of 70 based on test norms. Our cutoff is consistent with the thought that cognitive delay in high-risk infants should increase with the introduction of new standardized tests due to the overall increase in IQ scores over time (Flynn effect).31 There are multiple factors that impact child development including family structure and stability.32 Over one third of the preterm infants in our study received early intervention services during the first 2 years of life and over a quarter of the infants continued to receive services at 4 years. Our study did not assess the impact of these services. We acknowledge that our childhood follow-up was early. Future studies need to address the ability of the Bayley-III administered at age 2 years to predict school age cognitive outcomes.

CONCLUSIONS These results suggest that developmental outcomes at 2 years may be a more reliable endpoint for assessing outcomes in preterm infants than previously thought. This would allow more timely evaluations of effectiveness of therapeutic perinatal and neonatal interventions and provide information about long-term prognosis to families. Journal of Developmental & Behavioral Pediatrics

REFERENCES 1. Hack M, Taylor G, Drotar D, et al. Poor predictive validity of the Bayley Scales of Infant Development for cognitive function of extremely low birth weight children at school age. Pediatrics 2005; 116:333–341. 2. Roberts G, Anderson PJ, Doyle LW; the Victorian Infant Collaborative Study Group. The stability of the diagnosis of developmental disability between ages 2 and 8 in a geographic cohort of very preterm children born in 1997. Arch Dis Child. 2010;95:786–790. 3. McCormick MC. The outcomes of very low birth weight infants: are we asking the right questions? Pediatrics 1997;99:869–876. 4. Ment LR, Vohr B, Allan W, et al. Change in cognitive function over time in very low-birth-weight infants. JAMA. 2003;289:705–711. 5. Kitchen WH, Rickards AL, Ford GW, et al. Selective improvement in cognitive test scores of extremely low birthweight infants aged between 2 and 5 years. Aust Paediatr J. 1989;25:288–291. 6. Harris SR, Langkamp DL. Predictive value of the Bayley mental scale in the early detection of cognitive delays in high-risk infants. J Perinatal. 1994;14:275–279. 7. Potharst ES, Houtzager BA, Van Sonderen L, et al. Prediction of cognitive abilities at the age of 5 years using developmental followup assessments at the age of 2 and 3 years in very preterm children. Dev Med Child Neurol. 2012;54:240–246. 8. Synnes AR, Anson S, Arkesteijn A, et al. School entry age outcomes for infants with birth weight #800 grams. J Pediatr. 2010;57:989–994. 9. Bayley N. The Bayley Scales of Infant Development. San Antonio, TX: The Psychological Corporation; 1969. 10. Lorenz JM, Wooliever DE, Jetton JR, et al. A quantitative review of mortality and developmental disability in extremely premature newborns. Arch Pediatr Adolesc Med. 1998;152:425–435. 11. Anderson PJ, DeLuca CR, Hutchinson E, et al. Underestimation of developmental delay by the new Bayley-III scale. Arch Pediatr Adolesc Med. 2010;164:352–356. 12. Bayley N. The Bayley Scales of Infant Development-II. San Antonio, TX: The Psychological Corporation; 1993. 13. Slater A. Individual differences in infancy and later IQ. J Child Psychol Psychiatry. 1995;36:69–112. 14. McGrath E, Wypij D, Rappaport LA, et al. Prediction of IQ and achievement at age 8 years from neurodevelopmental status at age 1 year in children with d-transposition of the great arteries. Pediatrics. 2004;114:e572–e576. 15. Bayley N. Bayley Scales of Infant and Toddler Development. 3rd ed. San Antonio, TX: The Psychological Corporation; 2006.

Vol. 35, No. 9, November/December 2014

16. Bode MM, D’Eugenio DB, Forsyth N, et al. Outcome of extreme prematurity: a prospective comparison of 2 regional cohorts born 20 years apart. Pediatrics. 2009;124:866–874. 17. Ellison PH, Horn J, Browning C. Construction of an infant neurological international battery (INFANIB) for the assessment of neurological integrity in infancy. Phys Ther. 1985;65:1326–1331. 18. Palisano R, Rosenbaum P, Walter S, et al. Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol. 1997;39:214–223. 19. Wechsler D. Weschler Preschool and Primary Scale of Intelligence-Third Edition (WPPSI-III). San Antonio, TX: NCS Pearson, Inc; 2002. 20. Moore T, Johnson S, Haider S, et al. Relationship between test scores using the second and third editions of the Bayley scales in extremely preterm children. J Pediatr. 2012;160:553–558. 21. Fleiss JL, Levin B, Paik MC. Statistical Methods for Rates and Proportions. 3rd ed. Hoboken, NJ: John Wiley & Sons; 2003. 22. Taylor R. Interpretation of the correlation coefficient: a basic review. J Diagn Med Sonogr. 1990;1:35–39. 23. Aylward GP. The relationship between environmental risk and developmental outcome. J Dev Behav Pediatr. 1992;13:222–229. 24. Factors Contributing to Differences Between Bayley-III and BSIDII Scores. Bayley-III Technical Report. San Antonio, TX: Pearson Education; 2008. 25. Gauthier SM, Bauer CR, Messinger DS, et al. The bayley scales of infant development II: where to start? J Dev Behav Pediatr. 1990; 30:75–79. 26. Aylward GP, Aylward BS. The changing yardstick in measurement of cognitive abilities in infancy. J Dev Behav Pediatr. 2011;32:465–468. 27. Aylward GP. Continuing issues with the Bayley-III: where to go from here. J Dev Behav Pediatr. 2013;34:697–701. 28. Acton BV, Biggs WSG, Creighton DE, et al. Overestimating neurodevelopment using the Bayley-III after early complex cardiac surgery. Pediatrics. 2011;128:e794–e800. 29. Vohr BR, Stephens BE, Higgins RD, et al. Are outcomes of extremely preterm infants improving? Impact of Bayley assessment on outcomes. J Pediatr. 2012;161:222–228. 30. Gross SJ, Slagle TA, D’Eugenio DB, et al. Impact of a matched control group on interpretation of developmental performance in preterm infants. Pediatrics. 1992;90:681–687. 31. Flynn JR: Searching for justice. The discovery of IQ gains over time. Am Psychol. 1999;54:728–748. 32. Gross SJ, Mettelman BB, Dye TD, et al. Impact of family structure and stability on academic outcome in preterm children at 10 years of age. J Pediatr. 2001;138:169–175.

© 2014 Lippincott Williams & Wilkins

575

Predictive validity of the Bayley, Third Edition at 2 years for intelligence quotient at 4 years in preterm infants.

To determine the predictive validity of the Bayley Scales of Infant and Toddler Development, Third Edition (Bayley-III) at age 2 years for cognitive a...
165KB Sizes 0 Downloads 13 Views