Ana Beatriz Rodrigues ReisI; Rosane Reis de MelloI; Denise Streit MorschII; Maria Dalva Barbosa Baker MeioI; Kátia Silveira da SilvaI
IInstituto Fernandes Figueira- Fundação Oswaldo Cruz, Rio de Janeiro, Brasil
IIPontifícia Universidade Católica (PUC), Rio de Janeiro, Brasil
The aim of this study was to investigate the stability of mental performance of very low birth weight premature infants during the first two years of life, and to identify factors associated with mental performance.The study included 109 children. The Mental Scale of Bayley Scales of Infant Development - Second Edition was administered at 6, 12 and 18-24 months of corrected age. The stability of the scores between assessments was verified by the analysis of variance for repeated measures.The association of the major social and neonatal characteristics with mental development was confirmed using multivariate analysis by linear regression, considering the following outcomes: mental development indices at 6 months, 12 months and between 18-24 months of corrected age. The mean Mental Developmental Index (MDI) was 83.4 (SD: 12.4) at 6 months, 86.4 (SD: 13.9) at 12 months, and 73.4 (SD: 14.5) at 18-24 months. A significant decrease in the mental developmental index (13 points) at 18-24 month corrected age was observed. The Mental development index did not show stability during the first two years of life in this population of preterm infants, except for children with neonatal pneumonia whose performance was unsatisfactory in all assessments. Among the risk factors investigated only male gender and neonatal pneumonia were associated with outcomes.
Keywords: Child development. Premature. Cognition. Follow-up studies. Infant. Psychomotor performance.
Growth and development are some of the concerns faced by parents and professionals who take care of preterm infants. The scientific and technological advances combined with the changes occurred in perinatal care have contributed to increase survival of newborns with increasingly reduced weight1. The increased survival rate of infants with ever more reduced gestational age draws greater interest concerning the influence of prematurity on child development, leading to increasing research about the uniqueness of this population. Biological risks, including organic immaturity, use of respirator, bronchopulmonary dysplasia, cerebral hemorrhage, septicemia, among others, and the socioeconomic and environmental conditions determine the prognosis of children born prematurely concerning their mental and motor development after birth2,3. Family income, family support, and maternal education are important factors that influence the development4.
Preterm infants with very low birth weight are more likely to develop cognitive and behavioral problems5-9. The dynamic characteristic of the development of premature babies, due to malleability of the neuropsychomotor development, calls for serial evaluations in the first years of life 3,10. The variability in mental performance was also observed in high-risk children, regardless of the gestational age, which reinforces the need for this follow-up 11-13.
There was a poor correlation among the results obtained from an assessment on the mental performance stability of a group including preterm infants, carried out in the first and second years of life. The authors suggest that specific high-risk groups may have different performances over two years, and that future research must be directed to other groups of children at high risk to provide information studies about development and to early intervention programs. 14
Given the dynamics of development, this paper suggests that there is stability in mental performance over the first two years in premature infants, especially in high- risk groups.
The purpose of this study was to assess mental performance stability in infants with very low birth weight in the Bayley Scale of Infant Development over the first two years of life and to identify factors associated with this performance at 6, 12, and 18-24 months .
A hospital-based prospective cohort study was held including 303 premature infants (gestational age less than 37 weeks) with very low birth weight (less than 1500 grams) born between January 2004 and January 2008. The infants were hospitalized in the Neonatal Intensive Care Unit and monitored in the Outpatient Clinic for Monitoring Newborns under Risk, and assessed for their development.
The study excluded 68 newborn infants with genetic syndromes or congenital malformations or infections, and 59 deaths in neonatal period. Moreover, it excluded four post-neonatal deaths and four children with bilateral blindness diagnosed in this period. Therefore, the total population under evaluation was composed of 167 children.
The Bayley Scale15 was applied to 139 children at 6 months old, to 126 children at 12 months old, and to 117 children between 18 and 24 months corrected age. The difference between the number of children in the initial population and the number of children who underwent the development test at certain ages is due to failure to attend appointments to perform the test. This study only considered children who completed the three assessments of development (at 6, 12, and between 18 and 24 months corrected age), and due to this criterion the population under study consisted of 109 children.
The neonatal and social characteristics of the children in the study were compared with those of loss group, and there were no statistically significant differences between them, suggesting the absence of selection bias.
All children received outpatient pediatric follow-up every month up to 12 months of corrected age, and every three months from 13 to 24 months. On the same day of the routine medical examination previously scheduled, the Bayley Scales of Infant Development-Second Edition (BSID-II)15 were applied by two trained psychologists, at 6, 12, and between 18 and 24 months corrected age. This study considered the BSID-II Mental Scale only, which is composed of 178 items (distributed in patterns according to age in months), which assess memory, habituation, problem solving, primitive concepts of numbers, generalizations, classification, vocalizations, language, and social skills. The performance achieved by the child in the BSID-II was measured using the Mental Development Index (MDI). The mean of this index is 100 and the standard deviation is 15. According to the scores obtained, the performance is classified as normal if equal or greater than 85, moderate delay if between 70 and 84, and serious delay if below 70. The psychologists who applied the scale were unaware of the children's clinical history, but they were aware that they were premature with very low birth weight. The interobserver reliability (between the two psychologists) measured by the intraclass correlation coefficient (ICC) was 0.94 (CI: 0.80 to 0.98).
Although Bayley Scales are considered the gold standard in assessing development in children up to 42 months, they were not standardized for the Brazilian population or to high-risk children. The performance classification recommended by the scale author and used in this study was a result from the standardization for the American population of low risk. Due to the need to compare it with other studies and lack of standardized scales for our population in this age group, we chose to classify the performance as proposed by the scale's author (normal, moderate delay, and serious delay), though cross-cultural differences are considered.
To describe the mental performance evolution, all index values of the mental development were converted for the performance rating (normal, moderate delay, severe delay) proposed by author of the scale15. Children with regular performance scored Mental Development Index equal or above 85, those with moderate delay scored MDI between 70 and 84, and those with severe delay scored below 70.
The following information were collected in the hospitalization period, and they were obtained by consulting hospital records and interviews with the infants' mothers: a) maternal data: age, number of pregnancies, parity, last menstrual period, estimated gestational age by obstetric ultrasonography conducted before 20 weeks gestation, obstetric complications, number of prenatal appointments, and use of antenatal corticosteroids; b) neonatal: sex, gestational age at birth (by somatic and neurological characteristics of the newborn), birth weight, type of delivery, Apgar scores, use of tracheal intubation in delivery room, resuscitation procedures, use of mechanical ventilation, duration of oxygen therapy, use of surfactant, patent ductus arteriosus, hyaline membrane disease, neonatal pneumonia, septicemia, use of oxygen therapy for over 28 days (bronchopulmonary dysplasia)16, and use of oxygen therapy with 36 weeks of corrected age (moderate/severe bronchopulmonary dysplasia) 16, peri-intraventricular hemorrhage, leukomalacia, hospitalization period. After release from the Neonatal Unit, the study obtained information about breastfeeding and social characteristics: family income, maternal education, father interaction, and daycare attendance.
The gestational age was estimated by last menstrual period, or in case of uncertainty about this date, by the obstetrical ultrasound performed up to 20 weeks of gestation. In the absence of both, the gestational age considered was the one obtained by the method described by Ballard17, which consists in the assessment of the somatic features and neurological evaluation of the newborn. The classification of weight adjustment to the gestational age was based on Alexander's intrauterine growth curve 18, which was considered small for the gestational age (SGA) when the birth weight was below the 10th percentile for gestational age, and infants born weighing between 10 and 90 percentiles were considered appropriate for the gestational age (AGA). The diagnosis of peri-intraventricular hemorrhage was obtained by serial cranial ultrasound exams up to 40 weeks of corrected age. The study considered light levels of hemorrhage (1 and 2) and severe ones (3 and 4), according to Papile's classification19. The septicemia was considered in the presence of positive blood cultures. The corrected age was calculated by subtracting, from the chronological age, the number of weeks missing necessary for the child to reach the term (40 weeks of gestational age).
A sample size of 109 children was estimated to detect an 8 points difference in the Bayley Scale between male and female children, assuming a standard deviation of 15 points, a=0.05, and power = 0.80.
The data were analyzed using STATA 10 (Stata Corp, 2007)20 and SPSS (SPSS for Windows, Inc, Chicago, Illinois, USA). The description of the main features of the population was made by frequency measures, mean, median, and standard deviation. The study calculated the prevalence of change in the Mental Development Index (MDI) at 6, 12, and 18 to 24 months of corrected age.
Comparisons were made to investigate the stability among MDI means in the three periods assessment using analysis of variance for repeated measures. The stability in the MDI was considered when there was no statistically significant difference among the averages in the three moments.
To compare the proportions according to the classification of altered and normal performance between 6 to 12 months, 12 and 24 months, and 6 and 24 months, the study used the Cochran's Q test. To identify among which age the differences could be observed in the assessments, the study performed the nonparametric McNemar test for dependent samples.
To verify the association of the main neonatal and social characteristics of mental development, the study conducted linear regression analysis considering as outcomes the Mental Development Index at 6 months, 12 months, and between 18-24 months of corrected age. The independent variables selected were those that presented p <0.20 in the bivariate analysis. The Stepwise strategy was used in the multivariate analysis to select the variables, and the level of significance for the Forward procedure was of 0.05, and for Backward procedure was 0.10. In the final model, the level of significance was 0.05.
The mean maternal age was 26 years old (SD = 7.0), and the maternal schooling level was 8 years of study for the 109 newborns studied. The mean household income was 988 Brazilian real. The antenatal corticosteroid therapy was used by 94% of these mothers. Table 1 presents the main characteristics of the population studied. The children remained hospitalized for an average period of 60 days.
The mean age of children in the first application of the Bayley Scale was 6.5 months (SD = 0.89), 12.6 months (SD = 1.08) in the second application, 23.2 months in the third one (SD = 3.32).
The mean Mental Development Index (MDI) at 6 months was 83.4 (SD: 12.4), at 12 months was 86.4 (SD = 13.9). The assessment carried out between 18-24 months of corrected age showed a decrease of 10 and 13 points, respectively, in the mental performance and a significant difference when compared to previous assessments. There were also differences in MDI means in the first two years of life for all subgroups analyzed, except for the subgroup of 13 children with neonatal pneumonia, who showed poor performance in the three assessments. Even for the subgroup of children without the risk factors analyzed, which are female and with no neonatal morbidity, there were differences in the MDI means. Therefore, there was no stability of MDI over time for most very low birth weight preterm infants. The mean MDI of the total population and of subgroups at 18-24 months ranged from 65.7 to 77.7 (Table 2).
In the unadjusted model, boys showed worse performance than girls did at 6 and 12 and 18-24 months. Children with bronchopulmonary dysplasia (BPD) showed the worst performance at 6 and 12 months compared to those without BPD. The same was true when comparing the performance of those with birth weight below and above 1000g. Children with neonatal pneumonia showed low mental performance in the three ages, and this difference was significant for the three ages compared to children who did not show such pathology (Table 3).
Fifty-five percent of children underwent breastfeeding with milk formula supplementation and the average time of mixed milk feeding was 4 months. At 24 months, the mean MDI of children who were breastfed was 72.6 (SD: 14.2), and those not breastfed 69.9 (SD = 14.7) (p = 0.35). There were also no significant differences in relation to means IDM at different ages evaluated when analyzing mother's education, living with the father, and day care attendance (data not shown in the table).
In the multivariate analyzes conducted with outcomes the Mental Development Index, the study found out that at 6 months of age, the association with the outcome took place with the neonatal pneumonia variable. At 12 months of age, the association was with neonatal pneumonia and male sex. At 18-24 months of corrected age, the association occurred only with the male sex variable (Table 3).
The MDI results converted to the mental performance classification are shown in Table 4, with a significant increase in the frequency of children with delay in the last assessment. To make this comparison, the study considered the grouping of moderate delay and severe delays. There was significant difference between the proportions of abnormal tests in the three ages assessed. The comparison of proportions of children with delays at 6 months and 18-24 months, and children at 12 months with 18-24 months showed significant statistically difference.
The assessments over the two years corrected age showed two groups of results in relation to performance. A group of children (n = 34) remained stable, as they kept the same classification of development in three the ratings: 19 (17.4%) remained under regular performance since the first assessment, 7 (6.4 %) remained under moderate delay, and 8 (7.3%) with serious delay since the first evaluation. Therefore, 75 (68.8%) of the children showed change in the mental performance classification. It is noteworthy the increase in the percentage of children with delay in the third assessment. Out of 49 children with abnormal performance (moderate and severe delays) at 6 months, after reaching 24 months, 45 of them were still classified as having developmental delay.
There was no stability in this studied population, if comparing the MDI means in the assessments held in the first two years of life, even for those subgroups of children that showed no neonatal morbidities. Moreover, the majority (76.1%) of the children reached, at 18-24 months, indexes lower than 85 in the mental development assessment using the Bayley Scale. In the adjusted analysis, only the sex was associated with the outcome, showing a worse performance for boys, with an 8.5 MDI average difference. Between 18-24 months, mean MDI values for all subgroups were below the cutoff of "normality" defined for infants born at term and of low risk. About two thirds of the group showed a change in the classification of development in the assessments, mainly due to a tendency of worse performance of children classified as having "normal development" at 6 months, if compared with the results at 18-24 months.
In Brazil, there is little academic research on mental performance along time of preterm infants13,21. Few international studies were conducted to investigate the stability of this population concerning mental performance11,14,22.
Bayley Scales assess each stage development level and their main feature is to enable the diagnosis of developmental delay and enable the planning of intervention strategies15.The measure of stability in repeated assessments is important to assess the progress of these children. Upon detecting any abnormality in the exam/test, the team that assists children in this study provides guidance to the relatives concerning the stimulation toward the cognitive, motor, and behavioral areas. The team also sends them to proper therapies. However, at two years old, the index obtained were still poor.
As mentioned before, there is not a standard scale for the Brazilian population or to preterm infants, so the results that classify mental performance must be interpreted with caution, as cross-cultural aspects may influence the attributes considered to classify the "normal development". The application of the Bayley Scale enables to compare the assessment of children' performance along time with other studies. The use of corrected age by two years is one of the recommended measures for partially minimize the effect of prematurity, approaching the development of preterm children to term. However, this study cannot state that adopting the cutoff used by the standardized scale for American children to term is appropriate for premature infants. Therefore, the possibility to have an "abnormal" overestimated frequency of performance is not ruled out.
The purpose of this study was not to assess the predictive value of the Bayley Scale, but to know the development pattern of the population studied along the first two years of life, using the mental performance dynamics in this scale. The BSID-II manual states that, over time, there is high correlation between the scores of children considered normal, but the manual does not include studies for high-risk populations15. The author reports that a two point increase is expected in the Scale score as age increases. These results reflect the stability in relation to children considered at low-risk, and are accordingly to the work of Harris et al.22 These authors reported a MDI increase from the first to the second assessment in children at low-risk (95 to 100.8). Still, with high-risk children, the reverse occurred, as the MDI decreased in the first assessment at 7 months for the second assessment at 18 months (85.7 to 82.3).
The stability of the Bayley Scales of children with Down syndrome and children with clinical problems (44% preterm infants), referred to an early intervention program, was evaluated in the first and second years of life. The correlation between these assessments using the BSID-II was considered moderate for the group with Down syndrome (r = 0.65), and poor for the second group (r = 0.37)14. The authors suggest that specific high-risk groups may have different performances over two years, and that future research must be directed to other high-risk children groups to provide information to studies about development and to early intervention programs.
Most studies detects a reduction of mean scores in the population of children with birth weight below 2000 g23, preterm and extremely low birth weight children24, with high risk11 (use of drugs during prenatal or perinatal complications), or with HPIV12, assessed by the Bayley Scale, either in the comparison of first year with the second11,12, or of second year with third year24. The deterioration of mental performance tends to be greater as greater the risk in prenatal and neonatal periods 11, which may also be associated with lower weight25. Regarding the MDI value, there was an average decrease of 10 points 11, 12. In our study, the occurrence of neonatal pneumonia showed similar results on the mental index, but it seems that this impact is more important in the first year of life.
In this study, the comparison of group of children weighing less than 1000 grams with those weighting higher showed a greater change in the Bayley Scale scores, with a declining tendency for both groups. The group with extremely low birth weight (weight<1000grams) showed a performance classified as delay since the first assessment compared to the group weighing more than 1000g, which showed such delay only in the third assessment, a finding similar to another study24.
Moreover, the lack of stability was mainly due to the significant decrease in the third assessment score (18-24 months), which suggests a cognitive function worsening of high-risk infants studied in this age group. There was a 13 points reduction in mental performance at 18-24 months corrected age, partly explained by the greater complexity of tasks performed in this age. Among the items analyzed, those related to language seem to be among the most affected ones in these premature infants, which possibly has a greater impact in the reduction of the mental development index.
Concerning the comparison between boys and girls, boys' history of mental development is more regular than girls, and a predictive ability of development at 6 months in relation to development at 18 and 36 months, not reported for girls25. This difference was due to the fact that areas of language are more accelerated in girls than in boys. One explanation would be the sex related genetic difference in the ability of the expressive language and that genetic, biological, and environmental factors may contribute to differences between boys and girls in language development25. In this study, among characteristics investigated such as factors associated with MDI, the male sex was the variable that remained associated in older ages after controlling for confounding variables.
There was no association between environmental and social factors and cognitive impairment in this population. This differs from literature: association between the highest level of maternal education and higher scores in intelligence tests12, and higher scores in assessments using BSID-II.26 Hack et al.27 observed that at two years of age, the environmental influences become more apparent and a critical transition begins to work with the development of skills related to symbolic functions, language development, and early formation of concept. Environmental factors that could be associated with mental performance were not included in this model, such as the stimulation of the child by the legal guardians, providing toys, organization of a supportive environment, and parental practices and styles. 28
Although there was a decline in performance observed in this study and in most studies with infants at risk, Procianoy et al.21 reached different results by comparing two groups of Brazilian children born with very low birth weight (one appropriate for gestational age and another small for gestational age), assessed with the BSID-II at 8, 12, 18, and 24 months corrected age. Both groups showed an increase in MDI scores from 8 to 24 months of corrected age, and in the last assessment, half of the population obtained a score classified as normal. This indicates a need for broader studies with premature infants in Brazil, investigating the characteristics of such delay for a better understanding of which it may be associated to. However, it is important to note that in this study, the majority (91.8%) of children with moderate and severe delays in the first assessment kept this level of performance over time, suggesting that the test results at 6 months may be useful to indicate early intervention.
In the cohort study, many children did not attend the three recommended assessments, but the social and neonatal characteristics of children from the loss group did not differ from children who remained in the study.
Bayley Scales are considered the gold standard 11 to assess the development, but have not been standardized for the Brazilian population yet, or to children born prematurely. Although several studies in the country apply this scale, the findings cannot be generalized to the Brazilian population, due to differences in relation to the U.S. population, where the scale was standardized. In this study, the result refers to a population at high risk and lower socioeconomic status, assisted in a public hospital of reference for fetal risk, so it cannot be generalized to the general population of newborns with very low weight.
The use of instruments is essential for mental development assessment. In Brazil, there is a notable lack of updated and standardized instruments that have been tested, translated, and adapted to our reality29.
For the premature infants with very low birth weight studied, the Mental Development Index in the three ages did not show stability, so there was a worse performance at 18-24 months. Except for the subgroup with neonatal pneumonia, whose MDI average remained stable with poor performance in the three assessments. The male children were most affected in the last two ages assessed.
Although the Bayley Scale is not standardized for the Brazilian or high risk population, the stability analysis of the mental index can be performed, as the comparisons were made by observing the means MDI of the same group of preterm infants at different ages. This study cannot assure that the performance rating used as normal/abnormal is appropriate, facing cross-cultural aspects and characteristics peculiar to preterm infants.
Accurate measures of child development in children at risk are necessary for early identification and intervention, so actions can be developed to guide families on how to stimulate the infants, sending them to specialized therapies to minimize the consequences of premature birth.
This work is part of a longitudinal, prospective study, approved by the Ethics Committee of Instituto Fernandes Figueira in 2004, and it was renovated in 2006 (CAAE 0066.0.008.000-03/ 0005.0.008.000-06). The legal guardians responsible for the children who participated in the cohort study signed an informed consent.
We declare that this work was free from connection or financing agreement between authors and companies or individuals who may have interests in the material discussed in this paper.
1. Wilson-Costello D, Friedman H, Minich N, Fanaroff AA, Hack M. Improved survival rates with increased neurodevelopmental disability for extremely low birth weight infants in the 1990s. Pediatrics 2005; 115(4): 997-1003.
2. Eickmann SH, Lira PIC, Lima MC. Desenvolvimento mental e motor aos 24 meses de crianças nascidas a termo com baixo peso. Arq Neuropsiquiat 2002; 60(3-B): 748-54.
3. Halpern R, Giugliani CGV, Barros FC, Horta BL. Fatores de risco para suspeita de atraso no desenvolvimento neuropsicomotor aos 12 meses de vida. J Pediatr 2000; 76(6): 421-8.
4. Santos LM, Santos DN, Bastos ACS, Assis AMO, Prado MS, Barreto ML. Determinants of early cognitive development: hierarchical analysis of a longitudinal study. Cad Saúde Pública 2008; 24(2): 427-37.
5. Méio MDBB, Lopes CS, Morsch DS. Fatores prognósticos para o desenvolvimento cognitivo de prematuros de muito baixo peso. Rev Saúde Pública 2003; 37(3): 311-8.
6. Chapieski ML, Evankovich KD. Behavioral effects of prematurity. Semin in Perinatol 1997; 21(3): 221-39.
7. Bordin MBM, Linhares MBM, Jorge SM. Aspectos cognitivos e comportamentais na média meninice de crianças nascidas pré-termo e com muito baixo peso. Psic Teor Pesq 2001; 17(1): 49-57.
8. Bhuta AT, Cleves AM, Casey PH, Cradock MM, Anand KJS. Cognitive and behavioral outcomes of school-aged children who were born preterm. JAMA 2002; 288 (6): 728-37.
9. Goto MMF, Gonçalves VMG, Netto AA, Morcillo AM, Moura-Ribeiro MVL. Neurodesenvolvimento de lactentes nascidos a termo pequenos para a idade gestacional no segundo mês de vida. Arq Neuropsiquiatr 2005; 63(1): 75-82.
10. Fraga DA, Linhares, MBM, Carvalho, AEV, Martinez, FE. Desenvolvimento de bebês prematuros relacionado a variáveis neonatais e maternas. Psicologia em Estudo 2008, 13(2): 335-44.
11. Koseck K, Harris SR. Changes in performance over time on the Bayley Scales of Infant Development-II when administered to infants at high risk of developmental disabilities. Pediatr Phys Ther 2004; 16(4): 199-205.
12. Ment LR, Vohr B, Allan W, Katz KH, Schneider KC, Westerveld M et al. Change in cognitive function over time in very low-birth-weight infants. JAMA 2003; 289(6): 705-11.
13. Silveira RC, Procianoy RS, Koch MS, Benjamin ACW, Schlindwein CF. Growth and neurodevelopment outcome of very low birth weight infants delivered by preeclamptic mothers. Acta Paediatr 2007; 96(12): 1738-42.
14. Niccols A, Latchman A. Stability of the Bayley mental scale of infant development with high risk infants. Br J Dev Disabil 2002; 48: 3-13.
15. Bayley N. Bayley scales of infant development. 2nd Ed. San Antonio: Harcourt; 1993.
16. Jobe AH, Bancalari E. Bronchopulmonary dysplasia. Am Rev Respir Crit Care Med 2001; 163(7): 1723-9
17. Ballard JL, Khoury JC, Wedig K, Wang L, Eilers-Ealsman BL, Lipp R. New Ballard score, expanded to include extremely premature infants. J Pediatr 1991; 119 (3): 417- 23.
18. Alexander GR, Himes JH, Kayman RB, Mor J, Kogan MA. United States national reference for fetal growth. Obstet Gynecol 1996; 87(2): 163-8.
19. Papile L, Burstein J, Burstein R, Koffler H. Incidence and evaluation of subependymal haemorrhage: a study of children with a birthweight less than 1500g. J Pediatr 1978; 92(4): 529-34.
20. StataCorp.2007. Stata Statistical Software: Release 10. College Station, TX. StataCorp LP.
21. Procianoy RS, Koch MS , Silveira RC. Neurodevelopmental outcome of appropriate and small for gestational age very low birth weight infants. J Child Neurol 2009; 24: 788-94.
22. Harris SR, Megens AM, Backman CL, Hayes VE. Stability of the Bayley II Scales of Infant Development in a sample of low-risk and high-risk infants. Dev Med Child Neurol 2005; 47(12): 820-23.
23. Halpern R, Barros AJD, Matijasevich A, Santos IS, Victora CG, Barros FC. Developmental status at age 12 months according to birth weight and family income: a comparison of two Brazilian birth cohorts. Cad Saúde Publica 2008; 24: S444-50.
24. Constantinou JC, Adamson-Macedo E, Mirmiran M, Ariagno RL, Feisher BE. Neurobehavioral assessment predicts differential outcome between vlbw and elbw preterm infants. J Perinatol 2005; 25: 788-93.
25. Lung FW, Shu BC, Ching TL, Chen PF, Lin LL. Predictive validity of Bayley scale in language development of children at 6-36 months. Pediatr Inter 2009; 51:666-9.
26. Wang L, Wang S, Huang C. Preterm infants of educated mothers have better outcome. Acta Paediatr 2008; 97: 568-73.
27. Hack M, Taylor G, Drotar D, Schluchter M, Cartar L, Wilson-Costelo 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 (2): 333-41.
28. Santos DN, Assis AMO, Bastos ACS, Santos LM, Santos CAST, Strina A et al. Determinants of cognitive function in childhood: a cohort study in middle income context. BMC Public Health 2008; 8: 202-16.
29. Duarte CS, Bordin IAS. Instrumentos de avaliação. Rev Bras Psiquiatr 2000; 22(SII): 55-8. Correspondência: Received: 23/05/11
Rosane Reis de Mello
Instituto Fernandes Figueira
Fundação Oswaldo Cruz
Rio de Janeiro, Brasil
Av. Rui Barbosa 716, Flamengo
Rio de Janeiro, CEP:22250-020
E-mail: email@example.com ou firstname.lastname@example.org
Final version: 04/11/11
Approved em: 09/12/11
Financiamento: PAPES IV- convênio FIOCRUZ/CNPq; Processo No. 400115/2006-9.
Acknowledgment: To Juliana Rocha Verçosa Delamonica by applying the Bayley Scales and to Raquel Ferreira de Carvalho for the assistance in the assessment of development of children participating in the research.