Abstract
Early-life exposures are critical for later child cognitive development. McCarthy Scales of Children's Abilities (MSCA) were used to assess cognitive development of 700 preschoolers (Mage = 4.2 years), derived from the “Rhea” birth cohort, in Greece. Principal component analysis (PCA) was applied on prospectively collected exposure data. Six components were extracted; five of them were associated with child cognition. Higher parental social status, preschool attendance and less TV watching, nonsmoking during pregnancy and breastfeeding, and parental involvement in child life were protective factors of child cognition at 4 years. Increased child birth order was negatively associated with child cognition. Offspring's size at birth was not associated with any cognitive outcome. These findings reveal the importance of early-life exposures to child cognitive development.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 1063-1073 |
| Number of pages | 11 |
| Journal | Child Development |
| Volume | 89 |
| Issue number | 4 |
| DOIs | |
| State | Published - Jul 1 2018 |
ASJC Scopus subject areas
- Pediatrics, Perinatology, and Child Health
- Education
- Developmental and Educational Psychology
Access to Document
Other files and links
Fingerprint
Dive into the research topics of 'Patterns of Early-Life Social and Environmental Exposures and Child Cognitive Development, Rhea Birth Cohort, Crete, Greece'. Together they form a unique fingerprint.Cite this
- APA
- Standard
- Harvard
- Vancouver
- Author
- BIBTEX
- RIS
In: Child Development, Vol. 89, No. 4, 01.07.2018, p. 1063-1073.
Research output: Contribution to journal › Article › peer-review
}
TY - JOUR
T1 - Patterns of Early-Life Social and Environmental Exposures and Child Cognitive Development, Rhea Birth Cohort, Crete, Greece
AU - Kampouri, Mariza
AU - Kyriklaki, Andriani
AU - Roumeliotaki, Theano
AU - Koutra, Katerina
AU - Anousaki, Despoina
AU - Sarri, Katerina
AU - Vassilaki, Maria
AU - Kogevinas, Manolis
AU - Chatzi, Leda
N1 - Funding Information: The authors would like to thank all Rhea mother–child cohort participants for their generous collaboration. The “Rhea” project was financially supported by the European projects (EU FP6-2003-Food-3-NewGeneris, EU FP6. STREP Hiwate, EU FP7 ENV.2007.1.2.2.2, Project No. 211250 Escape, EU FP7-2008-ENV-1.2.1.4 Envirogenomarkers, EU FP7-HEALTH-2009-single stage CHICOS, EU FP7 ENV.2008.1.2.1.6, Proposal No. 226285 ENRIECO) and the Greek Ministry of Health (Program of prevention of obesity and neurodevelopmental disorders in preschool children, in Heraklion District, Crete, Greece: 2011–2014; “Rhea Plus”: Primary prevention program of environmental risk factors for reproductive health, and child health: 2012–2015). Funding Information: Children's cognitive and motor development at 4?years was assessed through the McCarthy Scales of Children's Abilities (MSCA; MacCarthy,). The MSCA represent an age-appropriate instrument, developed for children of ages 2??8??years, which assesses children's present cognitive development to identify possible developmental delay in different domains. The MSCA contain 18 subscales, which provide six scales: (a) verbal scale (verbal expression and comprehension), (b) perceptual performance (reasoning), (c) quantitative scale (numerical aptitude and interest), (d) general cognitive scale (overall cognitive ability), (e) memory (verbal and nonverbal short-term memory), and (f) motor scale (gross and fine motor ability). The MSCA were administered individually to the participants by two trained psychologists. The participants were randomly distributed between them. The interobserver variability was <?1%. At the end of the neuropsychological assessment the examiners completed a standard form regarding the assessment's conditions used to evaluate the ?quality of assessment? (excellent, bad, very bad). Families received detailed feedback on their children's performance. The translation and cross-cultural adaptation of the MSCA were performed according to the internationally recommended methodology. Internal consistency (Cronbach's alpha) varied between ra?=.76 and ra?=.89, showing adequate reliability for all the scales. Confirmatory factor analysis was performed using AMOS software (v.22, IBM SPSS Statistics, Armonk, NY) and maximum likelihood estimation method. A model with five correlated latent variables was confirmed, and the confirmatory factor analysis indices supported good fit of the model (?2/df?=?2, comparative fit index?=.83, goodness-of-fit index?=.97, root mean square error of approximation?=.034). Raw scores of the MSCA were standardized for child's age at test administration using a parametric method for the estimation of age-specific reference intervals (Royston & Wright,). The parameters of the distribution were modeled as fractional polynomial functions of age and estimated by maximum likelihood. Standardized residuals were then typified having a mean of 100 points with an SD?=?15 to homogenize the scales. Children's cognitive and motor development at 4?years was assessed through the McCarthy Scales of Children's Abilities (MSCA; MacCarthy,). The MSCA represent an age-appropriate instrument, developed for children of ages 2??8??years, which assesses children's present cognitive development to identify possible developmental delay in different domains. The MSCA contain 18 subscales, which provide six scales: (a) verbal scale (verbal expression and comprehension), (b) perceptual performance (reasoning), (c) quantitative scale (numerical aptitude and interest), (d) general cognitive scale (overall cognitive ability), (e) memory (verbal and nonverbal short-term memory), and (f) motor scale (gross and fine motor ability). The MSCA were administered individually to the participants by two trained psychologists. The participants were randomly distributed between them. The interobserver variability was <?1%. At the end of the neuropsychological assessment the examiners completed a standard form regarding the assessment's conditions used to evaluate the ?quality of assessment? (excellent, bad, very bad). Families received detailed feedback on their children's performance. The translation and cross-cultural adaptation of the MSCA were performed according to the internationally recommended methodology. Internal consistency (Cronbach's alpha) varied between ra?=.76 and ra?=.89, showing adequate reliability for all the scales. Confirmatory factor analysis was performed using AMOS software (v.22, IBM SPSS Statistics, Armonk, NY) and maximum likelihood estimation method. A model with five correlated latent variables was confirmed, and the confirmatory factor analysis indices supported good fit of the model (?2/df?=?2, comparative fit index?=.83, goodness-of-fit index?=.97, root mean square error of approximation?=.034). Raw scores of the MSCA were standardized for child's age at test administration using a parametric method for the estimation of age-specific reference intervals (Royston & Wright,). The parameters of the distribution were modeled as fractional polynomial functions of age and estimated by maximum likelihood. Standardized residuals were then typified having a mean of 100 points with an SD?=?15 to homogenize the scales. Information on children's ADHD difficulties was gathered through the ADHD test (ADHDT; Gilliam,), which was completed by the participants? mothers. The questionnaire is based on Diagnostic and Statistical Manual of Mental Disorders (4th ed.) criteria and designed to identify and evaluate ADHD symptoms at ages 3?23?years. It includes 36 items, providing three subscales (hyperactivity, inattention, and impulsivity) and a total ADHD difficulties index. The ADHDT has been translated and adapted for the Greek population (Maniadaki & Kakouros,). Of the 1,363 live singleton births, 875 children participated in the 4-year follow-up, 18 children were excluded due to neurodevelopmental disorder or other medical condition (i.e., plagiocephaly, microcephaly, hydrocephaly, brain tumor) and 8 children due to incomplete neuropsychological assessment, resulting in a cohort of 849 children with valid outcome data at 4?years. Of the 849 children with valid data on cognitive development, 700 had available data on multiple early-life exposures (Mage?=?4.2, SD?=?0.20). Of these participants, 358 were male (Mage?=?4.2, SD?=?0.19) and 342 female (Mage?=?4.2, SD?=?0.21). Of the 700 participants, 610 provided data on ADHD difficulties, of which 300 were female and 310 male. The total of the 700 children were Greek speaking, and the majority of them resided in urban centers (71%). Almost half of the children were the only or the firstborn child. The vast majority of the participants (85.7%) attended preschool at 4?years and used to spend their free time at weekends with both their parents (84.9%). Participating parents were mainly Greek (fathers 97% and mothers 87%), married at the time of the pregnancy, and living in urban areas, and almost all fathers were employed, similarly to half of the mothers. Half of the mothers had completed postsecondary education, likewise 43% of the fathers. We have compared the demographic characteristics at baseline between participants and nonparticipants at the 4-year follow-up (Table?S1). Parents of children who participated were more likely to be older, more educated, and married at child birth, in comparison with nonparticipants. No statistically significant differences have been identified between the excluded participants due to neurodevelopmental disorder diagnosis and nonexcluded children, in relevance with the exposure variables, with the exception of breastfeeding duration and preschool attendance at 4?years, which can be attributable to the specific difficulties these children face (Tables and S2). The ?Rhea study? is an ongoing prospective birth cohort based in Heraklion, Crete. Pregnant women, living in the prefecture of Heraklion, were recruited within 1?year (February 2007?January 2008; Chatzi et?al.,). The inclusion criteria were residence in the study area, good understanding of Greek, and maternal age above 16?years. The first contact with the families was made at the 15th gestational week (first major ultrasound examination) and the second at the 24th gestational week. The participants were invited again at child follow-up assessments when the children were 9 and 18?months old and 4?years old. Participants? assessments, at the 4-year follow-up, took place at the University Hospital of Heraklion or at health centers for the families residing in rural areas. The study was conducted according to the guidelines of the Declaration of Helsinki, and all procedures were approved by the ethical committee of the University Hospital in Heraklion. Written informed consent was obtained from the participants of the study. Of the 1,363 live singleton births, 875 children participated in the 4-year follow-up, 18 children were excluded due to neurodevelopmental disorder or other medical condition (i.e., plagiocephaly, microcephaly, hydrocephaly, brain tumor) and 8 children due to incomplete neuropsychological assessment, resulting in a cohort of 849 children with valid outcome data at 4?years. Of the 849 children with valid data on cognitive development, 700 had available data on multiple early-life exposures (Mage?=?4.2, SD?=?0.20). Of these participants, 358 were male (Mage?=?4.2, SD?=?0.19) and 342 female (Mage?=?4.2, SD?=?0.21). Of the 700 participants, 610 provided data on ADHD difficulties, of which 300 were female and 310 male. The total of the 700 children were Greek speaking, and the majority of them resided in urban centers (71%). Almost half of the children were the only or the firstborn child. The vast majority of the participants (85.7%) attended preschool at 4?years and used to spend their free time at weekends with both their parents (84.9%). Participating parents were mainly Greek (fathers 97% and mothers 87%), married at the time of the pregnancy, and living in urban areas, and almost all fathers were employed, similarly to half of the mothers. Half of the mothers had completed postsecondary education, likewise 43% of the fathers. We have compared the demographic characteristics at baseline between participants and nonparticipants at the 4-year follow-up (Table?S1). Parents of children who participated were more likely to be older, more educated, and married at child birth, in comparison with nonparticipants. No statistically significant differences have been identified between the excluded participants due to neurodevelopmental disorder diagnosis and nonexcluded children, in relevance with the exposure variables, with the exception of breastfeeding duration and preschool attendance at 4?years, which can be attributable to the specific difficulties these children face (Tables and S2). The ?Rhea study? is an ongoing prospective birth cohort based in Heraklion, Crete. Pregnant women, living in the prefecture of Heraklion, were recruited within 1?year (February 2007?January 2008; Chatzi et?al.,). The inclusion criteria were residence in the study area, good understanding of Greek, and maternal age above 16?years. The first contact with the families was made at the 15th gestational week (first major ultrasound examination) and the second at the 24th gestational week. The participants were invited again at child follow-up assessments when the children were 9 and 18?months old and 4?years old. Participants? assessments, at the 4-year follow-up, took place at the University Hospital of Heraklion or at health centers for the families residing in rural areas. The study was conducted according to the guidelines of the Declaration of Helsinki, and all procedures were approved by the ethical committee of the University Hospital in Heraklion. Written informed consent was obtained from the participants of the study. Of the 1,363 live singleton births, 875 children participated in the 4-year follow-up, 18 children were excluded due to neurodevelopmental disorder or other medical condition (i.e., plagiocephaly, microcephaly, hydrocephaly, brain tumor) and 8 children due to incomplete neuropsychological assessment, resulting in a cohort of 849 children with valid outcome data at 4?years. Of the 849 children with valid data on cognitive development, 700 had available data on multiple early-life exposures (Mage?=?4.2, SD?=?0.20). Of these participants, 358 were male (Mage?=?4.2, SD?=?0.19) and 342 female (Mage?=?4.2, SD?=?0.21). Of the 700 participants, 610 provided data on ADHD difficulties, of which 300 were female and 310 male. The total of the 700 children were Greek speaking, and the majority of them resided in urban centers (71%). Almost half of the children were the only or the firstborn child. The vast majority of the participants (85.7%) attended preschool at 4?years and used to spend their free time at weekends with both their parents (84.9%). Participating parents were mainly Greek (fathers 97% and mothers 87%), married at the time of the pregnancy, and living in urban areas, and almost all fathers were employed, similarly to half of the mothers. Half of the mothers had completed postsecondary education, likewise 43% of the fathers. We have compared the demographic characteristics at baseline between participants and nonparticipants at the 4-year follow-up (Table?S1). Parents of children who participated were more likely to be older, more educated, and married at child birth, in comparison with nonparticipants. No statistically significant differences have been identified between the excluded participants due to neurodevelopmental disorder diagnosis and nonexcluded children, in relevance with the exposure variables, with the exception of breastfeeding duration and preschool attendance at 4?years, which can be attributable to the specific difficulties these children face (Tables and S2). Parental and child sociodemographic characteristics were collected through questionnaires administered by interviewers. The variables were entered in PCA, after they were coded as it is referred below. Information on maternal age at child birth (in years), maternal parity at birth (primiparous, multiparous), and maternal smoking status during pregnancy (smoking, quit during pregnancy, no smoking) was collected during pregnancy. Information on breastfeeding duration (in months) was collected at 9?months postpartum and was updated at 18?months. Information regarding maternal and paternal education (low: mandatory schooling [9?years], medium: postsecondary education [12?years], high: university/technical college degree), residence (rural, urban), maternal occupational status (unemployed, employed), birth order (only child, first child, other), and maternal marital status (single, married) was obtained during the first trimester of pregnancy and updated at 4?years. Data on child preschool attendance (yes, no), television watching during weekdays (less than 1?hr, 1?2?hr, more than 2?hr), and parental participation at leisure time during weekends (both parents, one parent, other person) were obtained at the 4-year follow-up assessment. Gestational age and anthropometric measurements at birth were collected from clinical records at delivery (used as z scores, adjusted for gestational age and gender). Children's cognitive and motor development at 4?years was assessed through the McCarthy Scales of Children's Abilities (MSCA; MacCarthy,). The MSCA represent an age-appropriate instrument, developed for children of ages 2??8??years, which assesses children's present cognitive development to identify possible developmental delay in different domains. The MSCA contain 18 subscales, which provide six scales: (a) verbal scale (verbal expression and comprehension), (b) perceptual performance (reasoning), (c) quantitative scale (numerical aptitude and interest), (d) general cognitive scale (overall cognitive ability), (e) memory (verbal and nonverbal short-term memory), and (f) motor scale (gross and fine motor ability). The MSCA were administered individually to the participants by two trained psychologists. The participants were randomly distributed between them. The interobserver variability was <?1%. At the end of the neuropsychological assessment the examiners completed a standard form regarding the assessment's conditions used to evaluate the ?quality of assessment? (excellent, bad, very bad). Families received detailed feedback on their children's performance. The translation and cross-cultural adaptation of the MSCA were performed according to the internationally recommended methodology. Internal consistency (Cronbach's alpha) varied between ra?=.76 and ra?=.89, showing adequate reliability for all the scales. Confirmatory factor analysis was performed using AMOS software (v.22, IBM SPSS Statistics, Armonk, NY) and maximum likelihood estimation method. A model with five correlated latent variables was confirmed, and the confirmatory factor analysis indices supported good fit of the model (?2/df?=?2, comparative fit index?=.83, goodness-of-fit index?=.97, root mean square error of approximation?=.034). Raw scores of the MSCA were standardized for child's age at test administration using a parametric method for the estimation of age-specific reference intervals (Royston & Wright,). The parameters of the distribution were modeled as fractional polynomial functions of age and estimated by maximum likelihood. Standardized residuals were then typified having a mean of 100 points with an SD?=?15 to homogenize the scales. Information on children's ADHD difficulties was gathered through the ADHD test (ADHDT; Gilliam,), which was completed by the participants? mothers. The questionnaire is based on Diagnostic and Statistical Manual of Mental Disorders (4th ed.) criteria and designed to identify and evaluate ADHD symptoms at ages 3?23?years. It includes 36 items, providing three subscales (hyperactivity, inattention, and impulsivity) and a total ADHD difficulties index. The ADHDT has been translated and adapted for the Greek population (Maniadaki & Kakouros,). Potential factors affecting cognitive development were subjected to PCA, with varimax rotation, and the score loadings to extract the components were derived using a regression based method. Bartlett's test of sphericity (p?<.001), the Kaiser?Meyer?Olkin measure (KMO?=.596), and the correlation matrix determinant (determinant?=.049) supported sampling adequacy for PCA. The selection criterion for the extracted components was eigenvalues >?1, and the accepted factor loadings over 0.3. Multivariable linear regression models were implemented to examine the associations between the extracted components and each outcome, after the components were classified in tertiles to facilitate interpretation. The reference group was the first tertile, which corresponds to the lowest tertile of scores on each component. The components were entered simultaneously in each model and adjusted for confounding variables (child's gender, examiner, quality of assessment). Sensitivity analysis was conducted excluding preterm births to distinguish confounding by prematurity. The main analyses was repeated in a subsample of children (N?=?346) with available maternal intelligence data (Raven's Standard Progressive Matrices; Raven & Court,) to adjust for any confounding effect by maternal intelligence. The potential modification effect of child sex was examined by including an interaction term with each component and child's sex in the models. Subsequently, multivariate models stratified by child sex were implemented. Estimated associations are described in terms of ? coefficients and 95% confidence intervals (CIs). All hypothesis testing was conducted assuming a.05 significance level and a two-sided alternative hypothesis. Benjamini?Hochberg correction was performed post hoc to control for false discovery rate. The statistical analyses were performed using SPSS 22 (IBM Corporation, Armonk, NY) and Stata 13 (Stata Corp. LP, College Station, TX) statistical software. Publisher Copyright: © 2017 The Authors. Child Development © 2017 Society for Research in Child Development, Inc.
PY - 2018/7/1
Y1 - 2018/7/1
N2 - Early-life exposures are critical for later child cognitive development. McCarthy Scales of Children's Abilities (MSCA) were used to assess cognitive development of 700 preschoolers (Mage = 4.2 years), derived from the “Rhea” birth cohort, in Greece. Principal component analysis (PCA) was applied on prospectively collected exposure data. Six components were extracted; five of them were associated with child cognition. Higher parental social status, preschool attendance and less TV watching, nonsmoking during pregnancy and breastfeeding, and parental involvement in child life were protective factors of child cognition at 4 years. Increased child birth order was negatively associated with child cognition. Offspring's size at birth was not associated with any cognitive outcome. These findings reveal the importance of early-life exposures to child cognitive development.
AB - Early-life exposures are critical for later child cognitive development. McCarthy Scales of Children's Abilities (MSCA) were used to assess cognitive development of 700 preschoolers (Mage = 4.2 years), derived from the “Rhea” birth cohort, in Greece. Principal component analysis (PCA) was applied on prospectively collected exposure data. Six components were extracted; five of them were associated with child cognition. Higher parental social status, preschool attendance and less TV watching, nonsmoking during pregnancy and breastfeeding, and parental involvement in child life were protective factors of child cognition at 4 years. Increased child birth order was negatively associated with child cognition. Offspring's size at birth was not associated with any cognitive outcome. These findings reveal the importance of early-life exposures to child cognitive development.
UR - http://www.scopus.com/inward/record.url?scp=85017273931&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85017273931&partnerID=8YFLogxK
U2 - 10.1111/cdev.12782
DO - 10.1111/cdev.12782
M3 - Article
C2 - 28369793
AN - SCOPUS:85017273931
SN - 0009-3920
VL - 89
SP - 1063
EP - 1073
JO - Child Development
JF - Child Development
IS - 4
ER -