Do childhood depressive symptoms interfere with intelligence in adulthood?

ABSTRACT

OBJECTIVE

To investigate the effects of depressive symptoms in childhood on the intellectual development of young adults.

METHODS

Study conducted with a birth cohort of São Luís, Maranhão, Brazil, composed of 339 participants evaluated between 7 and 9 years and between 18 and 19 years. Structural equation modeling (young adult education, sex, race/color) and childhood variables (nutritional status, depressive symptoms, cognitive function, head of household’s and mother’s education, family income) were used. In addition, head of household’s occupation, mother’s age, and presence of partner were tested as determinants of adults’ intelligence quotient (IQ).

RESULTS

Presence of depressive symptoms in childhood triggered a reduction of 0.342 in standard deviation (SD) and -3.83 points in the average IQ of adults (p-value < 0.001). Cognitive function in childhood had a total and direct positive effect (standardized coefficient [SC] = 0.701; p-value < 0.001) on IQ, increasing 7.84 points with each increase in level. A positive indirect effect of child nutritional status (SC = 0.194; p-value = 0.045), head of household’s (SC = 0.162; p-value = 0.036), and mother’s education was identified, the latter mediated by cognitive function in childhood (SC = 0.215; p-value = 0.012) on the IQ of young people.

CONCLUSION

Presence of depressive symptoms in childhood triggered a long-term negative effect on intelligence, reducing the IQ score in adulthood.

Child; Adult; Depression; Intelligence

RESUMO

OBJETIVO

Investigar os efeitos dos sintomas depressivos na infância no desenvolvimento intelectual do adulto jovem.

MÉTODOS

Estudo realizado com uma coorte de nascimentos de São Luís, Maranhão, Brasil, composta por 339 participantes avaliados entre 7 e 9 anos e entre 18 e 19 anos. Utilizou-se modelagem de equações estruturais (escolaridade do adulto jovem, sexo, raça/cor) e variáveis da infância (estado nutricional, sintomas depressivos, função cognitiva, escolaridade do chefe da família e da mãe, renda familiar). Além disso, ocupação do chefe da família, idade da mãe e presença de companheiro foram testadas como determinantes do quociente de inteligência (QI) dos adultos.

RESULTADOS

A presença de sintomas depressivos na infância gerou redução de 0,342 no desvio-padrão (DP) e -3,83 pontos no QI médio dos adultos (valor de p < 0,001). A função cognitiva na infância apresentou efeito total e direto positivo (coeficiente padronizado [CP] = 0,701; valor de p < 0,001) sobre o QI, elevando 7,84 pontos a cada aumento do nível. Identificou-se efeito indireto positivo do estado nutricional infantil (CP = 0,194; valor de p = 0,045), escolaridade do chefe da família (CP = 0,162; valor de p = 0,036) e da mãe da criança, este último mediado pela função cognitiva na infância (CP = 0,215; valor de p = 0,012) sobre o QI dos jovens.

CONCLUSÃO

A presença de sintomas depressivos na infância gerou efeito negativo de longo prazo sobre a inteligência, reduzindo a pontuação do QI na idade adulta.

Criança; Adulto; Depressão; Inteligência

INTRODUCTION

The development of cognitive skills, mainly represented by intelligence and academic performance, has been widely studied due to its impact on socially relevant outcomes such as leadership ability, success at work, and social life¹,².

Although intelligence is strongly influenced by heredity, parental stimulation, education, and nutritional and psychological status are known to also affect its development³. Among these variables, the mental health of children and adolescents has been the subject of many studies⁴ due to the high worldwide prevalence of mental disorders in this age group in recent years, especially depression, which was identified in 6.2% of individuals aged 5 to 17 in 38 countries⁴. In Brazil, depressive disorders have been the most prevalent among the mental problems that affect children and adolescents⁵,⁸.

There is evidence that symptoms of internalization and cognitive ability are interrelated variables in child development until adolescence⁶. However, a study conducted with Canadian children did not find significant evidence that the presence of internalization symptoms interfered with academic performance in adolescence, which is directly related to intellectual development. The only significant effect was a positive association of internalization with academic performance from 4 to 11 years of age⁹.

Changes caused by depressive symptoms usually cause significant damage to a child’s life, affecting their behavior at home, at school and with friends. The drop in school performance is one of the first indicators of depression in children, in addition to the development of dysphoria, isolation, and sadness¹⁰.

It remains unclear how depressive symptoms can interfere with intellectual development until adulthood, with the pathways of association between these variables, and how one variable influences another. It is also necessary to understand if the difficulties of one cause difficulties in another or if they relate only because they share causes.

Thus, the objective of this study is to answer the following questions: does the presence of depressive symptoms in childhood have a direct effect on the adult’s intelligence? Do parents’ socioeconomic variables during childhood have a greater effect on the adult’s intelligence than the presence of depressive symptoms in childhood? Is the association between childhood depressive symptoms and adult intelligence mediated by childhood cognition?

METHODS

Study Design

This is a cohort study conducted with individuals born in the city of São Luís, Maranhão, Brazil, involving two different periods: childhood and adulthood. This cohort is part of the research Determinantes ao longo do ciclo vital da obesidade, precursores de doenças crônicas, capital humano e saúde mental (Determinants throughout the life cycle of obesity, precursors of chronic diseases, human capital and mental health), developed by the Universidade Federal do Maranhão (UFMA), Faculdade de Medicina de Ribeirão Preto da Universidade de São Paulo (FMRP/USP) and the Universidade Federal de Pelotas (UFPel)..

Study Population and Sample

The first phase of the cohort was initiated at birth in 10 public and private hospitals in the city, from March 1997 to February 1998, including 96.3% of births of the period through systematic sampling with proportional stratification according to the number of births in each maternity in one out of seven deliveries. Multiple births, stillbirths, and twin births were excluded. The final sample totaled 2,443 births¹¹.

The second phase occurred when the children were 7 to 9 years old, in 2005 and 2006, through a complex sampling design, using the variable birth weight to define the sample necessary for evaluation at school age. The final sample totaled 805 children in this phase, 673 being followed since birth and 132 children born between 1997 and 1998 included in the retrospective cohort⁸.

The third phase occurred from January to November 2016, when the subjects were 18–19 years of age. Of the 805 participants from the previous stage, 339 participated in this follow-up. At this stage, we investigated socioeconomic and demographic status, income, lifestyle habits, cognitive skills, mental illness, among others¹². The study included subjects who, at the time of evaluation, underwent tests of mental health, cognitive function in childhood and intellectual development in adulthood.

Data Collection

In the school phase, a standardized questionnaire was applied to parents or guardians of children containing demographic questions. Weight and height were measured, with the children barefoot and in light clothing, using a periodically calibrated precision scale and an anthropometer⁸. To assess cognitive function in childhood, the Raven’s Colored Progressive Matrices (RCPM) was applied, which assesses general aspects of intelligence¹³; and the Draw-a-Person test (DAP test), which assesses emotional maturity and psychomotor development¹⁴. To assess the presence of depressive symptoms in childhood, the Children’s Depression Inventory was used¹⁵.

Adult intelligence was assessed by the Wechsler Adult Intelligence Scale (WAIS III), which depicts a punctual measure of intelligence level. To verify the results, the raw data of the verbal and execution scales were initially analyzed. From the sum of scores of all subtests, the crude results were converted into weighted results and analyzed by the table corresponding to the WAIS III manual. The total intelligence quotient (IQ) was obtained by summing the raw values of the verbal and execution scales, and analyzing the values in the table by age, reflecting the participants’ intelligence levels¹⁶. In cases where the participant had any condition that made it impossible to perform the test, it was not applied.

Variables

The dependent variable was the IQ of the young adult, obtained through the application of WAIS-III. For the theoretical model tested, the variable was treated as continuous numerical. For classification purposes, it can be categorized as: lower (≤ 89 points); medium (90 to 109 points); higher (≥ 110 points)¹⁶.

The explanatory variables observed in childhood were: sex (female; male), race/color (black; mixed/yellow/oriental; white); nutritional status: low weight (< 17 kg/m²) or adequate (> 17 kg/m²)⁸; depressive symptoms: absent (≥ 17 points) or present (< 17 points)¹⁵; education of the head of household and of the mother in full years (0–4 years; 5–8; 9 or more); family income in minimum wages (continuous numerical); occupation of the head of household (unskilled and unemployed manual; qualified and semi-skilled manual; non-manual); age of the mother (continuous numerical); presence of a partner (no; yes).

The variable cognitive function in childhood was constructed from the variables cognitive function measured by the RCPM test, treated as categorical variable: below average (0–25); average (26–74); above average (75–100)¹³ and cognitive function measured by the DAP test, treated as categorical variable: below average (< 25^th percentile); average (25^th percentile); above average (> 25^th percentile)¹⁴. The education variable of the young adult was treated as categorical (0–4 years; 5–8; 9 or more).

Statistical Analysis

The descriptive analysis of the data was performed using version 14 of the Stata program (StataCorp., CollegeStation, United States of America). Categorical variables were presented through absolute and relative frequencies, numerical variables by mean and standard deviation.

Structural equation modeling was used¹⁷ to investigate the association of depressive symptoms in childhood with covariates and their effects on intelligence in adulthood. A hybrid model was built, composed of confirmatory factor analysis used to construct the latent variable cognitive function in childhood and pathway analysis used to analyze the effects of depressive symptoms in childhood on the IQ in adulthood and estimate the linear relationships between the variables. For this, version 7 of the Mplus software was used.

According to the proposed theoretical model, the family socioeconomic variables observed during the participant’s childhood, occupied the most distal position and indicated the presence of depressive symptoms, nutritional status and cognitive function of the child, which determined the education of the young adult and their IQ (Figure).

Figure
Path diagram, with standardized coefficients, of the association of depressive symptoms in childhood with the intelligence quotient of young adults in the São Luís RPS Consortium Cohort. São Luís (Maranhão), Brazil, 2004–2016.

The estimator used was the robust weighted least squares mean and variance adjusted (WLSMV), used for analysis of the observed categorical variables. Theta parameterization (θ) was used to control differences in residual variances.

The missing data were imputed to the variables by Mplus software, based on the variables prior to the parameterization in the theoretical model, using frequency and Bayesian analyses¹⁸. With this imputation, the missing data did not hinder the result.

To evaluate the model fit, the following was considered as good: a) p-value > 0.05 for the Chi-square test (χ²); b) root mean square error of approximation (RMSEA) value < 0.05 and upper limit of the 90% confidence interval < 0.08; c) values greater than 0.95 for the comparative fit index (CFI) and the Tucker-Lewis index (TLI); and d) weighted root mean square residual (WRMR) values < 1¹⁷.

In the analysis of the standardized estimates for the construction of the latent variable cognition in childhood, factor load > 0.5 was considered with a p-value < 0.05 as indicative that the correlation between the observed variable and the construct is moderately high in magnitude¹⁷.

The modindices command was used to obtain suggestions for changes in the initial hypotheses¹⁸, however, there were no plausible proposals from a theoretical point of view. In the final model, the total, direct, and indirect effects of the latent and observed variables were evaluated. An effect was considered when p-value was < 0.05.

To interpret the effects of the variables on the IQ of young adults, the value of the standardized coefficient of the total effect obtained in the structural model was multiplied by the standard deviation of the IQ.

Ethical Aspects

The study met the criteria of Resolution¹⁹ 466/2012of the National Health Council of Brazil. The subjects’ guardians signed an informed consent form (ICF). Withdrawal was allowed without any prejudice to the interviewee at any stage of the research. The projects and the ICFs of both phases of the research were approved by the Research Ethics Committee of the Hospital Universitário da Universidade Federal do Maranhão: Consolidated Opinion No. 60, of April 18, 2005, and Consolidated Opinion No. 1,302,489, of October 29, 2015.

RESULTS

In the sample of 339 young adults, 55.16% were female, 67.85% were mixed/yellow/oriental and 83.78% had 5 to 8 years of study (Table 1). The mean IQ of the participants was 99.45 (11.19) points. The majority (60.47%) achieved results classified as average, with the lowest performance being 70 and the highest 137 points.

Table 1
Socioeconomic, demographic, and family characteristics, presence of depressive symptoms and cognitive development and intelligence quotient of young adults in the São Luís RPS Consortium Cohort. São Luís (Maranhão), Brazil, 2004–2016.

Characteristic	n (%)
Of the child
Sex
Female	187 (55.16)
Male	152 (44.84)
Race/color
Black	30 (8.85)
Mixed/yellow/oriental	230 (67.85)
White	79 (23.30)
Malnutritiona
Yes	25 (7.40)
No	312 (92.31)
Cognitive development (RCPM)a
Below average	88 (25.96)
Average	150 (44.25)
Above average	100 (29.50)
Cognitive development (DAP)a
Below average	102 (30.09)
Average	192 (56.64)
Above average	42 (12.68)
Depressive symptomsa
Absent	277 (81.71)
Present	62 (18.29)
Of the family during childhood
Mother’s education (years)a
0–4	45 (13.27)
5–8	89 (26.25)
≥ 9	198 (58.41)
Presence of partner
No	70 (20.71)
Yes	269 (79.29)
Education of the head of household (years)a
0–4	106 (31.27)
5–8	71 (20.94)
≥ 9	143 (42.18)
Family income (in minimum wages)a
< 1	66 (19.47)
1 < 4	230 (67.85)
≥ 4	27 (7.96)
Occupation of the head of householda
Unskilled manual or unemployed	196 (56.64)
Qualified or semi-qualified manual	100 (29.50)
Non-manual	45 (13.27)
Of the young adult
Education (years)
0–4	18 (5.31)
5–8	284 (83.78)
≥ 9	37 (10.91)
Intelligence quotient (IQ)
Below average	63 (18.58)
Average	205 (60.47)
Above average	71 (20.94)
Total	339 (100)

• RCPM: Raven’s colored progressive matrices test; DAP: Draw-a-Person test.
• ^a Excluded the ignored ones.

During childhood, 92.31% of the study participants had adequate nutritional status and 18.29% had depressive symptoms. According to the RCPM test, 44.25% presented average cognitive function in this phase and, according to the DAP, 56.64%. The mean age of the participants’ mothers in this phase was 23.15 (5,22) years. Most of them lived with a partner (79.29%) and had 9 or more years of study (58.41%). Regarding the heads of households, 56.64% were unemployed or engaged in unskilled manual occupation and 42.18% had 9 or more years of education. The income of 67.85% of families was between one and less than four minimum wages (Table 1).

The model proposed to investigate the pathways of the association between the presence of depressive symptoms in childhood and intelligence in adulthood presented a good fit for the indicators RMSEA, CFI, TLI, and WRMR (Table 2) and there were no plausible suggestions for modification.

Table 2
Model fit indices for intellectual development of young adults of the São Luís RPS Consortium Cohort. São Luís (Maranhão), Brazil, 2004–2016.

Fit indices	Modela
χ2b	265,810
Degrees of freedom	63
P-value	< 0.001
RMSEA	0.018
90%CI	0.000–0.049
P-value	0.957
CFI	0.985
TLI	0.968
WRMR	0.608

• RMSEA: root mean square error of approximation; CFI: comparative fit index; TLI: Tucker-Lewis index; WRMR: weighted root mean square residual; 90% confidence interval.
• ^a Best fit final model.
• ^b Chi-square test.

The factor analysis for the construction of cognitive function in childhood showed that the indicator variables correlated with the construct, presenting a factor load > 0.5 and p values < 0.001 (Table 3).

Table 3
Standardized coefficient, standard error, and p-values of the cognition construct in childhood and the direct effects of the indicator variables on the intelligence quotient of young adults in the São Luís RPS Consortium Cohort. São Luís (Maranhão), Brazil, 2004–2016.

Variables	Standardized coefficient	Standard error	p-value
Latent variable
Cognition in childhood
Cognitive development (DAP)	0.588	0.073	< 0.001
Cognitive development (RCPM)	0.542	0.065	< 0.001
Direct effects
Intelligence quotient
Depressive symptoms in childhood	-0.171	0.122	0.16
Cognition in childhood	0.662	0.141	< 0.001
Malnutrition in childhood	0.046	0.097	0.633
Partner presence	-0.072	0.067	0.28
Maternal education	-0.025	0.079	0.75
Head of household’s education	0.065	0.065	0.319
Young adult’s education	0.088	0.152	0.562
Cognition in childhood
Depressive symptoms in childhood	0.205	0.128	0.11
Malnutrition in childhood	0.276	0.125	0.027
Mother’s age	0.088	0.079	0.263
Mother’s education	0.325	0.095	0.001
Partner presence	0.059	0.092	0.523
Family income in childhood	-0.06	0.093	0.524
Head of household’s education	0.067	0.122	0.58
Occupation of the head of household	-0.048	0.089	0.592
Depressive symptoms in childhood
Sex	0.014	0.083	0.861
Race/color	0.12	0.081	0.141
Family income in childhood	0.092	0.081	0.256
Mother’s age	0.123	0.091	0.202
Maternal education	-0.065	0.093	0.484
Head of household’s education	0.116	0.091	0.202
Malnutrition in childhood
Family income in childhood	-0.066	0.187	0.726
Maternal education	-0.103	0.143	0.475
Head of household’s education	0.291	0.152	0.056
Occupation of the head of household	-0.021	0.118	0.861
Young adult’s education
Depressive symptoms in childhood	-0.315	0.133	0.018
Cognition in childhood	0.437	0.121	< 0.001
Family income in childhood	0.088	0.104	0.398
Occupation of the head of household	0.109	0.093	0.243
Head of household’s education	0.056	0.093	0.55
Maternal education	0.087	0.124	0.48
Partner presence	0.054	0.073	0.459

• RCPM: Raven’s colored progressive matrices test; DAP: Draw-a-Person test.

The presence of depressive symptoms in childhood had a total negative effect of -0.342 SD on the average IQ in adulthood (Table 4), which corresponds to -3.83 points (p-value < 0.001). Cognitive development in childhood had a total and direct positive effect on IQ (Table 4), revealing an increase of 0.701 SD in its average (p-value < 0.001), corresponding to an increase of 7.84 points at each elevation in the level of child cognitive development.

Table 4
Standardized coefficient, standard error, and p-values of the total, direct and indirect effects of depressive symptoms, cognition and malnutrition in childhood and levels of education of the mother and of head of household on the intelligence quotient of young adults in the São Luís RPS Consortium Cohort. São Luís (Maranhão), Brazil, 2004–2016.

Pathways	Standardized coefficient	Standard error	p-value
Total, direct, and indirect effects
Depressive symptoms in childhood
Total effect	-0.342	0.08	< 0.001
Direct effect	-0.171	0.122	0.16
Indirect effect	-0.171	0.103	0.096
Cognition in childhood
Total effect	0.701	0.102	< 0.001
Direct effect	0.662	0.141	< 0.001
Indirect effect	0.039	0.062	0.531
Malnutrition in childhood
Total effect	0.24	0.078	0.002
Direct effect	0.046	0.097	0.633
Indirect effect	0.194	0.097	0.045
Head of household’s education
Total effect	0.227	0.057	< 0.001
Total effect	0.065	0.065	0.319
Direct effect	0.162	0.077	0.036
Maternal education
Total effect	0.163	0.056	0.004
Direct effect	-0.025	0.079	0.75
Indirect effect	0.188	0.084	0.026
Via cognition in childhood	0.215	0.086	0.012

The nutritional status of the participant in childhood (SC = 0.240; p-value = 0.002) and the levels of education of the head of household (SC = 0.227; p-value < 0.001) and of the mother of the participant in this phase (SC = 0.163; p-value = 0.004) showed a total positive effect, predominantly indirect, on IQ in adulthood (Table 4). The positive indirect effect of maternal education on IQ was mediated by childhood cognition (SC = 0.215; p-value = 0.012) (Table 4).

DISCUSSION

In this study, the presence of depressive symptoms in children had negative long-term results in the development of intelligence, reducing the IQ score in adulthood. This effect was not mediated by cognitive function in childhood, whose impact on IQ was direct. Positive indirect effects of the participant’s nutritional status in childhood, levels of education of the mother and of the head of household on the adult IQ score were observed. Additionally, we observed that cognition in childhood seems to mediate the positive effect of maternal education on the IQ score in adulthood.

It is possible that the reduction in IQ observed in young adults who had depressive symptoms in childhood is explained by the impaired ability to think, concentrate, or make decisions that accompany these symptoms in children. In this age group, the drop in school performance is one of the first signs of depression and can be a manifestation of attention deficit¹⁰. In this study, depressive symptoms in childhood had a direct negative effect on the education of young adults. Another possible explanation is that depressive symptoms detected in childhood can persist and be intensified into adulthood, thus generating lifelong impacts²⁰. Previous studies have shown that changes in mental health in childhood and adolescence can compromise academic performance and the level of education achieved, and the latter, in turn, can impact the job opportunities and careers of these individuals²¹,²².

The direct effect of childhood cognition on adult intelligence observed in this study is in line with previous studies, which reveal high cognitive stability throughout life²³,²⁴. Yu et al.²³ observed, using a latent construct of intelligence, that there was considerable stability in intelligence over the four periods investigated: childhood, preschool, school, and adolescence. In addition, the longitudinal progression of intelligence from childhood to adolescence was completely mediated by the intelligence of the previous period. Additionally, data from 588 individuals in the Lothian Birth Cohort 1936 revealed that cognitive ability at age 11 was a predictor of cortical thickness approximately 60 years later and was responsible for more than two-thirds of the cross-sectional association between intelligence and cortical thickness in old age²⁴.

Several studies have already shown the importance of the family’s socioeconomic situation in development and cognition in childhood and throughout life²⁵,²⁶. In this context, education has been used as one of the main indicators of socioeconomic status²⁵. A study conducted with data from seven countries (n = 15.297), including Brazil, revealed that the education of parents could be more important than family wealth in shaping the intelligence of their children (4–22 years)²⁶.

The relationship between parenting and child intelligence can be explained by both genetic and environmental factors²⁵,²⁷. Parents with a higher level of education seem to invest more money and time in their children, resulting in better health and education conditions²⁸, as well as offering an intellectually more stimulating environment, which could result in superior performance on some tests²⁹. Evidence from low- and middle-income countries indicates that maternal level of education would be an important protective factor for the child well-being and development³⁰, with an effect apparently superior to that of paternal level of education³¹. Mothers with higher levels of education seem to provide their children with greater learning opportunities, higher quality interaction and less experience of parental stress³². In addition to these facts, maternal stimulation is among the main mechanisms that explain the relationship between maternal education and child development³¹,³³.

In agreement with our findings, recent studies show that child development can also be affected by the child’s nutritional conditions³⁴,³⁵. Proper nutrition, in a critical period like childhood, is critical for lifelong brain functioning³⁵. In addition, children with adequate nutritional status are able to interact better with their caregivers and the environment, fundamental factors for their full development³⁶.

A limitation of this study was the loss of follow-up of the subjects, especially during the third phase, due to the difficulty in locating young adults, despite all the search strategies used. With a larger sample, it would be possible to detect other effects of important determinants.

A strong point is the type of study. Cohort studies have advantages in relation to reverse causality and the possibility of monitoring the same population. Another point is the statistical method used: structural equation modeling. This modeling has the advantage of simultaneously dealing with several dependency relationships and can represent latent variables in these relationships, in addition to modeling the measurement error in the estimation process¹⁷.

This study demonstrated that IQ in adulthood is directly influenced by levels of cognitive development in childhood. During childhood, favorable socioeconomic status can positively influence IQ in adulthood, but the presence of depressive symptoms can negatively influence IQ in adulthood.

REFERENCES

• ¹
  Schelini PW, Almeida LS, Primi R. Aumento da inteligência ao longo do tempo: efeito Flynn e suas possíveis causas. Psico-USF. 2013;18(1):45-52. https://doi.org/10.1590/S1413-82712013000100006
  » https://doi.org/10.1590/S1413-82712013000100006
• ²
  Nakano TC, Moraes ID, Oliveira AW. Moraes IDT, Oliveira AW Relação entre inteligência e competências socioemocionais em crianças e adolescentes. Rev Psicol. 2019;37(2):407-24. https://doi.org/10.18800/psico.201902.002
  » https://doi.org/10.18800/psico.201902.002
• ³
  Papalia DE, Feldman RD. Desenvolvimento humano. 12a ed. [place unknown]: McGraw Hill; 2013.
• ⁴
  Erskine HE, Baxter AJ, Patton G, Moffitt TE, Patel V, Whiteford HA. The global coverage of prevalence data for mental disorders in children and adolescents. Epidemiol Psychiatr Sci. Published online 2017;26(4):395-402. https://doi.org/10.1017/S2045796015001158
  » https://doi.org/10.1017/S2045796015001158
• ⁵
  Thiengo DL, Cavalcante MT, Lovisi GM. Prevalência de transtornos mentais entre crianças e adolescentes e fatores associados: uma revisão sistemática. J Bras Psiquiatr. 2015;63(4):360-72. https://doi.org/10.1590/0047-2085000000046
  » https://doi.org/10.1590/0047-2085000000046
• ⁶
  Papachristou E, Flouri E. The codevelopment of internalizing symptoms, externalizing symptoms, and cognitive ability across childhood and adolescence. Dev Psychopathol. 2020 Oct;32(4):1375-89. https://doi.org/10.1017/S0954579419001330
  » https://doi.org/10.1017/S0954579419001330
• ⁷
  Barr PB. Early neighborhood conditions and trajectories of depressive symptoms across adolescence and into adulthood. Adv Life Course Res. 2017 Feb;2018(35):57-68. https://doi.org/10.1016/j.alcr.2018.01.005
  » https://doi.org/10.1016/j.alcr.2018.01.005
• ⁸
  Silva AA, Barbieri MA, Cardoso VC, Batista RF, Simões VM, Vianna EO, et al. Prevalence of non-communicable diseases in Brazilian children: follow-up at school age of two Brazilian birth cohorts of the 1990’s. BMC Public Health. 2011 Jun;11(1):486. https://doi.org/10.1186/1471-2458-11-486
  » https://doi.org/10.1186/1471-2458-11-486
• ⁹
  Weeks M, Ploubidis GB, Cairney J, Wild TC, Naicker K, Colman I. Developmental pathways linking childhood and adolescent internalizing, externalizing, academic competence, and adolescent depression. J Adolesc. 2016 Aug;51(1):30-40. https://doi.org/10.1016/j.adolescence.2016.05.009
  » https://doi.org/10.1016/j.adolescence.2016.05.009
• ¹⁰
  Cruvinel M, Boruchovitch E. Compreendendo a depressão infantil. Petrópolis: Vozes; 2014.
• ¹¹
  Silva AA, Coimbra LC, Silva RA, Alves MT, Lamy Filho F, Lamy ZC, et al. Perinatal health and mother-child health care in the municipality of São Luís, Maranhão State, Brazil. Cad Saúde Pública. 2001;17(6):1413-23. https://doi.org/10.1590/S0102-311X2001000600012
  » https://doi.org/10.1590/S0102-311X2001000600012
• ¹²
  Simões VM, Batista RF, Alves MTSS, Ribeiro CCC, Thomaz EBAF, Carvalo CA et al. Health of adolescents in the 1997/1998 birth cohort in São Luís, Maranhão State, Brazil. Cad Saude Publica. 2020;36(7):e00164519. https://doi.org/10.1590/0102-311X00164519
  » https://doi.org/10.1590/0102-311X00164519
• ¹³
  Angelini AL, Alves ICB, Custódio EM, Duarte WF. Matrizes manuais progressivas coloridas de Raven (Escala Especial): padronização brasileira. São Paulo: Casa do Psicólogo; 1987.
• ¹⁴
  Wechsler SM. HFD III: o desenho da figura humana: avaliação do desenvolvimento cognitivo de crianças brasileiras. São Paulo: Editora da Pontifícia Universidade Católica de São Paulo; 2003.
• ¹⁵
  Gouveia VV, Barbosa GA, Almeida HJ, Gaião AA. Children’s depression inventory - CDI: adaptation study with students of João Pessoa. J Bras Psiq. 1995;44(7):345-9. Portuguese.
• ¹⁶
  Wechsler D. WAIS-III: Escala de inteligência Wechsler para adultos. São Paulo: Casa do Psicólogo; 2004.
• ¹⁷
  Kline RB. Principles and practice of structural equation modeling. [place unknown]: The Guilford Press; 2011.
• ¹⁸
  Wang J, Wang X. Structural equation modeling: applications using mplus. [place unknown]: Thomson Digital; 2012. https://doi.org/10.1002/9781118356258
  » https://doi.org/10.1002/9781118356258
• ¹⁹
  Ministério da Saúde (BR). Conselho Nacional de Saúde. Resolução nº 466, de 12 de dezembro de 2012. Brasília, DF: Ministério da Saúde; 2012.
• ²⁰
  Veldman K, Reijneveld SA, Verhulst FC, Ortiz JA, Bültmann U. A life course perspective on mental health problems, employment, and work outcomes. Scand J Work Environ Health. 2017 Jul;43(4):316-25. https://doi.org/10.5271/sjweh.3651
  » https://doi.org/10.5271/sjweh.3651
• ²¹
  Groot S, Veldman K, Amick III, BC, Oldehinkel TAJ, Arends I, Bültmann U. Does the timing and duration of mental health problems during childhood and adolescence matter for labour market participation of young adults ? J Epidemiol Community Heal. 75(9):896-902. https://doi.org/10.1136/jech-2020-215994
  » https://doi.org/10.1136/jech-2020-215994
• ²²
  Minh A, Bültmann U, Reijneveld SA, Zon SKR Van, Mcleod CB. Depressive symptom trajectories and early adult education and employment: comparing longitudinal cohorts in Canada and the United States. Int J Environ Res Public Heal. 2021 Apr;18(8)\;4279. https://doi.org/10.3390/ijerph18084279
  » https://doi.org/10.3390/ijerph18084279
• ²³
  Yu H, Mccoach DB, Gottfried AW, Eskeles A. Stability of intelligence from infancy through adolescence: an autoregressive latent variable model. Intelligence. 2017 Sep;2018(69):8-15. https://doi.org/10.1016/j.intell.2018.03.011
  » https://doi.org/10.1016/j.intell.2018.03.011
• ²⁴
  Karama S, Bastin ME, Murray C, Royle NA, Penke L, Muñoz Maniega S, et al. Childhood cognitive ability accounts for associations between cognitive ability and brain cortical thickness in old age. Mol Psychiatry. 2014 May;19(5):555-9. https://doi.org/10.1038/mp.2013.64
  » https://doi.org/10.1038/mp.2013.64
• ²⁵
  Cave SN, Wright M, Stumm S. Change and stability in the association of parents ’ education with children’s intelligence. Intelligence. 2022;90(Jan-Feb 2021):101597. https://doi.org/10.1016/j.intell.2021.101597
  » https://doi.org/10.1016/j.intell.2021.101597
• ²⁶
  Rindermann H, Ceci SJ. Parents’ education is more important than their wealth in shaping their children’s intelligence: results of 19 samples in seven countries at different developmental levels. J Educ Gift. 2018;41(4):298-326. https://doi.org/10.1177/0162353218799481
  » https://doi.org/10.1177/0162353218799481
• ²⁷
  Stumm S, Ziada ES, Mcmillan A, Rimfeld K, Dale PS, Plomin R. Predicting educational achievement from genomic measures and socioeconomic status. Dev Sci. 2020 May;23(3):e12925: https://doi.org/10.1111/desc.12925
  » https://doi.org/10.1111/desc.12925
• ²⁸
  Prickett KC, Augustine JM. Maternal education and investments in children’s health. J Marriage Fam. 2016 Feb;78(1):7-25. https://doi.org/10.1111/jomf.12253
  » https://doi.org/10.1111/jomf.12253
• ²⁹
  Jeong J, McCoy DC, Fink G. Paternal and maternal education, caregivers’ support for learning, and early child development in 44 low- and middle-income countries. Early Child Res Q. 2017;41(June):136-48. https://doi.org/10.1016/j.ecresq.2017.07.001
  » https://doi.org/10.1016/j.ecresq.2017.07.001
• ³⁰
  Walker SP, Wachs TD, Grantham-McGregor S, Black MM, Nelson CA, Huffman SL, et al. Inequality in early childhood: risk and protective factors for early child development. Lancet. 2011 Oct;378(9799):1325-38. https://doi.org/10.1016/S0140-6736 (11)60555-2
  » https://doi.org/10.1016/S0140-6736 (11)60555-2
• ³¹
  Jeong J, Obradović J, Rasheed M, Charles D, Fink G. Maternal and paternal stimulation: mediators of parenting intervention effects on preschoolers ’ development. J Appl Dev Psychol. 2018 Nov;2019(60):105-18. https://doi.org/10.1016/j.appdev.2018.12.001
  » https://doi.org/10.1016/j.appdev.2018.12.001
• ³²
  Shin JY, Nhan NV. Predictors of parenting stress among Vietnamese mothers of young children with and without cognitive delay. J Intellect Dev Disabil. 2009 Mar;34(1):17-26. https://doi.org/10.1080/13668250802690914
  » https://doi.org/10.1080/13668250802690914
• ³³
  Magnuson KA, Sexton HR, Davis-Kean PE, Huston AC. Increases in maternal education and young children’s language skills. Merrill-Palmer Q. 2022;55(3):319-50. https://doi.org/10.1353/mpq.0.0024
  » https://doi.org/10.1353/mpq.0.0024
• ³⁴
  Burneo-Garcés C, Cruz-Quintana F, Pérez-García M, Fernández-Alcántara M, Fasfous A, Pérez-Marfil MN. Interaction between socioeconomic status and cognitive development in children aged 7, 9, and 11 years: a cross-sectional study. Dev Neuropsychol. 2019;44(1):1-16. https://doi.org/10.1080/87565641.2018.1554662
  » https://doi.org/10.1080/87565641.2018.1554662
• ³⁵
  Galler JR, Bringas-Vega ML, Tang Q, Rabinowitz AG, Musa KI, Chai WJ, et al. Neurodevelopmental effects of childhood malnutrition: a neuroimaging perspective. Neuroimage. 2021 May;231(January):117828. https://doi.org/10.1016/j.neuroimage.2021.117828
  » https://doi.org/10.1016/j.neuroimage.2021.117828
• ³⁶
  Prado EL, Dewey KG. Nutrition and brain development in early life. Nutr Rev. 2014 Apr;72(4):267-84. https://doi.org/10.1111/nure.12102
  » https://doi.org/10.1111/nure.12102

• Funding: Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq - process 400943/2013-1; grant 520664/98-1). Fundação de Amparo à Pesquisa do Estado de São Paulo (Fapesp - grants 93/0525-0; 97/09517-1; 00/0908-7).

Publication Dates

• Publication in this collection
  20 Oct 2023
• Date of issue
  2023

History

• Received
  28 June 2022
• Accepted
  19 Aug 2022