portuguese | english Revista Brasileira de Psiquiatria ISSN print 1516-4446
ISSN on-line 1809-452X
JCR IF 2016: 2.049
Fully open access
No submission fees
No publication charges
Restricted Access:

Rev Bras Psiquiatr. Ahead of Print

Maternal recognition of child mental health problems in two Brazilian cities

Isabel A. Bordin1; Bartira M. Curto1; Joseph Murray2,3

1. Departamento de Psiquiatria, Universidade Federal de São Paulo (UNIFESP), São Paulo, SP, Brazil
2. Department of Psychiatry, University of Cambridge, Cambridge, United Kingdom
3. Programa de Pós-Graduacão em Epidemiologia, Universidade Federal de Pelotas, Pelotas, RS, Brazil

Isabel A. Bordin
Departamento de Psiquiatria, Universidade Federal de São Paulo
Rua Borges Lagoa, 570
CEP 04038-030, São Paulo, SP, Brazil
E-mail: iasbordin@gmail.com

Submitted Mar 11 2016
Accepted Feb 11 2017

Descriptors: Adolescents; child psychiatry; epidemiology; families; community mental health
OBJECTIVE: To identify child behaviors and types of impairment that increase the likelihood of maternal recognition of emotional/behavioral problems (EBP) in children and adolescents.
METHODS: Maternal-reported data were obtained from two subsamples of 11-to-16-year-olds derived from cross-sectional studies conducted in two Brazilian municipalities: Itaboraí, state of Rio de Janeiro (n=480), and Embu, state of São Paulo (n=217). The Itaboraí study involved a representative sample of 6-to-16-year-olds (n=1,248; response rate = 86.0%) selected from the Family Health Program registry, which covered 85.5% of the municipal population. The Embu study was based on a probabilistic sample of clusters of eligible households (women aged 15-49 years, child < 18 years), with one mother-child pair selected randomly per household (n=813; response rate = 82.4%). The outcome variable was mother's opinion of whether her child had EBP. Potential correlates included types of child behaviors (hyperactivity/conduct/emotional problems as isolated or combined conditions) and impairment, assessed using the Strengths and Difficulties Questionnaire (SDQ); child's age and gender; maternal education and anxiety/depression (assessed using the Self-Reporting Questionnaire [SRQ]).
RESULTS: Multivariate regression models identified the following correlates of maternal perception of child EBP: comorbidity (co-occurring hyperactivity/conduct/emotional problems), emotional problems alone, and interference of problems with classroom learning and friendships.
CONCLUSION: Comorbidity of different problem types, emotional problems alone, and interference with classroom learning and friendships increase the likelihood of maternal recognition of EBP in children.


There are almost 63 million people under age 20 in Brazil.1 Of these, 8.3 million may have mental health problems, according to the prevalence of DSM-IV disorders (13.1%) found among schoolchildren in a study conducted in four of the five Brazilian regions (n=1,623; 6-16 years; response rate = 81.1%).2 However, the vast majority of Brazilian children who need mental health assistance do not receive appropriate care, mainly due to provider or system barriers,3 but also to poor maternal recognition of emotional/behavioral problems (EBP) in their children.4 In many countries, primary health care is the main setting of mental health treatment for children, who rely on adults such as parents and teachers to identify their problems and initiate service use.5

In Brazil, the Unified Health System provides universal access to health services for the entire Brazilian population. However, child and adolescent mental health services are still scarce. Specialized public services (Psychosocial Community Care Centers for Children and Adolescents) exist to assist severe cases, but are insufficient in number and distributed unequally within the country.6 Therefore, when parents decide to seek treatment or support for their children with mental health problems, primary care facilities rather than specialized services are the main source of help, particularly among more disadvantaged populations.

Parents are more likely to report intentions to seek help or use services when they recognize a problem in their children.7,8 Therefore, parental perception of mental health problems in their children is the first of several stages toward children receiving appropriate care, followed by seeking support from primary care services, recognition within primary care, and referral to specialized health services.5 Parental perception is thus critical to improving access to treatment and avoiding escalation and perpetuation of problems into adult life, which may compromise individuals' functioning and wellbeing, putting them at risk of unemployment and social exclusion.9 However, parental perception of child problems is not only influenced by child psychopathology7,10 and functional impairment but also by other factors, such as the child's age and gender, maternal education, and maternal depression.11 With increasing age, children's own perception of their functioning becomes increasingly important as a motive for seeking help.10 Help is more often sought for boys during childhood and for girls during adolescence,12 which is consistent with the higher prevalence of hyperactivity and conduct problems in young boys and depression in adolescent girls.13

Within this context, the present study aims to identify specific child behavior profiles and types of impairment identified by mothers that increase the likelihood of maternal recognition of EBP in their children, taking into account other potential correlates.


Study design and setting

Cross-sectional studies were conducted in two low-income municipalities of Southeast Brazil, both located in the greater metropolitan area of state capitals: Itaboraí, state of Rio de Janeiro (population 218,008 in 2010,14 located 40 km from the state capital Rio de Janeiro) and Embu, state of São Paulo (population 207,663 in 2000,15 located 24.5 km from the state capital São Paulo).


The original Itaboraí study (2009-2010) included a sample of 6-to-16-year-olds (n=1,248; response rate = 86.0%) randomly selected from the Family Health Program (FHP) registry. The FHP, adopted as a nationwide policy in the 1990s, is a strategy for reorganizing primary care in order to increase public access to health care. According to official information from the Itaboraí municipal government, 85.5% of the population was covered by FHP units in 2010.16 Each unit was composed of one or two teams, with each team including a physician, a nurse, and around six community health workers responsible for making home visits to registered families. In 2009, 44 FHP teams were active in Itaboraí, 31 of which (70.5%) were involved in the study. For each team, 45 to 48 eligible families (having at least one child aged 6-16 years) were randomly selected, depending on the number of community health workers in the team. The selection of teams was based on the location of units, which were stratified according to the level of neighborhood violence: low violence (7/7 teams, 100%), high violence (13/14 teams, 92.9%), and intermediate violence (11/23 teams, 47.8%).

In Itaboraí, a three-stage sampling plan was applied. The first stage was determined by the level of violence (low, intermediate, high) in each neighborhood served by an FHP unit; these three geographical areas were based on census units. In areas with low and high levels of violence, all units were invited to participate, with 7/7 (100.0%) and 13/14 (92.9%) acceptance respectively. In the intermediate-violence area, 11/23 (47.8%) units were randomly selected to enter the study. The second stage was to conduct a random selection of families among all eligible families registered with the participating FHP units in each of the three geographical areas (low: 454/3,859, intermediate: 307/2,041, high: 487/4,061). The third stage was to randomly select a child among all eligible children in each participating family, in each of the three geographical areas (low: 1/1.70, intermediate: 1/1.55, high: 1/1.62).

The sample size of the original Itaboraí study (n=1,276) was calculated on the basis of the minimum expected prevalence (7%) of its primary outcome (use of mental health services in the past 12 months by children with moderate/severe mental health problems), with an adopted relative precision of 20% (5.6-8.4%).17

The original Embu study (2001-2002) was conducted in a community in which the participant municipal health center was located. Based on census units, 24 clusters (geographic areas of maximum internal homogeneity and similar size) in the area were selected randomly. In these clusters, all eligible households were identified (residences in which a woman aged 15-49 years lived with at least one of her children < 18 years), and one mother-child pair was selected randomly per household. From this initial selection (n=987), 813 mothers participated in the study (response rate = 82.4%).

In Embu, mothers from all eligible households located in the randomly selected clusters (24/60) were invited to participate in the study. The mean number of eligible children in the participant households was two. The sample size of the original Embu study (n=864) was calculated on the basis of the expected prevalence (10%) of its primary outcome (child exposed to severe physical punishment), with an adopted relative precision of 20% (8.0-12.0%).17


The present study was restricted to 11-to-16-year-olds from Itaboraí (n=480) and Embu (n=217) for whom complete data on the variables of interest were available, based on reports by (biological, adoptive, or step) mothers. In Itaboraí, 201 of 681 11-to-16-year-olds were excluded because the informant was not the mother (n=157), because the child had severe mental retardation (n=4), or because of missing data (n=40). In Embu, 10 of 227 11-to-16-year-olds were excluded due to missing data; none had severe retardation, and the informant was the mother in all cases.

Variables and instruments

Both the Itaboraí and Embu studies collected information on the outcome variable (mother's opinion of whether her child had EBP) and potential correlates (different types of child behavior profiles and impairment, child's age and gender, and maternal education). Maternal opinion was assessed by the following question: ''Do you think < name of index child > has emotional or behavioral problems?'' Maternal reports about their child's specific behavior profiles were then measured on the Strengths and Difficulties Questionnaire (SDQ), a widely used screening instrument with good psychometric properties.18,19 The SDQ was developed by Goodman18 and validated in Brazil by Fleitlich-Bilyk & Goodman.20 The SDQ was used to identify children with clinical-level conduct problems, emotional problems, and hyperactivity, according to maternal reports (cutoff points defined clinical levels as detailed at www.sdqinfo.com). The SDQ was also used to identify impairments arising from those problems (child distress and interference with home life, friendships, classroom learning, and leisure activities), and scored positive when mothers reported ''quite a lot/a great deal'' of impact on children's functioning (versus ''only a little/not at all''). Data on maternal anxiety/depression was obtained, only in Embu, using the Self-Reporting Questionnaire (SRQ-20), a screening measure developed by the World Health Organization with a total score ranging from 0 to 20.21 The SRQ-20 was validated for the Brazilian population with a cutoff point > 7.22


All interviews were conducted in the child's household (in Itaboraí) or at the local health center (in Embu). All questions were asked verbally. For both subsamples of 11-to-16-year-olds, all interviewees were mothers, the great majority of whom were biological mothers (Itaboraí: 98.1%, Embu: 99.1%). In both studies, the question on the mother's opinion about her child's EBPs was asked before completion of the SDQ.

Statistical analysis

Chi-square tests identified significant differences between groups. A significance level of p < 0.001 was adopted due to the multiple tests conducted. All child and mother factors potentially associated with maternal recognition of child EBPs were entered into logistic regression models to estimate unadjusted and adjusted odds ratios (univariate and multivariate analysis respectively). Sampling weights applied to data analysis were calculated on the basis of the study sampling plan. In this paper, unweighted numbers of subjects are presented, but percentages are weighted to generate frequencies representative of the population of same-age children and adolescents living in Itaboraíi and Embu.

Ethics approval

The Research Ethics Committee of Universidade Federal de São Paulo approved the Itaboraí study (process number 0601/09) and the Embu study (process number 0740/02). Written informed consent was obtained from all participating mothers.


Table 1 reports the characteristics of children/adolescents and mothers in the Itaboraí and Embu samples. Table 2 shows the types of clinical-level mental health problems in each sample, occurring either alone or with other types (comorbidity). Clinical-level conduct problems were mostly comorbid in Itaboraí (71.4%) and in Embu (78.9%). Clinical-level hyperactivity was also mostly comorbid in Itaboraíi (83.4%) and in Embu (72.2%). Clinical-level emotional problems occurred combined with conduct/hyperactivity problems in a slightly higher proportion than alone in Itaboraí (55.7% vs. 44.3%), but occurred mostly alone in Embu (78.6%) (Table 2).

Among children with conduct problems alone, some 50% in both samples were not perceived as having emotional/behavioral difficulties by their mothers (Table 3). This lack of maternal recognition of problems also applied to about 70% of children with emotional problems alone in both studies, and to the majority of children with hyperactivity alone in Itaboraí (83.4%) and in Embu (60.0%). In both studies, children with emotional problems alone and comorbidity were significantly more likely to be perceived by their mothers as having EBPs, compared to children with no mental health problems (Table 3). In Itaboraí, maternal recognition of problems was much more common among children with comorbidity than among children with emotional problems alone or hyperactivity alone (p < 0.005), but children with conduct problems alone and children with comorbidity were similarly recognized by mothers as ''problematic.'' In Embu, no differences in maternal perception were found among the four categories of problems.

Univariate regression analysis showed the strength of association between each potential correlate and the study outcome (Table 4). It is interesting to note that the high effect size of all impairment components was drastically reduced in multivariate models (Table 5), confirming the existence of multiple associations within this group of variables. Multivariate regression models showed that, in both studies, comorbidity, clinical-level emotional problems alone, and problems' interference with classroom learning were independent correlates of mothers perceiving their children as ''problematic,'' adjusting for other potential correlates (Table 5). These factors were significant correlates in the presence of maternal anxiety/depression (Embu model 2). Comorbidity was a stronger correlate than isolated conditions across both studies. In Itaboraí, conduct problems alone were more influential than emotional problems alone, but in Embu, the opposite was noted. In Itaboraí, male gender, child distress, and interference with leisure activities were also significant correlates (Table 5). In Embu model 2, interference with leisure activities was a significant correlate, but in the opposite direction, which may be explained by collinearity with ''child distress'' (of the five subjects positive for interference with leisure activities, four were positive for child distress). When excluding interference with leisure activities from the Embu models, interference with friendships remained non-significant in model 1 (odds ratio [OR] = 2.25; 95%CI 0.53-9.48) and lost significance in model 2 (OR = 2.92; 95%CI 0.61-14.04).


According to the Grand Challenges in Global Mental Health initiative, improvement of children's access to evidence-based care by trained health providers in low- and middle-income countries is one of the top five challenges to improve the lives of people living with mental, neurological, and substance-use disorders.23 According to Patel et al.,24 the most viable strategy to address the treatment gap is through empowerment of existing human resources that are most intimately concerned with child care, including empowerment of parents.

In the present study, even when mothers reported child symptoms on the SDQ that indicated clinical-level conduct problems, emotional problems, or hyperactivity, mothers often did not perceive their child as having an EBP. This lack of maternal recognition is in accordance with a previous U.S. study, which found that, in a sample of children with mental disorders, only 39% of parents perceived their child to have a mental health problem.11

Nonetheless, in Itaboraí and Embu, children with conduct problems alone, emotional problems alone, and comorbidity were significantly more likely to be perceived by their mothers as having EBPs, compared to children with no problems. This is consistent with a previous study in the Netherlands25 that found an association between parents perceiving their children as having EBPs and children's mental health problems indicated both by a screening questionnaire (Child Behavior Checklist/6-18) and by a diagnostic instrument (Diagnostic Interview Schedule for Children, IV).

In our investigation, multivariate models identified specific types of child behavior profiles and impairment that correlated independently with the mother's opinion of whether the child had any EBPs. The implications of these findings are discussed in turn.

Problem type

Comorbidity was a stronger predictor of mothers perceiving their child as having an emotional/behavioral difficulty compared to isolated conditions, which is consistent with a recent Brazilian study examining the capacity of teachers to identify students in need of mental health evaluation/treatment.26

One interpretation of the finding that, in Itaboraí, conduct problems alone had a stronger association with mothers' perception of child EBPs compared to emotional problems alone might be that externalizing behaviors are particularly distressing for parents and, therefore, children with externalizing behaviors are more frequently perceived as ''problematic'' than children with internalizing problems.27,28 Hankinson28 examined thresholds for parents' perceptions of children's problems and subsequent help-seeking decisions based on children's behaviors presented in vignettes. Externalizing behaviors alone were rated by parents as more serious and representing a greater need for treatment than internalizing behaviors alone. As such, mothers may have a lower threshold for identifying externalizing symptoms as ''problematic'' than for internalizing symptoms, which may be less noticeable.29 Furthermore, conduct problems alone in the Embu sample were not a significant correlate of mothers' perception of child EBPs, probably due to the low number of these problems observed.

Impact of problems

In both studies, children's problems interfering with class-room learning was associated with mothers' perceptions of their child having an emotional/behavioral problem, independently of the presence of child psychopathology identified by clinical-level SDQ scales, other types of impact, maternal anxiety/depression, or demographic variables. This is consistent with findings from a general-population sample of Dutch 4-to-18-year-olds in which academic problems were strongly associated with perceived mental health service need and utilization, independently of child internalizing and externalizing behaviors or family stress.10 A review of influences on the help-seeking process concluded that school-related problems play an important role, influencing not only parental help-seeking but also problem recognition by general practitioners.12 In the Itaboraí sample, interference with friendships was also interpreted by mothers as a red flag of child EBPs. This is in accordance with Brazilian cultural norms, which expect children to be very social. In Embu, interference with friendships was not a significant correlate of mothers' perception of child EBPs in model 1 and lost significance in model 2 when the variable ''interference with leisure activities'' was excluded from the model due to collinearity with ''child distress.'' This lack of significance is probably due to the low number of subjects classified as having ''quite a lot/a great deal'' of interference with leisure activities in the Embu sample.

Child gender

In the Itaboraí sample, mothers were more likely to perceive EBPs in their children when the child was a boy, independently of clinical-level problems and impairment. These results are in agreement with a Brazilian study that examined a probabilistic sample of schoolchildren (n= 1,721; grades 2-6) and identified female gender as a correlate of low mental health service utilization in the presence of psychiatric disorders and/or neurodevelopmental problems.30 However, in Embu, gender was not a correlate of maternal perception of children's problems. This inconsistency in findings across sites might be explained by an association of gender with variables not measured in our study that may have differed between sites. For example, our study did not include correlates of antisocial behavior (e.g., alcohol/illicit drug use31 and involvement with criminality32), which are predominant in males compared to females, and are signs of severity that may make mothers more likely to recognize their child as having EBPs. If alcohol/illicit drug use and involvement with criminality were more common in Itaboraí than in Embu, this could explain why male gender was associated with maternal recognition of children's problems in the former but not in the latter. This hypothesis is coherent with the fact that, in 2006, Itaboraí recorded a higher homicide rate among adolescents (6.0 deaths per 1,000 in the age range 12-18 years) compared to Embu (2.8 deaths per 1,000 in the same age range).33

Parental psychopathology

Parental recognition of children's problems partly depends on levels of distress or burden experienced by the parents in raising their child.12,34 Parental distress reduces the threshold for parents perceiving their children's behaviors as ''problematic.''10,29,35,36 Also, high levels of stress reduce parental self-efficacy to cope with parenting demands and other daily challenges, making parents more likely to seek help from pediatric primary care services.37 Because parental distress and below-average self-efficacy in high-stress environments are associated with perceiving children as having problems, it is reasonable to hypothesize that mothers with anxiety/depression would be more likely to perceive their sons/daughters as having EBPs.


In conclusion, this study contributed new evidence on isolated and combined types of child behavior profiles in relation to maternal perception of children's EBPs. It emphasized the strong role of comorbidity in increasing the likelihood of a mother's recognition of her child as ''problematic.'' Our study also contributed new evidence on the specific types of impairment that are related to maternal perception of children's problems, highlighting the particular importance of impaired school performance and problem interference with friendships, instead of using only one measure of global functioning. Our findings are probably generalizable to other disadvantaged Brazilian populations living in similar socioeconomic contexts. However, the study has limitations, such as the small number of children with conduct problems alone in the Embu sample and with hyperactivity alone in both samples, which reduced statistical power to find significant associations between these isolated conditions and the study outcome. Other limitations include the absence of other relevant potential correlates (e.g., maternal beliefs/understanding about mental illness, signs of severity of child antisocial behavior such as alcohol/illicit drug use and criminality) and the lack of an independent assessment of whether the children actually had mental health problems, since only a screening instrument (SDQ) was completed by the mothers. Because the study outcome (mother's opinion of whether her child had EBPs) and main independent variables of interest (types of child behaviors and impairment) were measured on the basis of information provided by the same respondent, this data relationship may have influenced the study results. Nevertheless, while this last factor may be seen as a limitation, it also provides an interesting perspective on our results, because the mothers in this study did not always perceive their child as having problems even when they noticed symptoms (as reported on the SDQ).


We thank all study participants, health professionals, and authorities from Embu and Itaboraí and all research team members from the Social Psychiatry Division at the Department of Psychiatry, Universidade Federal de São Paulo, for their valuable collaboration. We are also grateful to Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq; MCT/CNPq/CT-Saúde/MS/SCTIE/DECIT grant no. 33/2008 - Saúde Mental) and to Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP; grant no. 00/14555-4) for financially supporting the Itaboraí study and the Embu study, respectively.


The authors report no conflicts of interest.


1 Instituto Brasileiro de Geografia e Estatística (IBGE). Censo demográfico 2010: características da população e dos domicílios - resultados do universo [Internet]; 2011. [cited 2013 Jul 27]. biblioteca.ibge.gov.br/visualizacao/periodicos/93/cd_2010_caracteristicas_populacao_domicilios.pdf.

2 Paula CS, Coutinho ES, Mari JJ, Rohde LA, Miguel EC, Bordin IA. Prevalence of psychiatric disorders among children and adolescents from four Brazilian regions. Rev Bras Psiquiatr. 2015;37:178-9.

3 Paula CS, Duarte CS, Bordin IA. Prevalence of mental health problems in children and adolescents from the outskirts of Sao Paulo City: treatment needs and service capacity evaluation. Rev Bras Psiquiatr. 2007;29:11-7.

4 Fatori D, Evans-Lacko S, Bordin IA, de Paula C. Child mental health care in Brazil: barriers and achievements. Lancet. 2012;379:e16-7.

5 Sayal K. Annotation: pathways to care for children with mental health problems. J Child Psychol Psychiatry. 2006;47:649-59.

6 Paula CS, Lauridsen-Ribeiro E, Wissow L, Bordin IA, Evans-Lacko S. How to improve the mental health care of children and adolescents in Brazil: actions needed in the public sector. Rev Bras Psiquiatr. 2012;34:334-51.

7 Thurston IB, Phares V, Coates EE, Bogart LM. Child problem recognition and help-seeking intentions among black and white parents. J Clin Child Adolesc Psychol. 2015;44:604-15.

8 Ryan SM, Jorm AF, Toumbourou JW, Lubman DI. Parent and family factors associated with service use by young people with mental health problems: a systematic review. Early Interv Psychiatry. 2015;9:433-46.

9 Royal College of Psychiatrists (RC Psych). No health without public mental health: the case for action. Position statement PS4/2010 [Internet]. 2010. [cited 2017 Mar 05]. rcpsych.ac.uk/pdf/Position%20Statement%204%20website.pdf.

10 Verhulst FC, van der Ende J. Factors associated with child mental health service use in the community. J Am Acad Child Adolesc Psychiatry. 1997;36:901-9.

11 Teagle SE. Parental problem recognition and child mental health service use. Ment Health Serv Res. 2002;4:257-66.

12 Zwaanswijk M, Verhaak PF, Bensing JM, van der Ende J, Verhulst FC. Help seeking for emotional and behavioural problems in children and adolescents: a review of recent literature. Eur Child Adolesc Psychiatry. 2003;12:153-61.

13 Merikangas KR, Nakamura EF, Kessler RC. Epidemiology of mental disorders in children and adolescents. Dialogues Clin Neurosci. 2009;11:7-20.

14 Instituto Brasileiro de Geografia e Estatística (IBGE). IBGE cidades [Internet]. 2013. [cited 2013 Jul 24]. ibge.gov.br/cidadesat/painel/painel.php?codmun=330190.

15 Instituto Brasileiro de Geografia e Estatística (IBGE). Censo demográfico 2000. População residente nos municípios de São Paulo [Internet]. [cited 2017 Apr 18]. http://www.ibge.gov.br/home/estatistica/populacao/censo2000/universo.php?tipo=31o/tabela13_1.shtm&paginaatual=1&uf=35&letra=E.

16 Rio de Janeiro, Prefeitura de Itaboraí. ''PSFs vai à sua rua'' investe em prevenção de doenças; 2010 Dec 22. [cited 27/07/2013]. prefeituraitaborai.blogspot.com.br/2010/12/psfs-vai-sua-rua-investe-em-prevencao.html.

17 CochranWG. Sampling techniques. 3rd edition. New York: Wiley; 1977.

18 Goodman R. Psychometric properties of the strengths and difficulties questionnaire. J Am Acad Child Adolesc Psychiatry. 2001;40:1337-45.

19 WoernerW, Fleitlich-Bilyk B, Martinussen R, Fletcher J, Cucchiaro G, Dalgalarrondo P, et al. The strengths and difficulties questionnaire overseas: evaluations and applications of the SDQ beyond Europe. Eur Child Adolesc Psychiatry. 2004;13:II47-54.

20 Fleitlich-Bilyk B, Goodman R. Prevalence of child and adolescent psychiatric disorders in southeast Brazil. J Am Acad Child Adolesc Psychiatry. 2004;43:727-34.

21 World Health Organization (WHO). Division of Mental Health. A User's guide to the self reporting questionnaire (SRQ) [Internet]. 1994 [cited 2017 Mar 05]. http://apps.who.int/iris/bitstream/10665/61113/1/WHO_MNH_PSF_94.8.pdf.

22 Mari JJ, Williams P. A validity study of a psychiatric screening questionnaire (SRQ-20) in primary care in the city of Sao Paulo. Br J Psychiatry. 1986;148:23-6.

23 Collins PY, Patel V, Joestl SS, March D, Insel TR, Daar AS, et al. Grand challenges in global mental health. Nature. 2011;475:27-30.

24 Patel V, Kieling C, Maulik PK, Divan G. Improving access to care for children with mental disorders: a global perspective. Arch Dis Child. 2013;98:323-7.

25 Douma JC, Dekker MC, De Ruiter KP, Verhulst FC, Koot HM. Help-seeking process of parents for psychopathology in youth with moderate to borderline intellectual disabilities. J Am Acad Child Adolesc Psychiatry. 2006;45:1232-42.

26 Vieira MA, Gadelha AA, Moriyama TS, Bressan RA, Bordin IA. Evaluating the effectiveness of a training program that builds teachers' capability to identify and appropriately refer middle and high school students with mental health problems in Brazil: an exploratory study. BMC Public Health. 2014;14:210-210.

27 Cornelius JR, Pringle J, Jernigan J, Kirisci L, Clark DB. Correlates of mental health service utilization and unmet need among a sample of male adolescents. Addicti Behav. 2001;26:11-9.

28 Hankinson JC. Child psychopathology, parental problem perception, and help-seeking behaviors [dissertation]. Tampa: University of South Florida; 2009.

29 Brestan EV, Eyberg SM, Algina J, Johnson SB, Boggs SR. How annoying is it? Defining parental tolerance for child misbehavior. Child Fam Behav Ther. 2003;25:1-15.

30 Paula CS, Bordin IA, Mari JJ, Velasque L, Rohde LA, Coutinho ES. The mental health care gap among children and adolescents: data from an epidemiological survey from four Brazilian regions. PloS One. 2014;9:e88241.

31 Lynskey MT, Fergusson DM. Childhood conduct problems, attention deficit behaviors, and adolescent alcohol, tobacco, and illicit drug use. J Abnorm Child Psychol. 1995;23:281-302.

32 Murray J, Menezes AM, Hickman M, Maughan B, Gallo EA, Matijasevich A, et al. Childhood behaviour problems predict crime and violence in late adolescence: Brazilian and British birth cohort studies. Soc Psychiatry Psychiatr Epidemiol. 2015;50:579-89.

33 Secretaria Especial dos Direitos Humanos da Presidência da República, Fundo das Nações Unidas, Observatório de Favelas, Laboratório de Análise da Violência. Índice de homicídios na adolescência (IHA): análise preliminar dos homicídios em 267 municípios brasileiros com mais de 100 mil habitantes. 2009 Jul [cited 2017 Mar 05]. https://www.unicef.org/brazil/pt/br_IHA.pdf.

34 Pfefferle SG, Spitznagel EL. Children's mental health service use and maternal mental health: a path analytic model. Child Youth Serv Rev. 2009;31:378-82.

35 Weisz JR, Suwanlert S, Chaiyasit W, Weiss B, Walter BR, Anderson WW. Thai and American perspectives on over- and undercontrolled child behavior problems: exploring the threshold model among parents, teachers, and psychologists. J Consult Clin Psychol. 1988;56:601-9.

36 Hennigan KM, O'Keefe M, Noether CD, Rinehart DJ, Russell LA. Through a mother's eyes: sources of bias when mothers with cooccurring disorders assess their children. J Behav Health Serv Res. 2006;33:87-104.

37 Janicke DM, Finney JW. Children's primary health care services: social-cognitive factors related to utilization. J Pediatr Psychol. 2003;28:547-57.