Social factors associated with child mental health problems in Brazil: cross sectional survey

BMJ 2001; 323 doi: (Published 15 September 2001) Cite this as: BMJ 2001;323:599
  1. Bacy Fleitlich, clinical researchera,
  2. Robert Goodman, professor of brain and behavioural medicine (r.goodman{at}
  1. a Department of Psychiatry, University of São Paulo, São Paulo 05403-010, Brazil
  2. b Department of Child and Adolescent Psychiatry, Institute of Psychiatry, King's College London, London SE5 8AF
  1. Correspondence to: R Goodman
  • Accepted 13 June 2001

The prevalence of child psychiatric disorder in the developed world is 10-20%, but in the developing world, where children and adolescents make up a higher proportion of the population, the prevalence may be higher.1 Relatively little is known about the extent to which social risk factors identified in the developed world also apply in the developing world.1 To guide healthcare planning we used three contrasting samples from the largest and most populous country in Latin America to examine the association between child mental health problems and social factors, such as poverty, family violence, and parental mental illness.

Participants, methods, and results

Three contrasting neighbourhoods were selected from a single district in south east Brazil: a new favela (shanty town) of crowded makeshift dwellings, lacking sanitation, and built on illegally occupied land; a stable urban community; and a rural village. We aimed to identify all 7-14 year olds (compulsory school years) in these three areas from school registers. For the children on school lists, we obtained informed parental agreement to participation for 75% of the children from the favela, 67% from the stable urban community, and 95% from the rural village. Supplementary house to house searches in the favela identified relatively few additional 7-14 year olds who were not on any school list—amounting to only 16% of the favela sample and not differing significantly in social or psychiatric characteristics. The total sample of 898 participating children comprised 488 from the favela, 346 from the stable urban area, and 64 from the rural area.

Children with probable psychiatric disorder were identified by parents, teachers, and self report versions of the strengths and difficulties questionnaire, using a predictive algorithm that has been validated in both developed and developing countries. 2 3 In this study an independent psychiatric assessment4 of randomly chosen children identified a psychiatric disorder in 23 of 41 of those whose questionnaire results suggested psychiatric disorder, compared with six of 40 of those whose results did not (χ2=13.1, 1 df, P<0.001). We assessed social class from affluence and parental education using standard Brazilian criteria—ranging from A/B (middle class) to E (abject poverty and illiteracy). Maternal psychiatric disorder was predicted with the validated Brazilian cut off point on a self report questionnaire (the SRQ-20).5 Parents were asked how the child was disciplined and whether the child had witnessed marital violence.


Rates of probable child psychiatric disorder according to social factors

The figure shows the univariate results and their significance. In forward conditional logistic regression, area (whether it was a favela or not) was no longer significant (P=0.84) once the effects of social class, maternal depression, domestic violence, and harsh discipline had been allowed for.


Poverty, maternal psychiatric illness, and family violence were all strongly associated with higher rates of probable psychiatric disorders among Brazilian 7-14 year olds. Though it is tempting to conclude that the social adversities caused the children's behavioural problems, the children's problems may have evoked maternal depression or harsh discipline. Another possibility is that the social adversities and behavioural problems both arose from unmeasured factors. The higher rate of behavioural problems in the favela could be accounted for by the associated poverty, maternal psychiatric illness, and domestic violence; future studies should investigate the possible role of social capital and social networks in buffering the impact of these social adversities. It did not need this study to identify the importance of tackling poverty, parental mental illness, or family violence in the developing world, but tackling these problems is likely to have the additional benefit of improving child mental health. Moreover, our findings help to identify marginalised groups of “at risk” children and families who should be targeted for preventive or curative services.


Contributors: Both authors devised the study and the data collection instruments and analysed the data. RG obtained the funding. BF oversaw data collection. RG is guarantor of the study.


  • Funding Wellcome Trust.

  • Competing interests None declared.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract