SciELO - Scientific Electronic Library Online

vol.113 issue2  suppl.Ischemic heart disease and diabetes mellitus: a statistical analysisPrivate health insurance plans and clinical practice author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Sao Paulo Medical Journal

Print version ISSN 1516-3180

Sao Paulo Med. J. vol.113 no.2 supl.0 São Paulo May 1995 



Prevalence studies of mental disorders in Brazil



Sergio Baxter Andreoli

Presidente do Centro de Estudos do Departamento de Psiquiatria e Psicologia Medica da Universidade Federal de São Paulo - UNIFESP



Community prevalence studies of mental disorder began with Javis in 1955 in Massachusetts. Before the second world war other studies were undertaken in USA, such as those by Lenkau (1958) in Baltimore and by Roth and Luton (1943) in Tennessee. These studies, although advanced for their time, were limited in terms of case identification. Javis calculated the frequency of "insanity" and "idiocy" by means of interviews with community leaders, and Lenkau estimated the frequency of "nervousness" using unstructured interviews with subjects. These studies used vague diagnostic concepts and did not take into account reliability and validity (Weissman, 1987).

After the second world war, greater methodological rigor was introduced, with more attention being paid to issues such as sampling, data collection, and statistical analysis. Despite these improvements, the diagnostic process continued to lack reliability. Global measures of assessment were used which did not fit neatly with the clinical picture (Weissman, 1987).

The divergent strands of epidemiological psychiatry, clinical psychiatry and academic research began to converge in the 1970's with the introduction of specific diagnostic criteria. Standardized interviews improved the identification and quantification of signs and symptoms, greatly increasing diagnostic reliability (Weissman, 1987).

In the 1970 's these instruments were used in epidemiological studies to identify psychiatric cases in the community. One important advance that occured during this period was that the use of convenience samples, such as hospitalized patients, gave an incomplete picture of the disorder. In nontreated individuals, psychopathology often differs from the textbook descriptions (Dohrenwend et al., 1980).

Despite these advances, it is only re~ently that investigators have begun to study Brazilian communities (Santana et al., 1988). In 1974, Coutinho (1976) estimated the prevalence of mental disorder in Salvador, using a two-step method (identifying suspected cases and then making definitive diagnoses). The methodological weaknesses of this study were that the instruments used had been insufficiently validated and the diagnostic system was undefined. In 1974, Luz estimated the prevalence of alcoholism in the population of Vila Vargas, Porto Alegre. In 1976 a series of population studies started in different sectors of the metropolitan city of Salvador, Bahia. The aim of these studies was to develop two screening instruments - the adult psychiatry morbidity questionnaire (QMPA) (Santana, 1982) and the children's psychiatric morbidity questionnaire (QMPI) (Almeida-Filho, 1985). These Dave been used in a range of descriptive and analytic epidemiological studies (Almeida-Filho, 1981, 1982, 1985; Santana, 1982; Almeida-Filho & Bastos, 1982; Almeida-Filho et al., 1983, 1984; Dunnighan, 1985; Santana & Almeida-Filho, 1985; Bastos, 1986). Before 1976, Brazilian studies were of low diagnostic reliability; even those using the twostep method tended to use non-standardized diagnostic critena.

During the early 1990's four population surveys were carried out whose methodological strengths, regarding case identification, stand out, . Blay et al's (1991) study of elderly people in the city of São Paulo; Almeida-Filho et al's. (1992) large scale epidemiological study in three Brazilian cities; Almeida and Coutinho's (1993) prevalence study of alcohol misuse on Governor's Island (Rio de Janeiro); and Pechansky's (1993) study of alcohol consumption among adolescents in the city Porto Alegre.

The Multicenter Study of Psychiatric Morbidity in Urban Areas is the first population prevalence survey undertaken in Brazil that uses standardized methodology for case definition and identification (Almeida-Filho et al., 1992). This study used a two-step approach to case identification. In the
first stage 6,470 individuals, above the age of 15, were interviewed with the QMPA screening instrument. In the second stage a sample of QMPA-identified "cases" were interviewed with the DSM III Symptom Checklist (American Psychiatric Association, 1980). The data collected in this study are currently being analyzed by several groups of researchers and some findings have already been published (Andreoli, 1993; Andreoli et al., 1994; Busnello et al, 1993; Coutinho, 1993 and Soares, 1994).

As can be seen, psychiatric epidemiology in Brazil is some way behind that in the USA or England but is beginning to develop and advance in terms of methodological stingency.

Supported by FAPESP (Grant n° 94/0526-0) Psychiatric of Departamento de Psiquiatria e Psicologia Medica da Universidade Federal de São Paulo - Escola Paulista de Medicina.

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License