Volume 5 - 2000
Editor: Giovanni Torello


Dezembro de 2000 - Vol.5 - Nº 12

No Paiz dos Yankees

Epidemiology of Schizophrenia in a semi-rural area in Northeast Brazil

Dr. Erick Messias

Caros leitores,

Nessas últimas semanas não tivemos Grand Rounds porque o departamento como um todo esta concentrado em aulas para os estudantes de medicina. Decidi então utilizar um artigo meu, acerca de epidemiologia da esquizofrenia no Nordeste do Brazil, publicado no comeco desse ano no Journal of Nervous and Mental Diseases, como minha coluna de dezembro. Desejo a todos boas festa. Até o próximo mês.

Erick Messias

PS o artigo está em inglês porque não tive tempo de traduzir, se alguem tiver alguma dúvida, manda para [email protected]


Epidemiology of Schizophrenia in a semi-rural area in Northeast Brazil

Erick Messias, M.D. 1
Jose Jackson Sampaio, M.D., Ph.D.
Nidia Cordeiro Messias, M.D. 3
Brian Kirkpatrick, M.D. 1

1Maryland Psychiatric Research Center - Department of Psychiatry - University of Maryland Baltimore, Maryland
2Department of Public Health - State University of Ceara - Fortaleza, Ceara, Brazil
3 Federal University of Ceara - Fortaleza, Ceara, Brazil

For submission to:

Address correspondence to:Erick Messias, MD
Maryland Psychiatric Research Center
P.O. Box 21228
Baltimore, MD
Phone: 410-402-7935
Fax: 410-402-7198
E-mail: [email protected]
Supported in part by PHS grant MH 40279.


The systematic study of the epidemiology of schizophrenia in tropical countries data back to at least 1904, when Kraepelin traveled to Java and interviewed 100 psychiatric patients, 39 of whom he diagnosed as exhibiting dementia praecox (Bendick, 1989). Understanding the distribution and presentation of schizophrenia in different cultures was also the goal of two international studies, the International Pilot Study of Schizophrenia (IPSS) and the Determinants of Outcome of Severe Mental Disorders (Sartorious et al. 1977, 1986). The IPSS found important similarities in the distribution and presentation of schizophrenia across cultures, but also showed differences regarding outcome, in particular a more favorable course in developing countries (Sartorious, 1986).

The epidemiology of schizophrenia in tropical regions has received relativley little study, and Brazil is no exception. In Brazil, the more developed south has been studied more than the poorer (Menezes, 1993, Menezes, 1996, Kerr-Correa, 1998; Santos, 1998), and urban areas have more studied more than rural ones (Sousa e Silva, 1998; Almeida-Filho & al., 1997, Araujo & Monteiro, 1995). Even the best designed epidemiological study to date concentrated on three large urban settings, none of them in the Northeast (Almeida-Filho, 1997). In the present study, we examined the epidemiology of schizophrenia in a rural region in northeastern Brazil.


Geographical Area

Our study was performed in Sao Camilo de Lelis Hospital in Mossoro, a city in the state of Rio Grande do Norte in the northeastern Brazilian hinterland (5 degrees 11 minutes south, 37 degrees 21 minutes west). In the 1990 census, Mossoro had 192,267 inhabitants, but Sao Camilo de Lelis Hospital serves a population of approximately a quarter million inhabitants, and is the only inpatient or outpatient psychiatric facility in this area. The hospital was opened in 1964; prior to that time, the region had no inpatient or outpatient psychiatric facility.

Information was obtained from the clinical records of the hospital, including date of birth, date of first admission, gender, marital status and clinical subtype for all patients admitted with a diagnosis of schizophrenia. Patients admitted from 1964, when the hospital opened to 1994 were included in the study. The diagnosis was made at the time of admission by the house staff and, in most of cases, an ICD subtype diagnosis was also recorded. The diagnostic criteria used were based on the International Classification of Diseases (ICD), which included paranoid, hebephrenic, residual, simple and catatonic subtypes. Psychiatry in Brazil has been heavily influenced by French and German psychiatry (Knobel, 1984), leading to the use of the ICD in the clinical setting. Ethnicity was not recorded.


There were 2856 patients admitted with a diagnosis of schizopihrenia. Of these, 55% were male and 45% female. The mean age at admission was 31.2 (+ 11.59). Males had an earlier age of onset (t=5.57; d.f. with unequal variance = 2296; P<.001), and a smaller percentage of males had been married prior to admission (chi-square = 62.0; d.f = 1; p < .0001; Table 1). In a logistic regression analysis both age (chi-square = 62.8; d.f.=1; p < .0001) and gender (chi-square = 48.5; d.f. = 1; p < .0001) were signficant predictors of marital status (as a 0/1 variable) at the time of admission.

The paranoid and simple subtypes were the most common (Table 2). The males and females did not differ relative to subtype (chi-square = 6.06; d.f = 8; p =.64).


In this sample from northeastern Brazil, we found males had an earlier age of onset, and had poorer premorbid function as measured by marriage prior to admission. The difference in marriage did not appear to be due to the younger age of the males, as the difference was still significant after accounting for the variance in marriage associated with age.

Limitations of the study should be considered when analyzing the results. First, these were clinical diagnoses given by psychiatrists in their everyday practice. The interrater reliability of these diagnoses is not known. Second, these patients constituted a hospitalized sample. As no outpatient facilities were availalbe during the time of these admission, the bias due to inpatient versus outpatient treatment should have been small. However, the size of any bias due to treatment versus no treatment is not known. Finally, although the data came from the patients’ first hospitalizations, some may have had previous episodes of illness which did not lead to treatment.

Other studies internationally have had findings similar to ours, relative to gender ratio, age of onset, and social function. However, the percentage of patients with the simple subtype is unusual. This subtype was considered for inclusion in the DSM IV (McGlashan & Fenton, 1991), where it is listed in an appendix for further study because of insufficient information. However, the simple subtype is included in the ICD 10 (CID 10, 1993), which is the system used in Brazil. The other study on hospital treated schizophrenia in South Brazil does not mention ICD subtype. (Menezes, 1993)

Despite the limitations of our data, the similarities to results from other countries is reassuring, and may permit health officials in northeastern Brazil to make plans on the basis population estimates that are likely to be rather similar to those in other countries and other regions of Brazil. (Menezes, 1993, Menezes, 1996, Kerr-Correa, 1998; Santos, 1998)


Almeida OP, Forlenza OV, Lima NK, Bigliani V, Arcusi SM, Gentile M, Faria NM, Lourenco CC, de Oliveira DA (1997) Psychiatric morbidity among the elderly in a primary care setting – report from a survey in Sao Paulo, Brazil. Int J Geriatr Psychiatry 12(7):728-736.

Almeida-Filho N, Mari JM, Coutinho E, Franca JF, Fernandes J, Andreoli SB, Busnello ED (1997) Brazilian Multicentric Study of Psychiatric Morbidity. Br J Psychiatry 17(1):524-529.

Araujo NP, Monteiro MG (1995) Family history of alcoholism and psychiatric co-morbidity in Brazilian male Alcoholics and controls. Addiction 90(9):1205-1211.

Bendick C (1989) Emil Kraepelin Forchungreise nach Java im Jahre 1904: Ein Beitrag zur Geschiste der Ethnopsychiatrie. Kolner medizinhistorische Beitrage. Buchandlung C. –E. Kohlhauer, Feuchtwagen

Bromet EJ, Dew MA, Eaton W (1995) Epidemiology of psychosis with special reference to schzophrenia in: Tsuang MT, Tohen M, Zahner GEP: Textbook of Psychiatric Epidemiology. Wiley-Liss, New York

Classificacao dos transtornos mentais e do comportamento da CID 10. Artes Medicas. Porto Alegre, 1993

Jablesky, A (1995) Schizophrenia: Manifestations, Incidence and Course in Different Cultures: a WHO Ten-Country Study. Psycho Med, Monograph Supplement. Cambridge: Cambridge University Press,.

Kerr-Correa F, Sousa LB, Calil HM (1998) Affective disorders, hospital admissions, and seasonal variation of mania in a southern area, southern hemisphere. Psychopathology 31(5):265-269.

Knobbed P (1984) Relations between French psychiatry and Brazil. Ann Med Psychol 142:499-503.

Lecrubier Y, Goldberg DP Form and Frequency of Mental Disorders Across Centers. In Mental Illness in General Health Care: an International Study, Editores Ustin T B & Sartorius N, Editora by Wiley, West Sussex, England.

McGlashan TH, Fenton WS (1991) Classical Subtypes of Schizophrenia: Literature review for DSM IV. Schizophrenia bulletin 17(4):609-623.

Menezes PR, Mann AH (1993) Characteristics of hospital-treated schizophrenia in Sao Paulo, Brazil. Soc Psychiatry Psychiatr Epidemiol 28(6):267-274.

Menezes PR, Mann AH (1996) Mortality among patients with non-affective functional psychoses in a metropolitan area of south-eastern Brazil. Rev Saude Publica 30(4):304-309.

Riecher A, Maurer K, Loffler W, Fatkenheuer B, An Der Heiden W, Munk-Jorgensen P, Stromgren E, Hafner H (1991) Gender differences in age at onset and course of schizophrenic disorders. In Hafner H, Gattaz WF (eds), Search for the Causes of Schizophrenia (pp14-33). Berlin: Springer-Verlag.

Santos DN, Blizard R, Mann AH (1998) The development of a scale to measure concepts of schizophrenia : experience among Brazilian psychiatrists. Rev. Saude Publica 32(3):201-8

Sartorius N, Jablenski A, Korten A, Ernberg G, Anker M, Cooper JE, Day R (1986) Early manifestations and first-contact incidence of schizophrenia in different cultures. Psychol Med 16:909-928.

Souza MF, Silva G de R (1998) Risk of minor psychiatric disorders in a metropolitan area of southeast Brazil. Rev Saude Publica 32(1):50-58.

Schizophrenia Subtypes in Sao Camilo de Lelis Hospital in Mossoro, Rio Grande do Norte, Brazil



n (%)


n (%)


N (%)


675 (44.4)

558 (43.9)

1233 (43.7)


589 (38.8)

488 (38.4)

1077 (38.2)


150 (9.9)

146 (11.5)

296 (10.5)


66 (4.3)

43 (3.4)

109 (3.9)


34 (2.2)

34 (2.7)

68 (2.4)


5 (0.3)

2 (0.2)

5 (0.3)


1530 (100)

1283 (100)

2813 (100)


Demographic Features of Patients in Sao Camilo de Lelis Hospital in Mossoro






Mean age at first

Admission (+SD)


29.95 (10.89)

32.67 (12.19)

31.20 (11.59)

Age at first



n (%)


n (%)


n (%)

< 20 yo

240 (18.0)

188 (16.6)

428 (17.4)

21-25 yo

329 (24.7)

186 (16.4)

515 (20.9)

26-30 yo

264 (19.8)

209 (18.5)

473 (19.2)

31-35 yo

186 (14.0)

147 (13.0)

333 (13.5)

36-40 yo

101 (7.6)

133 (11.8)

234 (9.5)

> 40 yo

211 (15.9)

268 (23.7)

479 (19.5)


1331 (100)

1131 (100)

2462 (100.0)