Avaliação das propriedades psicométricas da escala de rastreamento populacional para depressão CES-D em populações clínica e não-clínica de adolescentes e adultos jovens
No Thumbnail Available
Date
1997
Authors
Silveira Filho, Dartiu Xavier da [UNIFESP]
relationships.isAdvisorOf
Jorge, Miguel Roberto [UNIFESP]
item.page.type-of
Tese de doutorado
Journal Title
Journal ISSN
Volume Title
Abstract
Existem evidencias de aumento na prevalencia de transtornos depressivos na populacao geral e de que tem havido incidencia progressivamente mais precoce de depressao no transcorrer deste seculo. A grande maioria dos trabalhos cientificos sobre depressao nao focaliza especificamente populacoes de adolescentes e adultos jovens, sendo a literatura mais escassa ainda quando se trata de populacoes clinicas. Embora exista um grande contingente de estudos a respeito de farmacodependencia nesta faixa etaria, poucos se detem nos aspectos relativos a comorbidade psiquiatrica e as dificuldades diagnosticas nesta populacao. As escalas de rastreamento populacional para depressao tem sido amplamente utilizadas em etapas iniciais de identificacao de casos em levantamentos epidemiologicos, a despeito dos questionamentos sobre a concordancia do desempenho destas escalas com resultados de procedimentos diagnosticos mais complexos realizados em uma segunda etapa. Alem disto, poucos estudos examinaram o desempenho destas escalas em populacoes de adolescentes e adultos jovens, sendo a literatura cientifica extremamente restrita quando se trata de populacoes clinicas. Utilizamos neste estudo a CES-D, escala para depressao do Centro de Estudos Epidemiologicos dos EUA, por ser uma das escalas de rastreamento populacional mais utilizadas nos ultimos vinte anos. Propusemo-nos a estudar as propriedades psicometricas e a estrutura fatorial da escala em duas populacoes de adolescentes e adultos jovens. Uma populacao de 523 estudantes universitarios da cidade de São Paulo e uma amostra de 50 farmacodependentes que procuraram tratamento em um ambulatorio da rede publica assistencial (PROAD) responderam a CES-D. A amostra de farmacodependentes e uma sub-amostra da populacao de estudantes foram tambem submetidas a uma entrevista psiquiatrica. Avaliamos o desempenho da escala de rastreamento comparativamente ao diagnostico psiquiatrico segundo o RDC (Research Diagnostic Criteria) nas duas amostras. Comparando-se os dois grupos, as estimativas de prevalencia entre os estudantes universitarios foram de 7,9 % para transtornos depressivos atuais e de 19 % para transtornos depressivos ao longo da vida, enquanto que entre os farmacodependentes encontramos frequencias de 32 % para transtornos depressivos atuais e de 44 % para transtornos depressivos ao longo da vida. Com relacao a escala de rastreamento, concluimos que entre os estudantes universitarios (populacao nao clinica) a escala apresentou melhor desempenho quando utilizamos o ponto de corte 15, contrastando com a amostra de farmacodependentes (populacao clinica), na qual o ponto de corte 24 correspondeu ao melhor desempenho da escala. Quanto ao estudo da estrutura da escala, duas questoes da CES-D foram excluidas por apresentarem baixa correlacao com a 18 questoes remanescentes. Entretanto, a CES-D apresentou boa consistencia interna (alfa de Cronbach=0,85) e a analise da estrutura fatorial resultou em solucao de quatro dimensoes da escala, demonstrando assim relativa estabilidade quando utilizada em populacoes jovens em nosso meio. Baseando-nos nestes resultados, concluimos que a CES-D pode ser de utilidade em uma primeira etapa de levantamento de casos, desde que sejam consideradas algumas de suas limitacoes. Finalmente, levantamos algumas sugestoes a respeito da utilizacao deste tipo de instrumento em populacoes distintas daquelas para as quais foram concebidos.
An increase in prevalence and an earlier age of onset of depression throughout the twentieth century has been documented. Nevertheless, most scientific literature on depression do not focus specifically on adolescents or young adults. Depression studies involving populations of youngsters are even more difficult to find. Although there are many studies on addiction among young adults and adolescents, only a few go deeper in discussing comorbidity issues and diagnostic problems within these groups. Screening scales have been widely used as a first step tool for case identification in epidemiological surveys, despite the fact that some studies raised the problems of poor agreement between the results of scale screening and second step diagnostic procedures. Furthermore, validation studies of these scales are still lacking for both clinical and non-clinical populations of youngsters. In the present study we used the Center for Epidemiological Studies Depression Scale (CES-D) , one of the most used depression screening tools in the two last decades, aiming to evaluate its psychometric performance and factor structure when used with two different populations of young adults and adolescents. Fifty drug addicts attending a public outward clinic (PROAD) and 523 college students in the city of São Paulo were administered the scale. All drug addicts and a sample of college students underwent psychiatric interviews aiming to assess psychiatric diagnosis according to Research Diagnostic Criteria (RDC). Psychiatric diagnosis were used as a gold standard for evaluating CES-D performance in both samples. Among college students we found an estimated prevalence of 7.9 % for present depressive disorders and 19 % for lifetime depressive disorders. Among substance addicts, 32 % were diagnosed as being presently depressed and 44 % met diagnostic criteria for lifetime diagnosis of depression. While studying the concurrent validity of CES-D we observed that the scale was at its best performance when using a cut-off of 15 among college students and a cut-off of 24 when among addicts. Structure analysis evidenced that two questions could be excluded from the scale since they displayed low correlation with the remaining 18 questions. Nevertheless, internal consistency of the scale was high (Cronbach’s alpha = 0.85) and factor analysis resulted in a four dimension solution, findings which are similar to those found in previous studies involving general population. Despite some limitations, we concluded CES-D can be a useful tool to be used as a first step screening instrument for depression case identification among young adults and adolescents. Finally, we also pointed out some difficulties which may arrive when using an instrument under different conditions for which it was originally meant to be used.
An increase in prevalence and an earlier age of onset of depression throughout the twentieth century has been documented. Nevertheless, most scientific literature on depression do not focus specifically on adolescents or young adults. Depression studies involving populations of youngsters are even more difficult to find. Although there are many studies on addiction among young adults and adolescents, only a few go deeper in discussing comorbidity issues and diagnostic problems within these groups. Screening scales have been widely used as a first step tool for case identification in epidemiological surveys, despite the fact that some studies raised the problems of poor agreement between the results of scale screening and second step diagnostic procedures. Furthermore, validation studies of these scales are still lacking for both clinical and non-clinical populations of youngsters. In the present study we used the Center for Epidemiological Studies Depression Scale (CES-D) , one of the most used depression screening tools in the two last decades, aiming to evaluate its psychometric performance and factor structure when used with two different populations of young adults and adolescents. Fifty drug addicts attending a public outward clinic (PROAD) and 523 college students in the city of São Paulo were administered the scale. All drug addicts and a sample of college students underwent psychiatric interviews aiming to assess psychiatric diagnosis according to Research Diagnostic Criteria (RDC). Psychiatric diagnosis were used as a gold standard for evaluating CES-D performance in both samples. Among college students we found an estimated prevalence of 7.9 % for present depressive disorders and 19 % for lifetime depressive disorders. Among substance addicts, 32 % were diagnosed as being presently depressed and 44 % met diagnostic criteria for lifetime diagnosis of depression. While studying the concurrent validity of CES-D we observed that the scale was at its best performance when using a cut-off of 15 among college students and a cut-off of 24 when among addicts. Structure analysis evidenced that two questions could be excluded from the scale since they displayed low correlation with the remaining 18 questions. Nevertheless, internal consistency of the scale was high (Cronbach’s alpha = 0.85) and factor analysis resulted in a four dimension solution, findings which are similar to those found in previous studies involving general population. Despite some limitations, we concluded CES-D can be a useful tool to be used as a first step screening instrument for depression case identification among young adults and adolescents. Finally, we also pointed out some difficulties which may arrive when using an instrument under different conditions for which it was originally meant to be used.
Description
Citation
São Paulo: [s.n.], 1997. 119 f.