Efficacy of the Patient Health Questionnaire (PHQ-9) for screening depression in the quilombola population of Bahia State, Brazil
Monday, 18 August 2014
Exhibit hall (Dena'ina Center)
Sabrina M Barroso, MS
,
Universidade Federal do Triângulo Mineiro, Uberaba, Brazil
Ana Paula S Melo, MD
,
Universidade Federal de São João Del Rei, Divinópolis, Brazil
Mônia A Silva, MS
,
Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil
Mark Drew C Guimaraes, PhD
,
Universidade Federal de Minas Gerais, Belo Horizonte, Brazil
INTRODUCTION: Quilombolas are African descendants with a high degree of social vulnerability. They descend from slaves, are extremely poor and live in poor rural areas of Brazil, with limited access to health services. Incorporating the use of screening tools for depression among health practices for this population may improve access to treatment for people who suffer from depression. The Patient Health Questionnaire (PHQ-9) is a screening scale for Major Depressive Episode. It has been validated in Brazil and it is one of the most used tools in the research of depression in the country. However, there are no studies assessing its validity among rural populations in Brazil, including quilombola communities. This study aimed at investigating the psychometric properties of the PHQ-9 in quilombola communities of Vitória da Conquista (Bahia State), Brazil.
METHODS: Cross-sectional population based health survey in five quilombola communities. Analysis was carried out by Item Response Theory and we used the Graded-Response model to assess depression among the 764 adult residents who answered the PHQ-9. We estimated the discrimination ability and difficulty of each of the nine items.
RESULTS: The score on the PHQ-9 ranged from zero to 27 points, with a mean of 5.68 (SD + 5.67). All items of the PHQ-9 showed moderate to very high discrimination ability. The items “thoughts about hurting oneself and death” and “feeling depressed” showed the highest and lowest discrimination ability (a =1.716 and 0.902, respectively). On the other hand, “feeling tired” and “feeling depressed” showed the smallest and highest amount of information about depression (b=0.744 and 2.570, respectively).
CONCLUSIONS:
The PHQ-9 showed good performance in this quilombola rural population for assessing depression and this indicates that it can be used as a screening tool for public health professionals serving this population.
