O processo de coleta de dados sobre pessoas com deficiência na atenção básica: A confiabilidade dessa informação para uso em pesquisas científicas e definição de políticas públicas

Abstract

OBJECTIVES: In this study, we developed a form for data collection of people with disabilities (PcD) to use in primary care, and analyzed the process of collecting and the reliability of the information for use in scientific research and to establish public policies; the results were presented in three papers. RESULTS: In the systematic review of the literature on the Primary Care Information System (SIAB) in Brazil, 174 studies were identified, 53 was selected for analysis. In 73% of the studies, the SIAB addressed local issues, particularly on populations of Southeastern Brazil (47%). The category "the SIAB as a data source" included 35 studies (66%), which 18 used demographic data and social health records and nine got data on diseases, specifically hypertension and diabetes. In the category "SIAB as the object of study", there were altogether 18 papers of which 12 have used qualitative methods to describe knowledge and practices of health workers in relation to the information system. These authors pointed out problems in the information production and consolidation process, mainly due to lack of training and supervision of community health workers or alienation and bureaucratization of their work process. However, only two studies have investigated the reliability of the information concerning the registration of families generated in the system, showing that the information is reliable. In the second article, we evaluated the competences and skills of Community Health Workers (CHW) to perform scientific research. Eleven CHWs participated of a in-service training course based on the educational textbooks elaborated to qualify the information on disabled people. Each CHW interviewed two PcD, repeating the same procedure two weeks later (test and retest). Altogether 50 interviews were recorded on video (25 in step "test" and 25 in "retest"), being afterwards transcribed and analyzed. Four categories of analysis were created and subdivided into ten subgroups, describing the most appropriate competences and skills to collect epidemiological data. The outcomes showed that the majority of the participants had more than 60% of the competences and skills to carry out epidemiological research; however, there was wide variation in performance. This variation could be explained by differences regarding education, technical training of HWS and engagement to the research activities. In the third article, a cross-sectional population-based study, conducted in two communities in the backlands of Northeastern Brazil, the prevalence of DP was estimated by CHWs was 2.5% while the researchers obtained 6.2%, showing an underestimation of 4% . The value of agreement (Kappa), considering all the data (2,589 items), compared to the HWS researchers was 0.808 (p <0.01), showing an almost perfect agreement of the information gathered by primary care workers. CONCLUSION: The tools designed to qualify the information about disabled people were reliable. The CHWs showed skills and ability to collect and to register reliable information, although the capture of cases to determine prevalence had possibly been impaired due to the work process.