Nogueira Luciana Tolstenko, Rêgo Cristiane Fortes Napoleão do, Gomes Keila Rejane Oliveira, Campelo Viriato
Centro de Ciências e Saúde, Universidade Federal do Piauí, Teresina, Brazil.
Cad Saude Publica. 2009 Feb;25(2):366-74. doi: 10.1590/s0102-311x2009000200015.
Brazilian health policy depends on the accuracy of data in the health information system. This study aimed to assess the reliability and validity of data on underlying cause of death due to a specific neoplasm in the Municipality of Teresina, Piauí State, Brazil, based on all the death certificates from 2004 and 2005 that reported oral cancer as the underlying cause of death (total of 23 death certificates). The death certificates were recoded for underlying cause of death by using a form targeted at collecting clinical and laboratory data to confirm the medical diagnosis on patient charts. Diagnosis was confirmed by histopathologic examination and clinical history. Simple agreement was 91.3%, kappa coefficient 0.84, and positive predictive value 90.9%. Based on the findings, the mortality statistics were valid and reliable. One limitation to this study was the possible underreporting of cases with oral cancer as the underlying cause of death, given that such data are not included in the mortality information system in Teresina.
巴西的卫生政策依赖于卫生信息系统中数据的准确性。本研究旨在基于2004年和2005年所有将口腔癌列为根本死因的死亡证明（共23份死亡证明），评估巴西皮奥伊州特雷西纳市特定肿瘤所致根本死因数据的可靠性和有效性。通过使用一份旨在收集临床和实验室数据以确认病历上医学诊断的表格，对死亡证明的根本死因进行重新编码。诊断通过组织病理学检查和临床病史得以确认。简单一致性为91.3%，kappa系数为0.84，阳性预测值为90.9%。基于这些发现，死亡率统计数据是有效且可靠的。本研究的一个局限性是，鉴于特雷西纳市的死亡率信息系统未包含此类数据，可能存在将口腔癌作为根本死因的病例报告不足的情况。
