Queiroz Rejane Christine de Sousa, Mattos Inês Echenique, Monteiro Gina Torres Rego, Koifman Sergio
Departamento de Epidemiologia e Métodos Quantitativos em Saúde, Escola Nacional de Saúde Pública, Fundação Oswaldo Cruz, Rio de Janeiro, RJ, 21041-210, Brasil.
Cad Saude Publica. 2003 Nov-Dec;19(6):1645-53. doi: 10.1590/s0102-311x2003000600009. Epub 2004 Mar 3.
Mortality statistics are frequently employed in cancer epidemiology studies. The aim of this research was to evaluate the reliability and validity of oral cavity cancer as the underlying cause of death reported on death certificates in the Municipality of Rio de Janeiro, Brazil. All death certificates since 1999 reporting this underlying cause of death were selected from the Municipal Mortality Information System. The death certificates were re-coded by an independent technician, and a standardized questionnaire was designed to collect clinical and laboratory data for diagnostic confirmation from patient records. Histopathological results and/or clinical history were considered as the gold standard to establish the diagnosis of cancer of the mouth. Reliability analysis showed a simple percentage agreement of 95.1% and a kappa coefficient of 0.93. Positive predictive value was 86.5%. Mortality statistics for oral cancer in Rio de Janeiro Municipality demonstrate good reliability and validity, possibly because diagnostic confirmation is facilitated by easy access to the tumor site.
死亡率统计数据经常用于癌症流行病学研究。本研究的目的是评估口腔癌作为巴西里约热内卢市死亡证明上报的根本死因的可靠性和有效性。自1999年以来所有报告该根本死因的死亡证明均从市死亡率信息系统中选取。由一名独立技术人员对死亡证明进行重新编码，并设计了一份标准化问卷，以便从患者记录中收集临床和实验室数据用于诊断确认。组织病理学结果和/或临床病史被视为确立口腔癌诊断的金标准。可靠性分析显示简单百分比一致性为95.1%，kappa系数为0.93。阳性预测值为86.5%。里约热内卢市口腔癌死亡率统计数据显示出良好的可靠性和有效性，这可能是因为肿瘤部位易于接触便于进行诊断确认。
