Evaluating the use of verbal autopsy methods for producing cause of death estimates in Uganda

在过去的60年里，与其他区域相比，非洲显示出从一种人口和流行病制度向另一种人口和流行病制度的循证小幅转变。这在一定程度上是因为撒哈拉以南非洲(包括乌干达)的死因确定工作仍然很差，因为缺乏有效的民事登记系统。世界卫生组织建议在缺乏全面的重要登记系统的情况下使用口头尸检(VA)，并可将其嵌入卫生和人口监测系统(HDSS)。关于从退伍军人管理局面谈中确定死因的最佳方法，目前仍在争论中。我的博士项目将评估用于从退伍军人管理局访谈中确定死因的不同方法，然后估计乌干达的特定死因死亡率。它规定，该国的特定死因死亡情况随着时间的推移而变化，并因年龄、性别和地区而不同。为了实现这一目标，我的项目将分为四个主要研究。在研究1中，我将根据医生对口头尸检访谈数据的审查，估计乌干达拉凯的成人原因特定死亡率。研究2将评估从口头尸检访谈中确定潜在死因所使用的不同方法的表现。在研究3中，我将按年龄、性别和日历年估计和检查乌干达三个地点的特定原因死亡率，而研究4将根据叙述部分的VA访谈数据估计Masaka的特定原因死亡率。我将使用在该国三个HDSS中进行的VA访谈中收集的数据。我将使用先进的定量分析方法来实现上述目标。预计这项研究的结果将有助于现有的关于特定原因死亡的文献和辩论，清楚地表明多年来流行病学转变是如何演变的。

Over the past 60 years, Africa has demonstrated a small evidence-based shift from one demographic and epidemiological regime to another as compared to other regions. This is partly because cause-of-death determination in Sub-Saharan Africa (including Uganda) remain very poor due to the lack of effective civil registration vital systems. World Health Organization recommends use of verbal autopsy (VA) in the absence of a comprehensive, vital registration systems and it can be embedded within Health and Demographic Surveillance Systems (HDSS). There is an ongoing debate on the best method for assigning cause of death from VA interviews. My PhD project will evaluate the different methods used to assign cause of death from VA interviews and then estimate cause-specific mortality in Uganda. It stipulates that the cause specific mortality profile of the country has changed over time, and it differs by age, gender, and region. To achieve this, my project will be divided into four major studies. In study 1, I will estimate the adult cause specific mortality rates in Rakai, Uganda based on Physician reviews of the verbal autopsy interview data. Study 2 will evaluate the performance of the different methods used in assigning underlying cause of death from verbal autopsy interviews. In study 3, I will estimate and examine the cause specific mortality rates of the three Ugandan sites by age, sex and calendar year, while study 4 will estimate cause specific mortality rates of Masaka, based on VA interview data in the narratives section.I will use data collected in VA interviews conducted in three HDSS within the country. I will use advanced quantitative analysis methods to achieve the above objectives. It is expected that the findings of this study will contribute to the existing literature and debate on cause specific mortality clearly showing how the epidemiological transition has evolved over the years.