Halder Amal K, Kabir M
Department of Statistics, Jahangirnagar University, Dhaka 1342, Bangladesh.
J Health Popul Nutr. 2008 Mar;26(1):64-73.
Principal component analysis (PCA) was applied to assets and other household data, collected as part of the Bangladesh Demographic and Health Survey (BDHS) in 2004, to rank individuals according to a household socioeconomic index and to investigate whether this predicts access to the sanitation system or outcomes. PCA was used for determining wealth indices for 11,440 women in 10,500 households in Bangladesh. The index was based on the presence or absence of items from a list of 13 specific household assets and three housing characteristics. PCA revealed 35 components, of which the first component accounted for 18% of the total variance. Ownership of assets and housing features contributed almost equally to the variance in the first component. In this study, ownership of latrines was examined as an example of sanitation-intervention access, and rates of mortality of neonates, infant, and children aged less than five years (under-five mortality) as examples of health outcomes. The analysis demonstrated significant gradients in both access and outcome measures across the wealth quintiles. The findings call for more attention to approaches for reducing health inequalities. These could include reforms in the health sector to provide more equitable allocation of resources, improvement in the quality of health services offered to the poor, and redesigning interventions and their delivery to ensure that they are more pro-poor.
主成分分析（PCA）被应用于2004年作为孟加拉国人口与健康调查（BDHS）一部分收集的资产及其他家庭数据，以根据家庭社会经济指数对个人进行排名，并调查这是否能预测获得卫生系统服务的情况或结果。主成分分析用于确定孟加拉国10500个家庭中11440名妇女的财富指数。该指数基于13种特定家庭资产清单中的物品是否存在以及三种住房特征。主成分分析揭示了35个成分，其中第一个成分占总方差的18%。资产所有权和住房特征对第一个成分的方差贡献几乎相同。在本研究中，以厕所所有权为例考察卫生干预措施的获得情况，以新生儿、婴儿和5岁以下儿童的死亡率（五岁以下儿童死亡率）为例考察健康结果。分析表明，在各财富五分位数组中，获得服务情况和结果指标均存在显著梯度。研究结果呼吁更加关注减少健康不平等的方法。这些方法可能包括卫生部门的改革，以实现更公平的资源分配，提高为穷人提供的卫生服务质量，以及重新设计干预措施及其实施方式，以确保它们更有利于穷人。
