Ribeirão Preto College of Nursing of the University of São Paulo, Ribeirão Preto Campus (EERP/USP), Avenue dos Bandeirantes, 3900, 14040-902, Ribeirão Preto, São Paulo, Brazil; NOVA National School of Public Health, Universidade NOVA de Lisboa, Avenue Padre Cruz, 1600-560 Lisbon, Portugal.
Ribeirão Preto College of Nursing of the University of São Paulo, Ribeirão Preto Campus (EERP/USP), Avenue dos Bandeirantes, 3900, 14040-902, Ribeirão Preto, São Paulo, Brazil.
J Infect Public Health. 2020 Aug;13(8):1148-1155. doi: 10.1016/j.jiph.2020.03.010. Epub 2020 Apr 12.
Tuberculosis (TB) is one of the top 10 causes of death worldwide; in 2016, over 95% of TB deaths occurred in low- and middle-income countries. Although the incidence and deaths from TB have decreased in Brazil in recent years, the disease has increased in the vulnerable population, whose diagnosis is more delayed and the chances for abandonment and deaths are significantly higher. This study aimed to identify high-risk areas for TB mortality and evidence their social determinants through a sensitive tailored social index, in a context of high inequality in South Brazil.
A multistep statistical methodology was developed, based on spatial clustering, categorical principal components analysis, and receiver operating characteristic curves (ROC). This study considered 138 spatial units in Curitiba, South Brazil. TB deaths (2008-2015) were obtained from the National Information Mortality System and social variables from the Brazilian Human Development Atlas (2013).
There were 128 TB deaths recorded in the study: the mortality rate was 0.9/100,000 inhabitants, minimum-maximum: 0-25.51/100,000, with a mean (standard deviation) of 1.07 (2.71), and 78 space units had no deaths. One risk cluster of TB mortality was found in the south region (RR=2.64, p=0.01). Considering the social variables, several clusters were identified in the social risk indicator (SRI): income (899.82/1752.94; 0.024), GINI Index (0.41/0.45; 0.010), and overcrowding (25.07/15.39; 0.032). The SRI showed a high capacity to discriminate the TB mortality areas (area under ROC curve 0.865, 95% CI: 0.796-0.934).
A powerful risk map (SRI) was developed, allowing tailored and personalised interventions. The south of Curitiba was identified as a high-risk area for TB mortality and the majority of social variables. This methodological approach can be generalised to other areas and/or other public health problems.
结核病（TB）是全球十大死因之一；2016 年，超过 95%的结核病死亡发生在中低收入国家。尽管近年来巴西的结核病发病率和死亡率有所下降，但该疾病在弱势群体中的发病率有所上升，这些人群的诊断更为延迟，放弃治疗和死亡的可能性显著更高。本研究旨在通过一个敏感的定制社会指数，在巴西南部高度不平等的背景下，确定结核病死亡率的高风险地区，并确定其社会决定因素。
采用多步骤统计方法，基于空间聚类、类别主成分分析和接收者操作特征曲线（ROC）。本研究考虑了巴西南部库里蒂巴的 138 个空间单元。结核病死亡（2008-2015 年）数据来自国家死亡信息系统，社会变量来自巴西人类发展地图集（2013 年）。
研究中记录了 128 例结核病死亡：死亡率为 0.9/100,000 居民，最低-最高：0-25.51/100,000，平均值（标准差）为 1.07（2.71），78 个空间单元无死亡。在南部地区发现了一个结核病死亡风险集群（RR=2.64，p=0.01）。考虑到社会变量，在社会风险指标（SRI）中确定了几个集群：收入（899.82/1752.94；0.024）、基尼指数（0.41/0.45；0.010）和过度拥挤（25.07/15.39；0.032）。SRI 显示出较高的区分结核病死亡率区域的能力（ROC 曲线下面积 0.865，95%CI：0.796-0.934）。
制定了一个强大的风险图（SRI），可以进行定制化和个性化干预。库里蒂巴南部被确定为结核病死亡率和大多数社会变量的高风险地区。这种方法可以推广到其他地区和/或其他公共卫生问题。
