Ling Cheong Y, Gruebner Oliver, Krämer Alexander, Lakes Tobia
Geoinformation Science Lab, Geography Department, Humboldt-Universität zu Berlin, Berlin, Germany.
Department of Epidemiology, Mailman School of Public Health, Columbia University, New York, USA.
Geospat Health. 2014 Nov;9(1):131-40. doi: 10.4081/gh.2014.11.
Spatio-temporal patterns of dengue risk in Malaysia were studied both at the address and the sub-district level in the province of Selangor and the Federal Territory of Kuala Lumpur. We geocoded laboratory-confirmed dengue cases from the years 2008 to 2010 at the address level and further aggregated the cases in proportion to the population at risk at the sub-district level. Kulldorff's spatial scan statistic was applied for the investigation that identified changing spatial patterns of dengue cases at both levels. At the address level, spatio-temporal clusters of dengue cases were concentrated at the central and south-eastern part of the study area in the early part of the years studied. Analyses at the sub-district level revealed a consistent spatial clustering of a high number of cases proportional to the population at risk. Linking both levels assisted in the identification of differences and confirmed the presence of areas at high risk for dengue infection. Our results suggest that the observed dengue cases had both a spatial and a temporal epidemiological component, which needs to be acknowledged and addressed to develop efficient control measures, including spatially explicit vector control. Our findings highlight the importance of detailed geographical analysis of disease cases in heterogeneous environments with a focus on clustered populations at different spatial and temporal scales. We conclude that bringing together information on the spatio-temporal distribution of dengue cases with a deeper insight of linkages between dengue risk, climate factors and land use constitutes an important step towards the development of an effective risk management strategy.
在马来西亚雪兰莪州和吉隆坡联邦直辖区，我们在地址和分区层面研究了登革热风险的时空模式。我们对2008年至2010年实验室确诊的登革热病例进行地址层面的地理编码，并进一步按分区层面的风险人群比例汇总病例。运用Kulldorff空间扫描统计法进行调查，以确定两个层面登革热病例不断变化的空间模式。在地址层面，在所研究年份的早期，登革热病例的时空聚集集中在研究区域的中部和东南部。分区层面的分析显示，大量病例与风险人群成比例的一致空间聚集。将两个层面的分析联系起来有助于识别差异，并确认存在登革热感染高风险区域。我们的结果表明，观察到的登革热病例具有空间和时间流行病学成分，为制定有效的控制措施（包括空间明确的病媒控制），需要认识到并解决这一问题。我们的研究结果强调了在异质环境中对疾病病例进行详细地理分析的重要性，重点是不同空间和时间尺度上的聚集人群。我们得出结论，将登革热病例的时空分布信息与对登革热风险、气候因素和土地利用之间联系的更深入了解结合起来，是朝着制定有效的风险管理策略迈出的重要一步。
