Toubiana L, Flahault A
INSERM U 444, Institut Fédératif Saint-Antoine de Recherche sur la Santé, Paris, France.
Eur J Epidemiol. 1998 Jul;14(5):465-70. doi: 10.1023/a:1007481929237.
To develop a method based on a space-time criterion for early detection of epidemics of influenza-like-illness in France.
Since 1984, the French Communicable Diseases computer Network (FCDN) routinely detects epidemics of influenza-like-illness when the national incidence rate is, for 2 consecutive weeks, above a threshold computed by a periodic regression model. It appears that some areas reported early increases in incidence several weeks before the national epidemic. An optimised space-time criterion allows an early detection of the epidemic periods.
Applying this space-time criterion to the last 11 epidemics (from 1986), the sensitivity was 0.82 and the specificity was 0.99.
This simple procedure can be used as an additional tool for early detection of an epidemic taking into account the distribution of new cases in space and time.
开发一种基于时空标准的方法,用于在法国早期检测流感样疾病的流行情况。
自1984年以来,法国传染病计算机网络(FCDN)在全国发病率连续两周高于通过周期性回归模型计算出的阈值时,常规检测流感样疾病的流行情况。似乎有些地区在全国流行前几周就报告了发病率的早期上升。一种优化的时空标准可实现对流行期的早期检测。
将此时空标准应用于过去11次流行(从1986年起),灵敏度为0.82,特异度为0.99。
考虑到新发病例在空间和时间上的分布,这个简单的程序可作为早期检测流行的附加工具。