Outbreak detection algorithms based on generalized linear model: a review with new practical examples

Table 2 Characteristics of GLM-based outbreak detection algorithms in examples

Method	Package’s name; Command	Range^a(week)	Control parameters	Data	Reference to R packag and control parameters
Original Farrington	R package surveillance; algo.farrington()	270–319	b = 5, w = 3, weight, reweight = TRUE α = 0.05	Measles	[9, 19]
Original Farrington	R package surveillance; algo.farrington()	52–120	b = 2, w = 3, weight, reweight = TRUE α = 0.05	Covid-19	[9, 19]
Farrington Flexible	R package surveillance; farringtonFlexible()	270–319	b = 5, w = 3, weight threshold = 2.58, thresholdMethod = “nbPlugin”, α = 0.05	Measles	[18, 19]
GLR Poisson	R package surveillance; algo.glrpois()	270–319	ARL = 5, dir = “inc”	Measles	[19, 29]
GLR Poisson	R package surveillance; algo.glrpois()	52–120	ARL = 5, dir = “inc”	Covid-19	[19, 29]
Periodic Poisson GLM Method	-	270–319	m = 2	Measles	[26]

ISSN: 1471-2288