Trigonometric Regression for Analysis of Public Health Surveillance Data

Table 5

Simulation results for E. coli O157:H7 infection outbreaks assuming a negative binomial model for the entire state with overdispersion parameter , , or .

Pertussis-like outbreak				Overdispersion			Overdispersion			Overdispersion
Size ()	Duration (2)	Signature^4	Time	POD^1	CED^2	FPR^3	POD^1	CED^2	FPR^3	POD^1	CED^2	FPR^3
5	4	(3, 5, 5, 3): 16	Trough	0.579	1.338	1.287	0.591	1.362	2.476	0.626	1.421	4.186
5	5	(2, 4, 5, 4, 2): 17	Trough	0.518	1.453	1.273	0.570	1.566	2.404	0.632	1.688	4.127
5	6	(1, 3, 5, 5, 3, 1): 18	Trough	0.652	2.024	1.255	0.687	2.099	2.428	0.723	2.127	4.069
10	4	(5, 10, 10, 5): 30	Trough	1.000	1.771	1.306	0.999	1.764	2.460	0.997	1.773	4.220
10	5	(4, 7, 10, 7, 4): 32	Trough	0.996	2.252	1.290	0.994	2.232	2.435	0.988	2.229	4.146
10	6	(4, 7, 10, 10, 7, 4): 42	Trough	1.000	2.244	1.255	1.000	2.259	2.393	0.999	2.256	4.078
10	4	(5, 10, 10, 5): 30	Mid	0.957	1.863	1.269	0.943	1.831	2.434	0.919	1.809	4.099
10	5	(4, 7, 10, 7, 4): 32	Mid	0.915	2.249	1.260	0.909	2.203	2.358	0.892	2.144	4.046
10	6	(4, 7, 10, 10, 7, 4): 42	Mid	0.977	2.519	1.229	0.970	2.470	2.339	0.958	2.433	3.972
20	4	(10, 20, 20, 10): 60	Mid	1.000	1.248	1.271	1.000	1.305	2.387	1.000	1.367	4.127
20	5	(7, 14, 20, 14, 7): 62	Mid	1.000	1.622	1.255	1.000	1.639	2.333	1.000	1.689	4.018
20	6	(7, 14, 20, 20, 14, 7): 82	Mid	1.000	1.620	1.221	1.000	1.650	2.344	1.000	1.678	3.950
10	4	(5, 10, 10, 5): 30	Peak	0.701	1.542	1.237	0.736	1.585	2.327	0.749	1.571	4.006
10	5	(4, 7, 10, 7, 4): 32	Peak	0.682	1.845	1.180	0.726	1.887	2.263	0.762	1.914	3.827
10	6	(4, 7, 10, 10, 7, 4): 42	Peak	0.799	2.375	1.152	0.834	2.398	2.196	0.856	2.339	3.734
20	4	(10, 20, 20, 10): 60	Peak	0.999	1.599	1.222	0.998	1.624	2.331	0.991	1.672	3.980
20	5	(7, 14, 20, 14, 7): 62	Peak	0.996	2.061	1.187	0.991	2.055	2.261	0.980	2.061	3.850
20	6	(7, 14, 20, 20, 14, 7): 82	peak	1.000	2.083	1.142	1.000	2.107	2.190	0.996	2.126	3.732

Probability of detection (POD) per “outbreak.”
²Conditional expected delay (CED) per “detected outbreaks.”
³False-positive rate in terms of false-positive signals per year.
⁴The signature is the number of cases in each week during the outbreak, along with the total number of cases.