Optimizing the maximum reported cluster size for the multinomial-based spatial scan statistic

Table 6 A summary of the significant spatial clusters identified using the multinomial-based spatial scan statistic with elliptical windows at the MRCS suggested by (1) default setting, (2) SCIC₁, (3) SCIC₂, (4) elbow method, (5) MCS-P, and (6) MCHS-P

	MRCS	Cluster	Districts^a	LLR^b	p-value	Obs^c	RR^d of each category
Default	50	1	7	48.655	< 0.001	933	(0.68, 1.24, 1.45, 1.16)
		2	10	38.363	< 0.001	1200	(0.98, 1.60, 0.71, 0.91)
		3	25	40.119	< 0.001	3096	(1.19, 0.70, 0.87, 1.10)
SCIC₁, SCIC₂	10	1	4	50.148	< 0.001	501	(0.57, 1.24, 1.59, 1.47)
		2	6	37.323	< 0.001	798	(0.91, 1.76, 0.72, 0.96)
		3	5	28.589	< 0.001	694	(1.30, 0.77, 0.67, 0.75)
		4	1	19.396	< 0.001	126	(0.87, 0.27, 1.83, 0.43)
		5	2	19.119	< 0.001	237	(1.40, 0.69, 0.55, 0.61)
		6	3	17.032	0.015	385	(0.76, 1.00, 1.50, 0.80)
Elbow	4	1	2	26.842	< 0.001	240	(0.55, 1.25, 1.67, 1.07)
		2	3	22.751	< 0.001	274	(0.83, 2.01, 0.72, 0.87)
		3	1	19.396	< 0.001	126	(0.87, 0.27, 1.83, 0.43)
		4	2	23.128	< 0.001	318	(1.38, 0.80, 0.53, 0.63)
		5	2	19.119	< 0.001	237	(1.40, 0.69, 0.55, 0.61)
		6	2	17.539	0.002	269	(0.75, 0.72, 1.57, 1.46)
		7	3	15.558	0.016	322	(1.28, 0.98, 0.63, 0.45)
		8	2	13.309	0.017	220	(1.12, 1.45, 0.49, 1.09)
		9	1	12.712	0.025	108	(0.65, 0.69, 1.82, 1.19)
		10	2	12.000	0.046	299	(0.72, 1.44, 1.20, 1.18)
MCS-P, MCHS-P	2	1	1	19.396	< 0.001	126	(0.87, 0.27, 1.83, 0.43)
		2	1	19.383	< 0.001	130	(0.51, 1.15, 1.83, 0.99)
		3	1	19.061	< 0.001	116	(0.48, 1.04, 1.73, 2.09)
		4	1	14.115	0.011	139	(0.67, 2.05, 0.92, 1.06)
		5	2	12.870	0.022	135	(1.00, 1.89, 0.51, 0.68)
		6	1	12.712	0.025	108	(0.65, 0.69, 1.82, 1.19)
		7	1	11.991	0.039	109	(0.73, 1.32, 1.49, 0.00)

ISSN: 1476-072X