Optimizing the maximum reported cluster size in the spatial scan statistic for survival data

Lee, Sujee; Moon, Jisu; Jung, Inkyung

doi:10.1186/s12942-021-00286-w

International Journal of Health Geographics

Table 6 Simulation results for cluster model F (one irregular cluster, 20% of total area) with a mean survival time of 7

From: Optimizing the maximum reported cluster size in the spatial scan statistic for survival data

	% of cens		Maximum reported cluster size (MRCS)																Default Setting
	% of cens		3%	4%	5%	6%	8%	10%	12%	15%	20%	25%	30%	35%	40%	45%	50%	Overall	Default Setting
Circular window	10%	Frequency	20	6	13	27	69	37	57	38	33	45	17	20	37	32	45		496
		Sensitivity	0.005	0.141	0.266	0.279	0.360	0.383	0.418	0.490	0.543	0.656	0.697	0.792	0.963	0.966	0.978	0.563	0.285
		PPV	0.950	0.750	0.930	0.951	0.966	0.788	0.719	0.673	0.567	0.568	0.460	0.459	0.481	0.434	0.394	0.674	0.663
	30%	Frequency	20	2	15	12	53	21	70	33	49	55	11	17	43	19	40		460
		Sensitivity	0.004	0.154	0.226	0.282	0.373	0.374	0.409	0.501	0.584	0.641	0.769	0.810	0.964	0.931	0.983	0.575	0.267
		PPV	0.950	1.000	0.878	0.979	0.973	0.785	0.775	0.678	0.582	0.557	0.511	0.482	0.484	0.411	0.398	0.670	0.663
	50%	Frequency	41	1	36	27	31	11	135	30	50	39	3	15	59	15	26		519
		Sensitivity	0.008	0.154	0.220	0.271	0.360	0.385	0.404	0.474	0.500	0.602	0.667	0.821	0.979	0.903	0.950	0.498	0.263
		PPV	1.000	1.000	0.940	0.938	0.967	0.755	0.754	0.637	0.557	0.534	0.453	0.488	0.491	0.411	0.399	0.700	0.698
	70%	Frequency	66	0	16	4	204	30	106	23	24	37	1	1	10	31	109		662
		Sensitivity	0.015	–	0.226	0.308	0.331	0.382	0.411	0.452	0.583	0.605	0.615	0.769	0.923	0.849	0.924	0.473	0.317
		PPV	0.985	–	0.797	0.764	0.787	0.736	0.743	0.607	0.622	0.496	0.421	0.455	0.468	0.394	0.399	0.681	0.681
Elliptical window	10%	Frequency	4	6	14	16	31	49	60	71	104	80	39	49	41	29	26		619
		Sensitivity	0.001	0.154	0.247	0.313	0.347	0.424	0.512	0.574	0.673	0.779	0.862	0.906	0.934	0.942	0.965	0.668	0.414
		PPV	1.000	0.833	0.929	0.972	0.966	0.899	0.879	0.816	0.751	0.659	0.591	0.521	0.477	0.421	0.383	0.716	0.714
	30%	Frequency	9	5	5	9	14	34	62	72	79	50	63	60	42	24	33		561
		Sensitivity	0.002	0.231	0.231	0.325	0.352	0.459	0.529	0.585	0.631	0.800	0.865	0.919	0.951	0.952	0.974	0.704	0.396
		PPV	0.944	1.000	0.933	1.000	0.954	0.951	0.932	0.835	0.716	0.666	0.594	0.538	0.498	0.431	0.392	0.705	0.703
	50%	Frequency	30	12	10	8	22	26	65	74	93	89	69	50	23	18	24		613
		Sensitivity	0.009	0.224	0.223	0.317	0.332	0.396	0.501	0.577	0.663	0.792	0.845	0.906	0.953	0.932	0.968	0.650	0.402
		PPV	1.000	1.000	0.933	0.888	0.899	0.883	0.904	0.813	0.758	0.667	0.578	0.541	0.490	0.426	0.395	0.726	0.727
	70%	Frequency	12	6	5	15	84	210	29	100	247	32	34	17	8	29	18		846
		Sensitivity	0.007	0.205	0.277	0.303	0.362	0.458	0.480	0.600	0.685	0.762	0.697	0.810	0.817	0.918	0.932	0.578	0.490
		PPV	1.000	1.000	1.000	0.911	0.926	0.856	0.846	0.869	0.810	0.654	0.495	0.484	0.434	0.422	0.395	0.797	0.797

% of cens, percentage of censoring; PPV, positive predictive value
Cells most often selected as the optimal MRCS are shown in italics

Back to article page

ISSN: 1476-072X

Contact us

Submission enquiries: journalsubmissions@springernature.com