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 3 Simulation results for cluster model C (two circular clusters, 10% each 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	18	12	11	17	103	48	46	65	19	12	20	39	53	39	26		528
		Sensitivity	0.005	0.174	0.288	0.284	0.414	0.464	0.478	0.544	0.513	0.597	0.771	0.857	0.943	0.953	0.971	0.588	0.313
		PPV	0.982	0.875	0.947	0.931	0.971	0.819	0.724	0.711	0.547	0.484	0.495	0.468	0.451	0.393	0.361	0.694	0.690
	30%	Frequency	30	7	15	15	114	46	55	82	28	34	17	26	18	23	17		527
		Sensitivity	0.007	0.167	0.244	0.256	0.382	0.409	0.444	0.498	0.506	0.674	0.750	0.840	0.917	0.971	0.980	0.500	0.267
		PPV	0.967	1.000	0.778	0.967	0.986	0.799	0.736	0.724	0.525	0.525	0.485	0.455	0.439	0.393	0.360	0.734	0.734
	50%	Frequency	47	4	17	8	82	24	44	58	26	49	11	23	27	28	17		465
		Sensitivity	0.011	0.167	0.245	0.250	0.383	0.438	0.451	0.471	0.458	0.692	0.735	0.844	0.874	0.911	0.927	0.505	0.242
		PPV	1.000	1.000	0.878	0.833	0.937	0.824	0.733	0.687	0.487	0.535	0.485	0.453	0.431	0.381	0.343	0.697	0.692
	70%	Frequency	77	2	15	3	281	46	5	14	7	10	23	27	3	53	25		591
		Sensitivity	0.016	0.167	0.250	0.250	0.356	0.379	0.383	0.399	0.500	0.658	0.757	0.904	0.778	0.849	0.827	0.424	0.258
		PPV	0.987	1.000	0.750	0.600	0.764	0.697	0.582	0.506	0.482	0.492	0.486	0.477	0.378	0.357	0.321	0.691	0.691
Elliptical window	10%	Frequency	4	14	14	17	73	43	54	39	48	55	55	39	39	24	33		551
		Sensitivity	0.001	0.196	0.268	0.304	0.389	0.446	0.435	0.532	0.547	0.811	0.861	0.949	0.949	0.951	0.975	0.636	0.350
		PPV	1.000	0.905	0.869	0.985	0.908	0.826	0.709	0.692	0.574	0.623	0.573	0.530	0.460	0.388	0.354	0.664	0.662
	30%	Frequency	14	23	6	22	84	86	52	64	43	50	34	20	22	25	15		560
		Sensitivity	0.004	0.217	0.236	0.299	0.367	0.420	0.431	0.475	0.574	0.777	0.851	0.900	0.905	0.960	0.983	0.539	0.304
		PPV	1.000	1.000	0.944	0.967	0.936	0.871	0.751	0.666	0.581	0.609	0.548	0.491	0.439	0.400	0.359	0.729	0.729
	50%	Frequency	27	26	9	20	54	59	39	78	46	50	23	22	20	25	33		531
		Sensitivity	0.007	0.244	0.222	0.300	0.360	0.412	0.436	0.473	0.533	0.717	0.710	0.845	0.900	0.937	0.927	0.526	0.285
		PPV	1.000	1.000	0.861	0.905	0.887	0.802	0.740	0.647	0.542	0.570	0.440	0.465	0.437	0.389	0.339	0.673	0.670
	70%	Frequency	34	23	2	54	154	122	17	15	35	14	32	37	40	20	49		648
		Sensitivity	0.010	0.221	0.208	0.323	0.329	0.336	0.343	0.383	0.521	0.637	0.779	0.887	0.892	0.883	0.959	0.481	0.316
		PPV	1.000	1.000	0.708	0.973	0.762	0.594	0.529	0.498	0.536	0.478	0.492	0.484	0.432	0.372	0.357	0.646	0.645

% 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