Space-Time Distribution Characteristics of Tuberculosis and Its Socioeconomic Factors in Southern China from 2015 to 2019

Figures & data

Table 1 Trends in Reported Incidence of Tuberculosis in Guangxi, 2015–2019

Trend	Group	Year	Population (10,000)^a	TB Cases	Reported Incidence (1/100,000)	Z^c	P
Temporal trends	___	2015	4811^b	51,299	106.63	−24.24	<0.001
		2016	4857^b	46,688	96.13
		2017	4907^b	46,578	94.92
		2018	4947^b	45,700	92.38
		2019	4982^b	43,358	87.03
Sex trends	Males	2015	2913	36,810	126.36	−20.09	<0.001
		2016	2943	33,748	114.67
		2017	2951	33,766	114.42
		2018	2980	32,862	110.28
		2019	2998	31,053	103.58
	Females	2015	2605	14,489	55.62	−12.20	<0.001
		2016	2636	12,940	49.09
		2017	2649	12,812	48.37
		2018	2679	12,838	47.92
		2019	2697	12,305	45.62
Age trends	0-	2015	1059	254	2.40	−0.72	0.472
		2016	1068	286	2.68
		2017	1080	254	2.35
		2018	1087	280	2.58
		2019	1093	244	2.23
	15-	2015	3258	36,556	112.19	−19.87	<0.001
		2016	3288	33,979	103.33
		2017	3318	33,582	101.19
		2018	3349	33,169	99.05
		2019	3371	31,109	92.28
	65-	2015	478	14,489	303.01	−13.96	<0.001
		2016	481	12,423	258.07
		2017	486	12,742	262.15
		2018	491	12,251	249.70
		2019	496	12,005	242.04

Note: a, Data from Guangxi Statistical Yearbook; b, Data are permanent population; c, Cochran-Armitage Trend test.

Figure 1 Seasonal decomposition analysis TB cases in Guangxi from 2015 to 2019.

Figure 2 Proportion of occupational composition of TB cases.

Figure 3 Spatial distribution of the overall reported incidence of TB in Guangxi, 2015–2019 (A), and annual incidence of TB reported by county (B).

Table 2 The Global Spatial Autocorrelation and Trend in Reported Incidence of TB in Guangxi, 2015–2019

Year	Moran’s I	Var(I)	Z	P
2015	0.399	−0.009	6.877	0.001
2016	0.438	−0.009	7.163	0.001
2017	0.480	−0.009	7.906	0.001
2018	0.486	−0.009	7.952	0.001
2019	0.515	−0.009	8.522	0.001

Figure 4 Local spatial autocorrelation of the reported incidence of TB in Guangxi, 2015–2019.

Table 3 Spatial Clustering of TB Cases in Guangxi, 2015–2019

Clusters	No. Counties	Latitude (N)	Longitude (E)	Radius (km)	Time Frame	Population	Annual Cases/100,000	Observed/Expected	RR	LLR	P
Most likely cluster	22	24.66	107.86	147.45	2015/1/1 to 2017/12/31	7,217,833	145.0	1.52	1.60	2683.76	< 0.001
Secondary cluster 1	7	23.22	109.65	68.56	2015/1/1 to 2017/12/31	4,959,129	124.7	1.31	1.33	652.23	< 0.001
Secondary cluster 2	5	21.76	108.06	67.10	2015/1/1 to 2017/12/31	1,699,035	122.6	1.29	1.29	183.72	< 0.001

Abbreviations: RR, Relative Risk; LLR, Log likelihood ratio.

Figure 5 Three-dimensional space-time clustering area, 2015–2019.

Table 4 Stepwise Regression Model Analysis Was Influencing Factors of Reported TB Incidence in Guangxi from 2015 to 2019

Variable	Non-Normalized Coefficients		Standardized Coefficient Beta	t	P	Beta (95% CI)		Tolerance Value
	Beta	Standard Error				Upper	Lower
Constant	148.604	15.420		9.637	<0.001	108.035	179.173
The number of beds per 1000 population in local healthcare institutions	8.720	3.357	0.642	2.598	0.011	2.065	15.374	0.121
The number of medical and technical personnel at healthcare institutions per 1000 population	−4.898	2.118	−0.582	−2.313	0.023	−9.096	−0.699	0.116
The per capita net income of rural residents	−0.005	0.001	−0.382	−4.200	<0.001	−0.008	−0.003	0.889

Abbreviation: CI, confidence interval.

Table 5 Comparison of Goodness of Fit of Four Bayesian Models

Model	DIC
Non-spatial effect model	20,143.000
Spatial effect model	20,144.200
Space-time effect model	20,143.200
Space-time Interaction effect model	5382.400

Abbreviation: DIC, deviance information criterion.

Table 6 Bayesian Space-Time Interaction Model Analysis of the Effect of Covariates (β) on TB Risk

Variable	Mean	sd	MC Error	95% CI
				Upper	Lower
Intercept	97.43	1.603	0.09859	94.52	99.87
The number of beds per 1000 population in local healthcare institutions	4.452	0.272	0.01674	4.092	4.998
The number of medical and technical personnel at healthcare institutions per 1000 population	2.151	0.06546	0.004027	2.086	2.300
The per capita net income of rural residents	−0.02282	6.932E-5	4.264E-6	−0.02295	−0.02271

Abbreviation: CI, confidence interval.

Figure 6 Comparison of relative risk and actual reported incidence of TB in Guangxi, 2015–2019.