Table 4.

Comparative ITS analysis model parameters of population-standardized quarterly notification rates of all-form and bacteriologically confirmed TB cases for intervention vs. control districts ^¥. Hai Phong and Quang Nam, Viet Nam. January 2018 to March 2019.

Comparative ITS Analysis Model Parameters	Intervention vs. Control Districts
	IRR ‡	95% CI	p-Value þ
All Forms TB
Baseline rate (β0)	21.563	(20.108, 23.124)	<0.001
Preintervention trend, control (β1)	0.998	(0.990, 1.006)	0.590
Postintervention step change, control (β2)	1.116	(0.952, 1.308)	0.178
Postintervention trend, control (β3)	0.977	(0.923, 1.034)	0.427
Difference in baseline (β4)	1.378	(1.270, 1.495)	<0.001
Difference in preintervention trends (β5)	0.977	(0.967, 0.986)	<0.001
Difference in postintervention step change (β6)	1.221	(1.011, 1.475)	0.038
Difference in postintervention trends (β7)	1.015	(0.948, 1.086)	0.676
Bacteriologically confirmed TB
Baseline rate (β0)	11.107	(9.562, 12.901)	<0.001
Preintervention trend, control (β1)	0.992	(0.975, 1.009)	0.361
Postintervention step change, control (β2)	0.807	(0.587, 1.109)	0.186
Postintervention trend, control (β3)	1.043	(0.935, 1.163)	0.448
Difference in baseline (β4)	1.141	(0.956, 1.362)	0.144
Difference in preintervention trends (β5)	1.001	(0.981, 1.021)	0.928
Difference in postintervention step change (β6)	1.535	(1.067, 2.210)	0.021
Difference in postintervention trends (β7)	0.902	(0.796, 1.023)	0.108

^¥ The parameters were obtained for a segmented regression model with the following structure: $Y_t={\beta }_0+{\beta }_1{T}_t+{\beta }_2{X}_t+{\beta }_3{X}_t{T}_t+{\beta }_{4}Z+{\beta }_{5}Z{T}_t+{\beta }_{6}Z{X}_t+{\beta }_{6}Z{X}_t{T}_t+{ϵ}_t$. Here, Y_t is the outcome measure along time t; T_t is the monthly time counter; X_t indicates pre- and postintervention periods, Z denotes the intervention cohort, and ZT_t, ZX_t, and ZX_tT_t are interaction terms. β₀ to β₃ relate to the control group as follows: β₀, intercept; β₁, preintervention trend; β₂, postintervention step change; β₃, postintervention trend. β₄ to β₇ represent differences between the control and intervention districts: β₄, difference in baseline intercepts; β₅, difference in preintervention trends; β₆, difference in postintervention step changes; β₇, difference in postintervention trend. ^‡ IRR is based on a log-linear GEE Poisson regression with correlation structures, as determined by the Cumby–Huizinga test and Quasi-Information Criteria; ^Þ, Wald test.