Incidence of Tuberculosis and Associations with Indicators of Alcohol Consumption in Three Regions of Northwest Russia in 1975–2009: A Time-Series Analysis

Table 1

Estimated parameters for ARIMA model to forecast incidence of tuberculosis for alcohol psychosis and for alcohol poisoning deaths in Arkhangelsk, Vologda, and Murmansk regions, 1975–2009.

	Coefficient	SE (95% CI)
ARIMA models for alcohol psychosis*
TB Arkhangelsk region,; Model 1 ARIMA (0,1,0)	.24	.09 (0.10–0.37)
TB Vologda region,; Model 2 ARIMA (0,1,0)	.18	.04 (0.10–0.25)
TB Murmansk region,; Model 3 ARIMA (0,1,0)	−.10	.83 (−1.74–1.53)
ARIMA models for alcohol poisoning deaths*
TB Arkhangelsk region,; Model 1 ARIMA (0,1,0)	−.52	.80 (−2.1–1.0)
TB Vologda region,; Model 2 ARIMA (0,1,0)	−.19	.61 (−1.72–1.0)
TB Murmansk region,; Model 3 ARIMA (0,1,0)	−.10	.83 (−1.74–1.5)

∗0, 1, and 0 means 𝑝=0, 𝑑=1, and 𝑞=0, where p is the number of autoregressive terms, d is the number of nonseasonal differences, and q is the number of lagged forecast errors in the prediction equation.