Table 3. Estimates of Temporal Associations of Alcohol and Tobacco Consumption With Sex- and Age-Specific Cancer Mortality Based on the Cross-Correlation Lag Model.

Characteristic	Consumption, Coefficient (95% CI)		Model Specification	Box-Ljung Q (Lag 10)
	Alcohol	Tobacco		Q Statistic	P Valuea
Male by age, y
30-49	0.032 (−0.137 to 0.201)	0.137 (−1.080 to 1.354)	1,1,0	6.404	.70
50-69	0.095(0.040 to 0.150)b	0.170 (0.029 to 0.311)c	1,1,0	10.883	.28
≥70	0.016 (−0.035 to 0.067)	0.263 (0.075 to 0.451)d	0,1,0	8.350	.60
Subtotal	0.043 (0.012 to 0.074)c	0.266 (0.115 to 0.417)d	1,1,0	9.546	.39
Female by age, y
30-49	0.022 (−0.051 to 0.095)	0.070 (−0.181 to 0.321)	1,1,0	5.080	.75
50-69	0.059 (0.008 to 0.110)d	0.063 (−0.151 to 0.277)	0,1,0	8.576	.57
≥70	0.042 (0.020 to 0.064)c	0.067 (−0.015 to 0.149)	0,1,1	7.357	.60
Subtotal	0.035 (0.010 to 0.060)c	0.083 (−0.066 to 0.232)	0,1,1	4.996	.84

^aThe Box-Ljung Q test is a diagnostic tool used to test the lack of fit of a time series model, and a P value of the Box-Ljung Q test greater than .10 indicates the test rejects the null hypothesis of lack of fit of the time series model.

^bP < .001.

^cP < .05.

^dP < .01.