Table 4.

Multinomial regression results. Adjusted^a within-person change in alcohol consumption (change in days, drinks, binge occasions) for an increase in off-premise alcohol outlets within a 1.6 km buffer^b. N = 714^c

		Distribution	Adjusted	95% confidence interval		P
	N	in the sample	odds ratiod	low	high	value
Change in number of drinking days per week
1. No change	288	40%	Referent
2. Increased	206	29%	1.64	1.00	2.68	0.049
3. Decreased	220	31%	1.23	0.76	2.00	0.409
Change in number of drinks per week
1. No change	239	33%	Referent
2. Increased	244	34%	1.55	0.95	2.55	0.081
3. Decreased	231	32%	0.97	0.58	1.62	0.908
Change number of binge occasions (past 30 days)e
1. No change	483	68%	Referent
2. Increased	126	18%	1.16	0.68	1.97	0.588
3. Decreased	105	15%	1.01	0.57	1.79	0.984

Abbreviations: CI confidence interval

^aAdjusted for age at baseline, gender, race/ethnicity, per capita income, education, history of chronic disease (binary), moved from ZIP code at follow-up, state at follow-up (Pennsylvania vs non-Pennsylvania), and population density per 1.6 km area (quartiles)

^bThe exposure is a binary variable: increase in outlets vs. no increase in outlets (referent category) using the measure ‘count of outlets in 1.6 km buffer’. The category for ‘decrease’ in outlets was not included because very few participants experienced a decrease in outlets. Per population standardization was not needed for the exposure variable in longitudinal model because the exposure was within-person change in outlet exposure and population density did not change much (because participants remained in their state). Nevertheless, we included population density (quartiles) as an adjustment variable in the model

^cThe sample size for this table slightly decreased (from N = 772 to N = 714). We deleted the 58 participants who experienced a decrease in alcohol outlet density during follow-up

^dOdds ratios derived from multinomial logit regression as appropriate for 3-level outcome: alcohol consumption no change (referent category), decrease, increase. Change defined as | > 0| days per week, | > 0| drinks per week, | > 0| binge days per month

^eBinge refers to past 30 days consumed a large volume of alcohol during a single occasion (> = 5 drinks for males, > = 4 drinks for females)