Table 4.

Change in maximum alcohol consumed, number of alcohol‐related harms, drinking context and use of delivery services during COVID‐19 restrictions compared to February 2020: results of mixed‐effects models.

	Maximum alcohol consumed in month a	Number of alcohol‐related harms b	Drinking alone c	Drinking with others in person c	Drinking with others virtually c	Use of alcohol delivery services d
	Coeff. (95% CI)	IRR (95% CI)	Coeff. (95% CI)	Coeff. (95% CI)	Coeff. (95% CI)	OR (95% CI)
February 2020	Ref	Ref	Ref	Ref	Ref	Ref
May/June 2020	−1.58 (−1.99, −1.16)	0.65 (0.54, 0.79)	2.15 (0.50, 3.81)	−13.80 (−17.56, −10.04)	7.80 (5.57, 10.02)	0.68 (0.43, 1.08)

^a Analysed using linear mixed‐effects models, with results presented as a coefficient equating to the difference in the mean maximum alcohol consumed.

^b Analysed using negative binomial mixed‐effects models, with results presented as incidence rate ratios (IRRs).

^c Analysed using linear mixed‐effects models with results presented as coefficients equating to the difference in the mean proportion of consumption in each setting.

^d Analysed using logistic mixed‐effects models, and presented as odds ratios (ORs). Models are adjusted for covariates; full results are included in Supporting information, Table E4. Results are based on multiple imputation with estimates combined using Rubin's rules.