Message on a bottle: are alcohol warning labels about cancer appropriate?

BMC Public Health

Table 4 Factors predicting statement agreement: multivariate analysis

Statement	Factor	Agreement ratio (95 % CI)	Agreement difference (95 % CI)	p-value
Raise awareness about the link between regular alcohol consumption and cancer	Female sex	1.07 (1.04–1.10)	0.04 (0.01–0.07)	<0.001
	Risk	1.32 (1.22–1.44)	0.24 (0.17–0.30)	<0.001
Prompt conversations about the cancer risk involved in drinking alcohol regularly	Reassess	1.87 (1.61–2.18)	0.45 (0.36–0.53)	<0.001
Prompt me to drink alcohol less often	Reassess	8.59 (5.13–14.40)	0.56 (0.48–0.53)	<0.001
Prompt my friends to drink alcohol less often	≥43 years	1.40 (1.05–1.86)	0.09 (0.04–0.14)	0.024
	Risk	9.45 (4.50–19.90)	0.37 (0.30–0.46)	<0.001
	AUDIT-C <4	1.42 (1.06–1.89)	0.11 (0.02–0.16)	0.019
Prompt me to talk to my family/friends about the cancer risks associated with alcohol	Reassess	3.44 (2.61–4.54)	0.52 (0.22–0.63)	<0.001
Prompt me to educate my children about the cancer risk associated with alcohol	Reassess	14.71 (8.16–26.51)	0.49 (0.41–0.58)	<0.001

Notes. Agreement that with the statement about generally reassessing behaviour based on warning labels is abbreviated to ‘Reassess’; prefer to have product labels so I know what the risks are abbreviated to ‘Risk’
All models included age and sex

ISSN: 1471-2458