AIMS
To evaluate the effectiveness of a French mass media campaign in raising knowledge of both long-term alcohol-related harms (LTH) and low-risk drinking guidelines (LRDG), as well as in lowering alcohol consumption.

DESIGN
An 8-month longitudinal survey from February to October 2019, with three waves of online data collection (T0 before the campaign, T1 just after it ended, and T2 6 months after it ended).

SETTING
France PARTICIPANTS: 2,538 adult drinkers (18-75 years old) MEASUREMENTS: The main outcomes' variables were LTH knowledge (cancer, hypertension, brain haemorrhage), LRDG knowledge (two guidelines: 'maximum of 2 drinks a day' and 'minimum of 2 days without alcohol per week'), intention to reduce alcohol consumption, and self-declared consumption with respect to the French LRDG. At T1, exposure to the campaign was measured using self-reported campaign recall.

FINDINGS
In T1, we observed significant positive interactions between exposure group based on campaign recall and survey waves on knowledge of i) the 'maximum 2 drinks a day' guideline (adjusted odds ratio [aOR] 1.32 [95% confidence interval 1.08-1.62], p=0.008), ii) brain haemorrhage (aOR = 1.80 [1.44-2.25], p<0.001), and iii) hypertension (aOR = 1.41 [1.09-1.81], p=0.008) risks. Campaign exposure was also associated with a significant decrease in at-risk drinking in women (aOR=0.67 [0.50-0.88], p=0.004). No significant interaction was observed at T1 for the knowledge of the 'minimum of 2 days without alcohol a week' guideline, or of cancer risk. At T2, no significant interaction was observed for the main outcomes' variables.

CONCLUSIONS
There appears to be an association between exposure to a 2019 French mass-media campaign to raise knowledge of long-term alcohol-related harms and low-risk drinking guidelines (LRDG) and reduce alcohol consumption and i) improved knowledge of the 'maximum 2 drinks per day guideline', ii) knowledge of the risks of hypertension and brain haemorrhage, and iii) a reduction in the proportion of people exceeding LRDG (in the general population only). These associations were only observed over the short term and, in some cases, only for certain segments of the population.