Introduction
Risky drinking (RD) is associated with an increased risk of chronic and infectious diseases, injuries, and violence. This study aimed to assess changes in risky drinking (RD) in Brazil after COVID-19 outbreak, both overall and among individuals with self-reported chronic diseases and mental health disorders.

Methods
We conducted three independent, anonymous web surveys in Brazil including adult participants: S1 (April/2020, n=19,257), S2 (August/2020, n=1,590), and S3 (January/2021, n=859). Participants were recruited through adapted snowball sampling and sponsored social network advertisements. RD was assessed using the Alcohol Use Disorder Identification Test-Concise, designed to identify individuals at risk of alcohol-related problems. Logistic regression analyses with bootstrapping (B=2,000) were performed, with stratification by sex, age, education, employment, household size, and the presence of chronic and mental health conditions, as well as lifestyle factors, to address sample imbalances.

Results
The estimated prevalence of RD was 45.8% [95%CI 45.5, 46.1] in S1, 35.3% [95%CI 34.9, 35.6] in S2, and 33.7% [95%CI 33.3, 34.0] in S3. Participants with chronic diseases consistently presented lower RD prevalence across all three surveys, compared to those without such conditions. Conversely, individuals with mental health disorders presented higher RD prevalence than those without such diagnoses in S1 and S2, but not in S3.

Discussion
Despite the decrease in RD prevalence, monitoring of alcohol consumption trends remains essential for shaping effective public health policies. Additionally, the observed variations among individuals reporting chronic and mental health disorders highlight the need for targeted interventions in future crises.