Aims Due to the stressful nature of policing, police employees are at risk of mental health problems and problematic alcohol use. We aim to determine the prevalence of hazardous and harmful alcohol use in the UK Police Service, and to explore the associations with job strain and mental health problems. Methods Cross-sectional data from the Airwave Health Monitoring Study (N = 40 986) included measures of alcohol consumption (total units in past week), mental health (depression, anxiety and post-traumatic stress disorder [PTSD]) and job strain. The associations between mental health and job strain with alcohol consumption (i.e. abstinence, low-risk [<14 units per week, reference group], hazardous [>14 to 35 units for women, >14 to 50 units for men], harmful [>35 units for women, >50 units for men]), were analysed using multinomial logistic regressions, adjusting for potential confounders (i.e. age, gender, ethnicity, marital status, children under 18, income and smoking status). Results A total of 32.6% of police employees reported hazardous drinking, with 3.0% drinking at harmful levels. Compared to those without a mental health problem, police employees with depression, anxiety or PTSD were twice as likely to be harmful drinkers and were also 1.3 times more likely to report abstinence. Those reporting low strain (reference group) were more likely to drink hazardously compared to those reporting high strain, which was statistically moderated by mental health. When the sample was stratified by mental health status, the association between low strain (compared to all other categories) and hazardous drinking, was significant only in those without a mental health problem. Conclusions These findings indicate that police employees may be an occupational group at risk of alcohol harm, with one-third drinking hazardously. The J-shaped relationship between mental health and alcohol use highlights a need for an integration of mental health and alcohol services, tailored for the UK Police Service.