Introduction Work is considered a key determinant of mental health. However, evidence on the impact of work in mental health in Latin America is scarce.

Objectives To analyse the relationship between mental health status and work in Latin America through three indicators: labour relationship (employed/self-employed or employer), employment condition (formal/informal), and occupational category (non manual skilled/non manual non-skilled/manual skilled/manual non-skilled).

Methods Cross sectional study based on health and working condition surveys from 8 countries of Latin America. Poor mental health was defined as a score of > = 3 on the GHQ12; >= 10 on the PHQ9; or <= 13 on the WHO5 scales. Informal workers were defined as those unregistered or lacking a contract. Occupational categories were created following the international standard classification of occupation. Prevalence ratios and their IC95% were calculated stratified by sex and adjusted by age, educational level and marital status.

Results Final sample consisted of 72,452 workers, from Argentina (n=8966), Brazil (n=52832), Costa Rica (n=1503), El Salvador (n=1507), Guatemala (n=1510), Honduras (n=1507), Nicaragua (n=1500) and Panamá (n=1505). Poor mental health ranged from 4.5% in Brazilian men to 25.9% in Panamanian women. Manual non-skilled men showed higher prevalence of poor mental health than non-manual skilled men in Argentina (RPa 1.43 IC95% 1.04 - 1.95) and Central America (RPa 1.94 IC95% 1.12 - 3.34). In Brazil, men employers showed higher prevalence of poor mental health (RPa 1.25 IC95% 1.02 - 1.53) and informal workers showed higher prevalence of poor mental health irrespective of sex (Men RPa 1.16 IC95% 1.02 - 1.33; Women RPa 1.30 IC95% 1.05 - 1.60).

Conclusion Prevalence of poor mental health varies significantly between occupational groups and countries. Special attention should be pay to informal workers, workers in elementary occupations and employers.