Objectives: The purpose of this study was to determine the level, pattern and correlates (socio-demographic, lifestyle and work-related) of physical activity among middle-aged employees in Croatia. Material and Methods: In this cross-sectional study the data were collected using a household interview on a random sample of 766 middle-aged employees (52% female) living in Croatia. The International Physical Activity Questionnaire (IPAQ-long) was used to assess physical activity. An additional questionnaire was used to collect data on socio-demographic, lifestyle and work-related correlates. Results: The median (95% CI) total physical activity for the whole sample was 78.7 (69.1-88.3) MET-hours/week. Most physical activity was accumulated in the domain of work (26.4 (20.3-32.5) MET-hours/week) or in domestic activities (19.2(17.8-20.7) MET-hours/week), whilst a significantly lower physical activity was found in the transport (3.3 (2.9-3.7) MET-hours/week) and leisure-time domains (6.5 (5.7-7.3) MET-hours/week). The multiple regression analysis showed an inverse relationship between educational level and the size of settlements with the domestic-related and total physical activity (β range: -0.11 to -0.22; p < 0.01). Among lifestyle correlates, only alcohol consumption was positively related to the total physical activity (β = 0.12; p < 0.01), while the perceived level of stress outside work was the only one, which was inversely associated with the work-related (β = -0.09; p < 0.01) and leisure-time physical activity (β = -0.10; p < 0.01). Conclusions: Middle-aged employees in Croatia accumulate most of their daily physical activity in the work and domestic domains. Analysis of the relationship between physical activity and potential socio-demographic, lifestyle, and work-related correlates indicated that physical activity promotional activities should be primarily focused on males, employees living in smaller settlements and those with higher educational levels. The correlates of physical activity among middle-aged employees seem to be domain-specific. Therefore, future studies in this area should consider assessing physical activity in each domain separately.