Introduction Work in primary coffee factories exposes workers to various hazardous agents. At farms coffee is dried on mat or on the ground (Robusta coffee) or on sieves (Arabica coffee). Arabica is depulped and washed before drying. Afterwards, the coffee is brought to primary coffee factories, and the study will determine the level of exposure to dust and endotoxin and respiratory health effects among workers in these factories.

Methods Personal total dust samples were collected from four primary coffee factories n = 193. These were analysed gravimetrically and out of those; 154 samples were analysed for endotoxin. A standardised questionnaire adapted from ATS was used. Asthma symptoms questions from ERCHS questionnaire were added. Lung function test were performed (n = 133). Determinants of exposure and correlation to lung function were analysed using linear mixed-effect model and linear regression

Results The personal total dust and endotoxin exposure levels were higher in Robusta coffee factories (Geometric Mean 3.42 mg/m³ and 10800 EU/m³) compared to Arabica (GM 2.10 mg/m³ and 1380 EU/m³). Working with dry pre-processed coffee (Robusta) increased endotoxin levels 7 times compared to washed coffee. Workers in Robusta factories had more asthma symptoms (38% vs. 18% ORadj 3.5; 95% CI; 1.4–9.0) compared to Arabica workers. A decrease of 21.4 ml/s in FEV₁ and 0.22% in FEV₁/FVC ratio in lung function may be realised by a worker exposed to endotoxin in these factories for one season.

Conclusion Workers in coffee factories are exposed to high levels of endotoxin and likely to develop respiratory health effects. The respiratory symptoms are seen even at lower levels of dust exposure thus suggesting occupational exposure limit (OEL) for organic dust might be safer if derived from endotoxin levels.