Objectives To evaluate the possible associations between occupational exposure to cleaning products and cancers of the following sites: bladder, pancreas, prostate, colon, stomach, kidney, non-Hodgkin’s lymphoma, and lymphoid tissue (Hodgkin’s and Non-Hodgkin’s lymphomas and myelomas combined).

Methods We conducted a case-control study of occupational exposures and cancer in Montreal including 3730 cancer cases and 533 population controls. Occupational exposure to a set of agents was evaluated using a combination of subject-reported job history and expert assessment. We evaluated the association between each of eight cancer sites and each of eight cleaning-related occupations, and each of seven cleaning-related exposures (ammonia, hypochlorite, spray gases, waxes/polishes, caustic solutions, chlorine, and cleaning agents as a class). Analyses were done using multivariate logistic regression.

Results In general, we found no increased risk of cancer for the evaluated occupations as compared to never cleaners with the exceptions of pancreatic cancer (Odds Ratio; 95% CI (OR): 3.0; 1.3–6.8) and cancers of lymphoid origin (lymphomas and myelomas) (OR: 2.1; 1.1–4.1) in relation to long term employment as ‘janitors and cleaners’. Among specific agents, we found an indication of excess risk for substantial exposure to ammonia and hypochlorite with a pooled set of lymphoid cancers (Non-Hodgkin’s lymphoma, Hodgkin’s lymphoma and myeloma) (OR: 2.2; 1.2–4.0 and 2.0; 1.0–4.0, respectively). For the rest of the agents and cancer sites evaluated we found no evidence of increased risk.

Conclusions Overall, our results do not support a clear association between occupational exposure to cleaning products or cleaning-related occupations and the examined cancer sites, with the exception of a potential link between lymphoid cancers (Hodgkin’s and Non-Hodgkin’s lymphomas and myeloma) and exposure to ammonia and to hypochlorite. Elevated ORs among janitors and cleaners should be further explored. Our results must be interpreted in a context of multiple hypotheses testing.