In 1991, IARC published an evaluation of the carcinogenic risks to humans of chlorinated drinking water based on a number of animal toxicological and epidemiological studies. IARC concluded that, because of one or more methodological weaknesses, the epidemiological studies reviewed cannot constitute the basis of valid risk assessment.

The epidemiological investigation of the relation between exposure to chlorinated drinking water and cancer occurrence was considered problematic, because any increase in relative risk over that in people drinking unchlorinated water is likely to be small and therefore difficult to detect in epidemiological studies. In all of the studies evaluated, estimates of exposure to chlorinated by-products were imprecise, and surrogates of exposure (e.g., surface versus groundwater) may not reflect exposure during the relevant time periods for the etiology of the cancers in question. Many variables, such as smoking habits, dietary practices, use of alcohol, socioeconomic status, and ethnicity are known to affect cancer incidence and were not taken into account in most of the epidemiological studies (IARC 1991).

In its overall evaluation published in 1991, IARC concluded that there is inadequate evidence for the carcinogenicity of chlorinated drinking water in humans as well as in experimental animals (IARC 1991). In 1992, an international conference on the safety of water disinfection considered additional epidemiological and toxicological studies published after the IARC review and concluded that the current scientific information was still inadequate to establish the carcinogenicity of chlorinated drinking water (Craun 1993).