Health risks of titanium dioxide (TiO) dust exposure in occupational settings - A scoping review.

Titanium dioxide (TiO) is an inorganic compound with many applications, for example in paint, sunscreen or as food coloring. There have been concerns regarding its safety and according to IARC, the existing evidence is not substantial enough to rule them out, leading to the substance being classified as possibly carcinogenic to humans (2B). This work aims to provide a comprehensible overview about epidemiological studies on occupational health risks and methodological aspects. A literature search was conducted in two databases (MEDLINE and Web of Science). The search focused on occupational exposure since this setting provides the highest amounts of TiO exposure. Of 443 unique search results, ten were included in this study, with publication dates ranging from 1988 to 2022. Seven of them are retrospective cohort studies and three have a case-control study design. Main outcomes of most studies were all-cause mortality and lung cancer mortality. For all-cause mortality, most cohort studies reported no association with TiO exposure. For lung cancer mortality, a significantly increased risk was found in a study population from Europe. The analysis results of working cohorts from the US comparing exposed workers' mortality rates with those of the general population were unobtrusive. However, one US cohort found an elevated mortality risk for all causes and lung cancer based on a reference population of company workers unexposed to TiO. Case-control studies did not indicate an increased risk for cancer related to TiO. Recent publications partly questioned the validity of those earlier findings, claiming insufficient confounder analysis, most notably for smoking, as well as the presence of the healthy worker effect, masking a potential health risk. In conclusion, the associations between occupational TiO exposure and mortality are unclear, but concerns regarding possible health risks recently re-emerged based on new analytical approaches, highlighting methodological difficulties that could have limited the inferential value of previously conducted studies.