Background Prior meta-analysis and IARC reviews suggested no significant association between non-asbestiform talc exposure and increased risks of lung cancer, without including related Chinese literatures. A meta-analysis including eligible Chinese cohorts was performed to update the current evidence.

Methods We searched the MEDLINE, EMBASE, through 2015 for a meta-analysis examining the association between non-asbestiform talc exposure of workers and increased risks of lung cancer. We also searched China National Knowledge Infrastructure (CNKI) database and Wanfang Data, because many studies on talc exposure of workers have been published in Chinese journals only. Data from observational studies were pooled using a general variance based meta-analytic method. The outcome of interest was a summary standardised mortality ratios (SMRs) reflecting the risk of lung cancer death associated with non-asbestiform talc exposure of workers.

Results Seven observational studies with eight cohorts meeting protocol specified inclusion criteria were located via a comprehensive literature search. Analysis for heterogeneity demonstrated the data were heterogeneous (p = 0.0003). Pooling all cohorts yielded a summary SMRs of 1.58 (CI = 1.12–2.22, p = 0.0091, random effect model) for lung cancer among workers exposed to non-asbestiform talc, showing a 58% increase in stomach cancer mortality for these workers compared to general population. Subgroup analysis for co-exposures contamination showed workers exposed to talc without quartz/radon contamination had a summary SMRs of 2.17 (CI = 1.29–3.66, p = 0.0036, random effect model), and workers exposed to talc with quartz/radon contamination had a summary SMRs of 1.14 (CI = 0.87–1.50, p = 0.3345, random effect model).

Conclusion The available observational data showed increased lung cancer risk in workers exposed to non-asbestiform talc with quartz/radon co-exposures. However, existence of a causal relationship between non-asbestiform talc without quartz/radon co-exposures and increased risks of lung cancer cannot be established.