Background
Previous research indicates an association between higher-chlorinated polychlorinated biphenyls (PCBs) and type 2 diabetes (T2D). However, less is known about the extent to which PCB exposure in indoor air, composed primarily of lower-chlorinated PCBs, affects T2D risk. We assessed the association between indoor air exposure to PCBs in residential buildings and T2D incidence.

Methods
The register-based ‘Health Effects of PCBs in Indoor Air’ (HESPAIR) cohort comprises 51,921 Danish residents of two residential areas with apartments built with and without PCB-containing materials (reference apartments). We assessed exposure status by combining register-based information on relocation history with extrapolated values of exposure based on PCB-measurements in indoor air from subsets of the apartments. T2D cases were identified in the Danish registers during 1977–2018. We estimated adjusted hazard ratios (HR) and 95% confidence intervals (CI) using Cox regression analyses with time-varying exposure.

Results
We identified 2737 incident T2D cases during the follow-up. Exposure to ≥3300 ng/m3 PCB × year (3rd tertile of PCByear) was associated with higher risk of T2D (HR 1.15, 95% CI 1.02–1.30) compared with exposure to <300 ng/m3 PCB × year (reference). However, among individuals with lower cumulated PCByear, the risk was similar to residents with exposure <300 ng/m3 PCB × year (300–899 ng/m3 PCB × year: HR 0.98, 95% CI 0.87–1.11; 900–3299 ng/m3 PCB × year: HR 0.96, 95% CI 0.83–1.10).

Discussion
We observed a marginally higher risk of T2D, but there was no evidence of an exposure-response relationship. The results should be interpreted with caution until confirmed in other independent studies of PCB exposure in indoor air.