Environment/Climate Change
Association of PFAS Exposure with Prevalent Prediabetes or Diabetes among Adults in the United States: Effect Modification by Physical Activity and Dietary Fiber Intake Jennifer Arzu* Jennifer Arzu Arzu Brown University School Public Health
Per- and polyfluoroalkyl substances (PFAS) exposure have been linked to type 2 diabetes. However, few studies have investigated lifestyle behaviors that may modify these associations. We explored whether the associations of PFAS with prevalent prediabetes or diabetes differ by moderate-to-vigorous physical activity (MVPA) and dietary fiber intake.
This study analyzed data from 813 non-pregnant adults ages ≥20 years in the 2017–2020 U.S. National Health and Nutrition Examination Survey. Using survey-weighted modified Poisson regression models, we estimated the covariate-adjusted prevalence ratios of prediabetes/diabetes (i.e., self-report, fasting plasma glucose≥100 mg/dL, or HbA1c≥5.7%), diabetes only (i.e., self-report, fasting plasma glucose≥126 mg/dL, or HbA1c≥6.5%), and prediabetes only (i.e., self-report, fasting plasma glucose≥100 and <126 mg/dL, or HbA1c≥5.7% and <6.5%, excluding diabetes) per doubling in serum perfluorohexane sulfonic acid (PFHxS), perfluorooctane sulfonic acid (PFOS), perfluorooctanoic acid (PFOA), perfluorononanoic acid (PFNA) concentrations. We used quantile g-computation models to estimate these associations per 1-quartile increase in PFAS mixture.
We observed no overall associations of PFAS with prevalent prediabetes or diabetes. Among adults who were less physically active (MVPA<150 minutes/week), higher PFOS, PFOA, and PFAS mixture were associated with 7% [PR:1.07 (95% CI:1.00,1.14)], 15% [PR:1.15 (95% CI:1.06,1.26)], 11% [PR:1.11 (95% CI:1.00,1.22)] higher prediabetes/diabetes prevalence, but with 4% [PR:0.96 (95% CI:0.91,1.02); pinteraction=0.078], 9% [PR: 0.91 (95% CI:0.82,1.01); pinteraction=0.005], and 8% [PR:0.92 (95% CI:0.83,1.02); pinteraction=0.005] lower prevalence among adults meeting the MVPA guidelines (≥150 minutes/week). We found similar patterns of associations between PFAS and prevalent prediabetes across physical activity levels. The associations of PFHxS, PFOS, PFOA, PFNA, and PFAS mixture with prevalent diabetes were consistently positive among adults with lower fiber intake (≤median of 6.8 gram/1000 kcal) and inverse among adults with higher fiber intake (>median of 6.8 gram/1000 kcal) [pinteraction=0.0001, 0.001, 0.002, 0.033, and 0.002 respectively]. For example, each doubling in PFHxS was associated with 24% [PR:1.24 (95% CI:1.01,1.53)] higher diabetes prevalence in adults with lower fiber intake, but 21% [PR:0.79 (95% CI: 0.66,0.94)] lower prevalence in adults with higher fiber intake.
Physical activity levels and dietary fiber intake modified the associations of individual PFAS and their mixture with prediabetes or diabetes prevalence.
