Case-Only Approach Applied in Environmental Epidemiology: Two Examples of Interaction Effect Using the NHANES Dataset

Abstract

BACKGROUND AND AIM: By substituting the general ‘susceptibility factor’ concept for the conventional ‘gene’ concept in the gene-environment interaction case-only approach, this study can broaden the applicability of the case-only approach to environmental epidemiology. METHODS: This study collectively summarized the logics and equations previously reported in methodology studies with a case-only approach. RESULTS:(1) When 1 unit of both blood chromium level (mcg/L) and blood glycohemoglobin level (%) increased coincidentally, the multiplicative interaction risk ratio was 0.72 (95% CI 0.47-1.05), with no statistical significance. However, when only the cases were analyzed (case-only approach), the case-only interaction odds ratio was 0.67 (95% CI 0.44-0.90), with a statistical significance (a negative interaction effect). (2) When 1 unit of both blood cobalt level (mcg/L) and age in years increased coincidentally, the multiplicative interaction risk ratio was 1.13 (95% CI 0.99-1.37), with no statistical significance. However, the case-only interaction odds ratio of 1.21 (95% CI 1.06-1.51) was calculated using the case-only approach with a fine-tuning. CONCLUSIONS:If the independence assumption between susceptibility factor and environmental exposure in the entire population is kept, this case-only approach can provide (ⅰ) a more precise interaction effect estimate or (ⅱ) a reliable interaction effect estimate when appropriate controls were not available. The two examples using the NHANES dataset must be investigated meticulously in the following studies. In summary, the case-only approach can be useful either in environmental epidemiology, not only in genetic epidemiology. KEYWORDS: Environmental epidemiology, Heavy metals