Genetic predisposition to television watching increases the risk of type 2 diabetes: a bidirectional and multivariable Mendelian randomization study

Songzan Chen,Fangkun Yang,Tian Xu,Yao Wang,Kaijie Zhang,Guosheng Fu,Wenbin Zhang
DOI: https://doi.org/10.21203/rs.3.rs-58026/v1
2020-01-01
Abstract:Abstract Background: Excessive sedentary behaviors have been reported to be associated with increased risk of type 2 diabetes, but whether this association is causal remains unclear. In current study, we aimed to investigate the causal association between domain-specific sedentary behaviors and the risk of type 2 diabetes using a two-sample Mendelian randomization (MR) study. Methods: We identified 165 single nucleotide polymorphisms as instrumental variables for television watching, 43 for computer use and 5 for driving behavior from a recently published genome-wide association study (n = 408,815). Genetic association estimates for type 2 diabetes were obtained from the DIAbetes Genetics Replication And Meta-analysis (DIAGRAM) consortium (74,124 cases and 824,006 controls). The inverse variance-weighted method was used to estimate the effect of genetically predicted sedentary behaviors on the risk of type 2 diabetes. Reverse MR analysis was performed to investigate the reverse causation. The weighted median method, MR-Egger method, and MR Pleiotropy Residual Sum and Outlier method were employed in the sensitivity analyses. In addition, multivariable MR analysis and mediation analysis were conducted to explore the potential mechanistic elements.Results: Genetic predisposition to excessive television watching was associated with increased risk of type 2 diabetes. The OR (95% CI) per 1.5h (1 standard deviation) increment in television watching time was 1.82 (1.61, 2.07) for type 2 diabetes. This association was substantially attenuated after adjustment for anthropometric traits (adjusting BMI: OR = 1.35, 95% CI = 1.17 – 1.57, P = 4.1 × 10-5; adjusting WHR: OR = 1.26, 95% CI = 1.09 – 1.45, P = 1.4 × 10-3) and educational attainment (OR = 1.49, 95% CI = 1.16 – 1.91, P = 1.7 × 10-3). There was limited evidence of associations of computer use and driving behavior with the risk of type 2 diabetes. Conclusions: Our study clarifies the causal effect of excessive television watching on the increased risk of type 2 diabetes from a genetic perspective, which may be partly mediated via anthropometric and educational traits. Television watching may serve as a behavioral target to prevent incident diabetes.
What problem does this paper attempt to address?