Observational studies suggest that modifiable risk factors such as low vitamin D levels, coffee consumption, alcohol consumption and obesity may be important in cancer risk. However, observational studies can only demonstrate association between a risk factor and cancer, and association does not equal causation. We will apply a causal inference approached called Mendelian randomization. This approach will help determine which risk factors actually cause cancer. This work will help inform cancer prevention efforts, ultimately reducing the overall cancer burden.