Skin Cancer-Precentive Effects of Caffeinated Coffee in Men and Women

Skin cancer is the most prevalent cancer in the U.S., with an annual incidence of 5.5 million, exceeding all other cancers combined. Given the extremely high incidence of skin cancer, preventive measures are important. Intriguingly, multiple human epidemiological studies have demonstrated that caffeinated coffee, which is widely consumed in the U.S., is associated with decreased risk of developing skin cancer in a dose-dependent manner. Importantly, decaffeinated coffee had no such preventive effect. To determine the per-cup effect of caffeinated coffee consumption on skin cancer prevention, we performed linear regression analyses of previously published data. Males showed 3.0% reduced risk per cup of caffeinated coffee, whereas females showed 4.4% reduced risk.

To determine the average reduced risk among males and females in the U.S. population, I extracted coffee consumption data from the National Health and Nutrition Examination Survey (NHANES) 2015–2016. Subsequently, the average number of cups of caffeinated coffee consumed was combined with the per-cup effect on skin cancer prevention. Among caffeinated coffee drinkers, males and females consumed on average 2.9 and 2.0 cups, respectively. However, both male and female caffeinated coffee drinkers coincided with 9% reduced risk of skin cancer incidence even though coffee consumption and the per-cup effect on risk reduction were different between males and females. The cancer-preventive effects of caffeinated coffee per cup appear to be stronger in females than in males. However, greater coffee consumption in males with weaker per-cup effect on skin cancer prevention leads to the net effect in males being comparable to that in females. For both genders, current levels of caffeinated coffee consumption may be preventing thousands of skin cancers annually.

From this experience, I improved my research paper reading and writing skills, including figure making. Since I had to create Python code (unable to the public) to analyze data from 9,000+ participants, my Python skills also improved. I presented my findings at the University of Washington 2020 Undergraduate Research Symposium, so my presenting skills also improved as well.

Acknowledgements: Dr. Masoki Kawasumi and Jihoon Lee