Is obesity a risk factor for melanoma?

As McQuade et al. [18] suggest: “Although the impact of obesity associated inflammation on carcinogenesis has been well studied, the impact of energy balance on anti-tumor immune response has not been examined to date and should be investigated as a potential explanation underlying the observed interaction between BMI and both targeted and immune therapy.” (page 320).

This calculation transforms the variable to percentage points, so that obesity prevalence of X percent in the population (0 ≤ X ≤ 100) equals X%.

The information was obtained from Center for Disease Control and Prevention (CDC) [6]: United States Cancer Statistics, available at: https://​www.​cdc.​gov/​cancer/​uscs/​ [last accessed on December 22, 2022]. The reason for the white-black choice is our findings these groups have the most extreme differences in terms of melanoma prevalence. In this context, see the graph at Additional file 1: Appendix S1.

Incorporation of the time variable (Year − 2011) addresses the possibility of spurious or nonsense correlation in time series analysis. According to Johnston and Dinardo [14], series, responding to unrelated mechanisms, such as, death rates in England and Wales and the proportion of all marriages solemnized in the Church of England from 1866 to 1911 [29], may display contemporaneous upward or downward movement. This problem may be addressed by fitting trends to such series. Regression outcomes reported in Table 6, demonstrate that while melanoma prevalence rises by 0.336−0.384% per annum only for the white population, obesity prevalence increases by 0.382−0.465% per annum only for the black population.

Calculations of the figures were carried out in Stata software package version 16 via the following commands:

• margins White, at(Obesity_prevalence==(10(1)55))

• marginsplot, noci

• margins White, at(Obesity_prevalence==(10(1)55)) contrast

• marginsplot

The “margins” and “marginsplot” commands come after the corresponding regression analysis.