Fig. 2

Correlation between HDI and MIR and its transition from 2012 to 2018. The patterns of urologic cancer MIRs to national HDIs with the best-fit lines by modified nonlinear regression (“dose-to-response” model) were presented as following: a prostate cancer in 2018 (slope = − 1.962, HDI₅₀ = 0.639, R² = 0.687) and b in 2012 (slope = − 3.177, HDI₅₀ = 0.713, R² = 0.891); d bladder cancer in 2018 (slope = − 1.967, HDI₅₀ = 0.704, R² = 0.733) and e in 2012 (slope = − 1.720, HDI₅₀ = 0.640, R² = 0.835); g kidney cancer in 2018 (slope = − 2.178, HDI₅₀ = 0.736, R² = 0.737) and h in 2012 (slope = − 2.835, HDI₅₀ = 0.780, R² = 0.824). MIRs of c prostate, f bladder and i kidney cancer in the 4 HDI groups, with significant differences among the very high, high, medium and low groups and a decreasing tendency in certain groups between 2012 (light purple) and 2018 (dark purple). #### p < 0.0001, vs. very-high-HDI countries in 2018, one-way ANOVA followed by Tukey–Kramer post hoc test. The statistical significance among countries in 2012 was not indicated. **p < 0.01, ***p < 0.001, ****p < 0.0001, 2008 vs. 2018 in specific corresponding group, unpaired t-test