Risk adds up. This code piece answers, "how quickly?" https://twitter.com/statwonk/status/1160542394544267265

risk_adds_up.R

library(tidyverse)
expand.grid(
  risk = seq(0.001, 0.02, 0.001),
  units_of_exposure = seq_len(24*7)
) %>% as_tibble() %>%
  mutate(total_risk = map2_dbl(risk, units_of_exposure, ~ 1 - (1 - .x)^(.y))) %>%
  ggplot(aes(x = risk, y = units_of_exposure)) +
  geom_raster(aes(fill = total_risk), alpha = 0.90) +
  scale_fill_gradient2(name = "Total chance", 
                       low = "white", mid = "white", high = "purple4", midpoint = 0.5,
                       labels = scales::percent, limits = c(0, 1)) +
  scale_y_continuous(breaks = seq(1, 24*7, 24) - 1) +
  theme(panel.grid.major = element_line(color = "black")) +
  ylab("Units of exposure") + xlab("Risk per unit") +
  scale_x_continuous(labels = scales::percent)

The visualization shows that the risk in a given unit of time has a strong influence on how many units of exposure can be experienced before the event occur. It also shows how reduction of risk relates to the number of units of exposure we can take without experiencing the event.