The number of times that a person contracts a cold in a given year is a Poisson random variable with parameter λ=6. However a new drug has just been marketed that reduces the Poisson parameter to λ=3 for 35 % the population. For the other 65 % of the population, the drug has no effect on colds. If an individual tries the drug for a year and has 3 colds in that time, how likely is it that the drug is beneficial for him?