Nevertheless, it induces a spurious correlation between Smoking and Mortality and biases our estimate of the actual (direct) causal effect, Smoking Mortality. In fact, it biases the estimate to such a large extent that smoking actually appears beneficial.
