We're interested in modeling the relationship between number of cigarettes smoked per day and number of hospital visits over a person's lifetime.  Here, the predictor is number of cigarettes smoked per day and the outcome is number of hospital visits.  We run a regression and find that the regression coefficient is .03. This regression coefficient implies that, on average, increasing the number of cigarettes smoked per day by [选择] 1 .03 results in [选择] 1 .03 [选择] more fewer hospital visits, on average, over a person's lifetime.