Considering that a researcher can test at the <p = .05 (95%), p = .01 (99%), or p = .001 (99.9%) level for statistical significance, how do you use these three levels in relationship to evaluate risk to patients in implementing the decisions based on research findings?

The lower p is, the less likely that the the results are due solely to chance. Now you need to decide how to use the information.