The CBS Evening News (8/2, lead story) reported that healthcare officials have announced that "the AIDS epidemic is worse than we thought." Apparently, for many years, the "government has underreported the number of people infected by the AIDS virus."
The problem with statistics is that they are only as good as the input of data. With flawed data of any kind, estimates are guesstimates and paradigms are useless. What is the primary care physician to do? We are at the mercy of the data we are provided and then ultimately held accountable for any faux pas that occurs in our individual practices related to that data.
Clearly underestimating the number of HIV cases in a practice is not jugular to practice success, but consider the effect of reporting a "normal" lab range (supposedly within the 95% confidence interval of a normal population) with flawed input data of those normal observations? At worst diseases are missed or overlooked. So the application of statistics to medical practice, and understanding the data as is presented in the medical literature is a key skill for clinicians, But can we really grasp the reality of flawed data?
Note to public health teacher...the data is everything, the concepts are important...the calculations are useless. I suspect too that HIV is a more important clinical public health problem than the formula for finding the Pearson Correlation Coefficient.
Let us not look back in anger or forward in fear, but around in awareness. 
— 
