Fam Pract Manag. 2004;11(1):17
To the Editor:
As Mark Twain said, “There are lies, damned lies and statistics.” I think “How to Get All the 99214s You Deserve” [October 2003, page 31] “overcoded” the statistics a little when it referenced a 33-percent undercoding rate.1 In the study by King et al, expert coders reviewed family physicians’ choices of codes for hypothetical established patients.
Compared with the expert coders, physicians did undercode 33 percent of the time for established patients, but they undercoded only 1 percent of the time for new patients. If averaged over a representative patient population, that would yield less than a 33-percent rate.
The King study also references three other studies that weren’t mentioned by the author. In the first, trained nurses observing actual patient visits noted that physicians assigned incorrect codes 55 percent of the time, split equally between undercoding and overcoding.2 Two retrospective chart reviews found similar results.3,4 These studies suggest that undercoding is not as common as in the hypothetical situation the author referenced and that physicians under-code and overcode equally.