To the Editor:

Your ICD-9 “cheat sheet” (November/December 1997, preceding page 95) is quite an extensive list, and I like the format. But why provide so many unspecified codes? A few acute unspecified codes are probably OK, since it is likely that the patient would be sent to a specialist to obtain a more definitive diagnosis. Still, many primary care physicians manage chronic conditions, and one would hope they know what they are managing.

Yes, the cheat sheet notes that more specific codes exist, but most providers will not take the time to look up something else and will stick to the listed choices.

Authors' response:

True, there are numerous “unspecified” codes on our list, as discussed in our original article (“An Easy Reference to ICD-9 Codes,” October 1996). We designed this list to enable family physicians who don't have computerized patient records to pick ICD-9 codes themselves “on the fly” as they see patients. The benefits of doing this are quicker charge entry and claim filing; more accurate matching of CPT and ICD-9 codes, leading to better reimbursement; and lower overhead, since a coder does not have to be employed. More complete ICD-9 diagnosis coding, done by looking up more specific codes in the thick ICD-9 book, is too time-consuming to be done practically in real time while a family physician is seeing patients every 15 minutes. The summary list, however, does allow for real-time coding by physicians. A list that covers almost all family practice visits must by necessity have a lot of “unspecified” codes.

A potential risk of using this summary list is that an “unspecified” code could be rejected by an insurance company computer as insufficient to justify higher-level E/M codes (i.e., the 99215 office visit). We have not encountered that so far. Another risk is that using “unspecified” codes could undervalue the severity of illness of patients in an era when managed care plans are beginning to severity-adjust utilization rates. We believe, however, that including all diagnosis codes (i.e., comorbidities) on a claim is more important in severity adjustment than coding only one disease specifically and ignoring the others.