Procedural coding of spinal surgeries (CPT-4 versus ICD-9-CM) and decisions regarding standards: a multicenter study.

STUDY DESIGN: A comparison of procedural coding systems ( [ICD-9-CM] [CPT-4]) applied to lumbar spine surgery patients from six teaching institutions.

OBJECTIVE: To compare the detail reflected by coding systems used to describe spinal procedures.

SUMMARY OF BACKGROUND DATA: Administrative databases contain ICD-9-CM procedural codes, which are derived from hospital discharge abstracts. These databases are used, in part, to establish health care utilization patterns and set health care policy. Previous studies have demonstrated inaccuracies in ICD diagnosis coding. However, the literature is void of information regarding the accuracy of ICD procedural coding of spine procedures.

METHODS: Data were complete in 143 of 150 lumbar spine surgery patients (aged 17-84 years). Surgeons assigned CPT-4 procedural codes. These codes were compared with ICD procedure codes assigned by hospital medical records staff.

RESULTS: On average, in four of six hospitals, there were more CPT codes assigned to patient records by the surgeon than ICD codes assigned by hospital medical records staff. Overall, CPT codes reflected a greater level of detail than ICD codes.

CONCLUSIONS: These findings illustrate the increased detail of CPT coding over ICD coding in the spinal surgery cases reviewed. The ICD procedural codes contained in administrative databases tend to underrepresent the complexity of the surgical procedures actually performed.