The aims of this study were to assess the cost/benefit ratio for interinstitution pathology consultation (IPC) and to identify the types of specimens with little or no risk of diagnostic error in order to reduce the cost.All gynecologic oncology referrals having IPC from 1993 to 1998 were reviewed. Each case was evaluated by comparing both the original and the consulted pathology reports. A discrepancy was major if it led to treatment alteration. A minor discrepancy was defined as differences without clinical consequences. Consultation error was determined by comparison with the final diagnosis and clinical data obtained from the records. The cost per review was adjusted to 1998 dollars for all cases over the 5-year study period. Statistical data were obtained by Fisher's exact test and Pearson's correlation test.Five hundred sixty-nine pathology specimens from 498 patients were analyzed in this study. The major discrepancy rate was 6.5% and the minor discrepancy rate was 12.5%. Cytological specimens accounted for no major discrepancy and 13 minor discrepancies compared to 37 major and 58 minor discrepancies in histological specimens. The difference was statistically significant (P = 0.003). Consultation errors occurred in 5 cases with no alteration of clinical care. By excluding cervical and vaginal smears and cervical biopsy specimens in cases with clinically gross tumors, the cost can be reduced by 25% with no detriment to the clinical management.The types of specimens that do not need consultative pathology review include (1) cervical biopsy specimens in those patients with gross tumors and (2) cervical and vaginal smears.