This study retrospectively analyzes the treatment of advanced ovarian cancer (Stages III and IV) in elderly patients (> or = 65) compared to that in younger patients (< 65). The purpose of this study was to identify possible treatment bias toward the elderly and to statistically analyze the nature of these differences. Seventy patients were evaluated of which 29 were identified as elderly and 41 as young. All patients were treated with platinum-based chemotherapy. Chi 2, log rank, Kaplan-Meier, and Cox model analyses were performed for multiple variables including age, grade of tumor, adequacy of surgery, and dose intensity. The elderly significantly differed from the young in the following analyses: median length of hospitalization, 20 days vs 11 days (P < 0.001); optimum surgery, 79.3% vs 97.5% (P = 0.02); initial chemotherapeutic dose reduction, 15.4% vs 0% (P = 0.02); median survival compared to age, 19.2 months vs 36.7 months (P < 0.03). When survival analysis was performed comparing 17 elderly patients and 40 younger patients who had optimum surgery and optimum initial chemotherapy, the median survival remained essentially unchanged, 22.0 months vs 36.7 months. There were differences in treatment intensity between young and old, however, the indications generally were valid and when analyzed by the Kaplan-Meier and Cox model, these differences became insignificant. It was concluded that when elderly patients can undergo aggressive surgical and chemotherapeutic management, survival remained significantly decreased for aged compared to younger patients. Physician bias was not a major factor accounting for the poorer survival observed in elderly patients. Age was the most significant variable related to survival and could not be accounted for by differences in adequacy of surgery or dose intensity.