The purpose of this study was to describe the nature of the impact of race on the prognosis of women with ovarian carcinoma. Data for this study were selected from case reports voluntarily submitted to the National Cancer Data Base, which includes cases diagnosed from 1985 to 1988 and from 1990 to 1993 by 1,388 cancer facilities. This study used cases of women diagnosed with primary invasive epithelial ovarian tumors who had no prior diagnosis of malignant cancer. African-American women constituted 7.4% of ovarian carcinoma cases submitted. Two groups of white patients were used for comparison –those submitted by facilities whose African-American patients represented at least 7.4% of their total cases ("whites-same facility") and those submitted by facilities whose African-American patients represented less than 7.4% of their total cases ("whites-other facility").

One-third of African-Americans and "whites-same facility" lived in the South Atlantic region. African-American patients lived in lower income neighborhoods, on average, than white patents. One-third of African-Americans, 11% of the "whites-same facility", and 8% of the "whites-other facility" lived in areas with mean family incomes <$20,000 (in the patient's residential ZIP code area). The largest percentage of both African-American and "whites-same facility" patients received care from facilities with approval as Teaching Hospital Cancer Programs (38% and 34%, respectively), while the largest proportion of reports for "white-other facility" came from Community Hospital Comprehensive Cancer Programs.

There were no substantial differences in age group or histologic subtype among the three groups. About 50% of all patients were pathologically staged in 1985-8 and 75% in 1990-3. Among staged cases, African-American patients were more often diagnosed at Stage IV and less often at Stage 1-III than either group of white patients, even after controlling for grade, age and residential income. (The adjusted odds ratio (OR) for stage I versus higher stages was 1.61 for "whites-other facility" and 1.8 for "whites-same facility,” while the OR was 1.00 for African-Americans.) The African-American patients had proportionately more well-differentiated tumors at earlier stages than the white patients.

African-American patients were less frequently treated with combinations of surgery and chemotherapy than white patients. They were more frequently treated with surgery alone for stage I-II disease and chemotherapy alone for stage III-IV disease. Nearly 4% of African-American patients but only 1% of white patients had no record recommending surgical treatment be performed. Income and age were associated with not receiving the recommended surgical treatment. Nearly 6% of African-Americans, 5% of "whites-same facility" and 4% of "whites-other facility" did not receive recommended chemotherapy. Neither income nor age were associated with not receiving recommended chemotherapy. The adjusted OR for not receiving recommended treatment was 0.44 for "whites-other facility" and 0.48 for "whites-same facility," white the OR for African Americans was 1.00. Race was not associated with the type of surgery provided. African-Americans were less likely than whites to be treated with adjuvant therapy.

Five-year relative survival rates for African-Americans diagnosed from 1985 to 1988 was 30%, compared with 41% for "whites-same facility" and 40% for "whites-other facility." Survival analyses, adjusting for age, income, diagnostic status, and pathologic grade confirmed that African-American patients had poorer survival than white patients.

The authors conclude that African-American women experienced a poorer prognosis, were diagnosed at more advanced stages, were less likely to be treated with combinations of surgery and chemotherapy, and were less likely to receive the recommended treatment. This study could not evaluate potential reasons for these patterns. The authors made the following recommendations: "(1) women who present with suggestive symptoms of ovarian carcinoma should be referred to specialty care; (2) women who refuse treatment should be sent for consultation with cancer survivors of similar cultural backgrounds; (3) quality assurance boards should document dispensed services by race, gender and SES; (4) cultural and racial competency should be part of physician training; and (5) women should empower themselves by education about ovarian cancer and its treatment."