Tan et al. Page 1 MORTALITY RATES DUE TO GYNECOLOGIC CANCERS IN NEW YORK STATE 1 BY DEMOGRAPHIC FACTORS AND PROXIMITY TO A GYNECOLOGIC

32 Objective: To describe trends in mortality rates, in New York State, due to cervical, endometrial 33 and ovarian cancer and to assess how these rates varied with proximity to a comprehensive 34 cancer treatment center or population density (rural/urban). 35 Methods: Data were obtained from the Centers for Disease Control and Prevention (CDC)’s 36 Compressed Mortality Files, Census Bureau records, and online maps. Poisson regression 37 models were fitted to estimate death rates (mean number of deaths per 100,000 women per year) 38 due to gynecologic cancer type. Trends in death rates were compared with respect to driving time 39 to the nearest comprehensive cancer treatment center and population density, controlling for race, 40 county income level, and age at death. 41 Results: Cervical and endometrial but not ovarian death rates declined over time. For both 42 cervical and endometrial cancers, death rates varied significantly with driving time and between 43 rural and urban counties. In the case of cervical cancer, the decline over time was steeper in rural 44 than in urban counties. For endometrial cancer, the decline steepened with increasing distance 45 from a treatment center. 46 Conclusion: Improvements in cervical and endometrial cancer mortality from 1979-2001 47 followed increases in gynecologic cancer treatment research efforts, number of specialists trained 48 to treat such cases, and in the emphasis on gynecologic cancer in the training of physicians in 49 general. Our results are consistent with an interpretation that the progressive actions by leaders in 50 the gynecologic oncology profession during the late 1960’s and early 1970’s contributed to 51 improvements in mortality rates in subsequent decades. 52