Provider attitudes and practices regarding management of obesity in gynecologic cancer survivors

S114

Abstracts / Gynecologic Oncology 125 (2012) S3–S167

This reflects, in part, efforts made in publicly funded cervical cancer screening programs. Consideration should be given to the promotion of endometrial cancer symptom-based screening as a public health effort, given the evolving obesity epidemic and its disproportionate impact in poor populations.

radical abdominal hysterectomy by high-volume surgeons, which may contribute to a disparity in cervical cancer outcomes. The impact of these findings on short-term and long-term cancer survival outcomes should be further explored. doi:10.1016/j.ygyno.2011.12.273

Stage distribution of gynecologic cancers: urban county hospital vs national cancer database Cancer site/stage (%) Cervixp b 0.05

Uterusp b 0.05

Ovaryp — NS

I II III IV I II III IV I II III IV

JHSH 2008–2009

NCDB 2008

44.6 33.7 19.8 1.9 47.9 17.0 18.1 14.9 17.2 10.4 51.7 20.7

41.1 22.6 22.4 13.9 67.4 9.6 15.6 7.4 16.5 10.1 44.9 28.5

doi:10.1016/j.ygyno.2011.12.272

272 Patient factors associated with undergoing radical abdominal hysterectomy for cervical cancer by high-volume cancer surgeons A. Tergas, T. Diaz-Montes, E. Oni, O. Ibeanu. Johns Hopkins Medical Institutions, Baltimore, MD. Objective: Cancer patients who undergo technically complex operations by high-volume cancer surgeons generally have better postoperative outcomes. The study aim was to determine patient factors associated with undergoing radical abdominal hysterectomy by a high-volume gynecologic cancer surgeon. Methods: This was a retrospective study of patients who underwent radical abdominal hysterectomy for cervical cancer in Maryland state hospitals from 1994 to 2010, using the Maryland Health Services Cost Review Commission database. The primary variable was treatment by low-volume surgeon (LVS) or high-volume surgeon (HVS), defined by ≤ or N25 total radical abdominal hysterectomy cases (cases) during the study period. Other variables analyzed included age, race, length of stay (LOS), All Patient Refined Diagnosis-Related Group complexity and mortality risk score, hospital volume (low volume defined by ≤40 total cases during study period), payer status, and total charges. Chi-square and logistic regression were performed. Results: A total of 1170 women were included; 405 (34.8%) were operated on by a LVS and 759 (64.9%) were operated on by a HVS. Mean age was 47.6 years. By race, 773 (66.1%) were white, 261 (22.3%) were black, and 130 (11.1%) were classified as ‘other’. By payer status, 244 (20.9%) had Medicare or Medicaid, 548 (46.8%) were with a Health Maintenance Organization, 273 (23.3%) had commercial insurance, 15 (1.3%) were uninsured, and 84 (7.2%) were of another payer status. More HVS performed surgery at high-volume hospitals than at low-volume hospitals (68.9% vs. 56.5%, respectively; P b 0.001). More white women and fewer black women were operated on by a HVS compared to a LVS (white: 69.7% vs. 60.2%; black: 19.8% vs. 27.4%, respectively; P = 0.007). More women with complexity scores of 3 or 4 were operated on by a LVS (19.5% vs. 12.8%; P b 0.001). On multivariate analysis, there was no difference in likelihood of being operated on by a HVS by age, race, complexity score, or mortality risk score. Notably, LVS was 3.6 times more likely to operate on uninsured patients (P = 0.04), and 1.6 times more likely to operate at a low-volume hospital (P = 0.001). Conclusions: Patients with cervical cancer who are uninsured and those treated at low-volume hospitals are less likely to undergo

273 Racial and ethnic differences in treatment and survival among women with invasive squamous cell carcinoma of the vulva in the United States A. Tergas1, R. Bristow2. 1Johns Hopkins Medical Institutions, Baltimore, MD, 2University of California, Irvine Medical Center, Orange, CA. Objective: The objective of this study was to examine racial/ethnic differences in treatment and survival in women diagnosed with invasive vulvar cancer in the United States. Methods: Women with a diagnosis of invasive vulvar cancer were identified from the Surveillance, Epidemiology, and End Results database from 1992 to 2002. Exclusion criteria were as follows: cancers of unknown or non-squamous cell histology, women diagnosed at death or by autopsy, and unknown race. Statistical analysis using Chi-square, Fisher's Exact Test, Kaplan–Meier survival methods, and Cox regression proportional hazards models were performed using STATA statistical software. Results: Of the 2357 cases of invasive vulvar cancer included in this study, 1974 (83.8%) were non-Hispanic white, 209 (8.9%) were nonHispanic black, 119 (5.0%) were Hispanic, and 55 (2.3%) women of another race/ethnicity. Median age was 70 years. Black women were older than women of other races (P =b 0.001). There was no significant difference in stage, tumor size, or tumor grade by race/ ethnicity. After adjustment for stage, black women were half as likely (OR = 0.48, 95% CI 0.31–0. 74) to undergo surgery and 1.7 times more likely (OR = 1.67, 95% CI 1.18–2.36) to receive radiation than white women, whereas treatment patterns for Hispanic women and women of other races were similar to the treatment patterns for white women. In multivariate analysis, surgical treatment reduced the risk of death from vulvar cancer by 46% (HR 0.54, 95% CI 0.43–0.67), whereas radiation was not shown to impact the risk of death (HR 0.99, 95% CI 0.84–1.19), after adjusting for age, race, stage, and grade. There was no significant difference in risk of death by race/ethnicity group after adjusting for the previously described variables. Conclusions: According to this analysis of SEER data, race/ethnicity is not an independent risk factor for poor prognosis in women diagnosed with invasive vulvar cancer, despite differences in treatment modality by race/ethnicity. Further research to define the factors contributing to differences in treatment selection according to race/ethnicity and the resulting impact on quality of life is warranted. doi:10.1016/j.ygyno.2011.12.274

274 Provider attitudes and practices regarding management of obesity in gynecologic cancer survivors A. Jernigan1, A. Tergas1, A. Fader2. 1Johns Hopkins Medical Institutions, Baltimore, MD, 2Greater Baltimore Medical Center, Baltimore, MD. Objective: Obesity is associated with the development of and risk of death from endometrial, breast and several other female cancers. However, studies show that obese cancer survivors are not making healthy lifestyle changes. The study objective was to survey oncology

Abstracts / Gynecologic Oncology 125 (2012) S3–S167

providers regarding their attitudes and practices related to obesity counseling and management in cancer survivors (CS). Methods: A survey was sent to Society of Gynecologic Oncology members by electronic mail. Chi-square and Fisher's exact tests were used to analyze responses among provider groups. Results: Of the 245 (25%) respondents, 94% were practicing GYN oncologists (GON) or fellows, 5% allied health professionals and 1% other oncologists. Median age was 42 yrs and 51% were female. Responses were equally distributed amongst those in practice ≤3 yrs, 4–15 yrs and ≥16 yrs. 43% of GON reported measurements that classified themselves as overweight/obese. Providers reported that ≥50% of their CS were overweight/obese; 82% felt that discussing a patient's weight would not harm the doctor–patient relationship. The majority (95%) agreed that addressing lifestyle modifications with obese CS was important. Respondents felt that GONs (85%) and PCPs (84%) were responsible for addressing obesity. More providers ≤42 years old reported undergoing training in obesity management (p b .001) and were also more likely to believe that CS would benefit from obesity counseling than providers N42 years (72% vs 54%, p = .016). Further, providers of normal weight were more likely than overweight/obese providers to believe in the value of education on healthy eating and physical activity (P = .035) and associations of obesity with cancer (P = .043). Providers felt that 10 min was adequate to devote to counseling, but in practice, 91% spent ≤10 min addressing obesity issues. After initial counseling, most GON (82%) referred CS to other providers for obesity interventions. Conclusions: Oncology providers believe that addressing obesity with cancer survivors is important. Those ≤42 years and normal weight providers were most likely to perform obesity counseling and believe in its value. Providers felt responsible for initial counseling, but believed obesity interventions should be directed by other specialists. Further research is needed to identify barriers to care for obese cancer survivors and to improve physician engagement with obesity counseling in the “teachable moment” provided by a new cancer diagnosis.

S115

as non-Hispanic White. There was no difference in median age, stage at diagnosis, or tumor grade between Black and NHW women. Rates of primary surgical resection to disease residual less than 1 cm were 45.4% for Black women compared with 56.5% for NHW, 51.1% for Asian, and 62.5% for Hispanic/Latina women. Black, Asian, and Hispanic/Latina women had lower median household income than NHW women, but there was no racial or ethnic difference in insurance type, and less than 5% of all patients were uninsured. Combining all stages of disease, there was no difference in outcome between Black and NHW patients; median OS for Black women was 53.6 months vs 53.5 months for NHW women. Median OS for was longer for Hispanic/Latina women (122.6 months) than for Asian (74.8 months), Black, or NHW women with EOC (p b .001). Conclusions: Disparities in outcome between Black and White women with EOC described in population-based studies are eliminated when care is provided by gynecologic oncologists in a multidisciplinary setting. Women of other racial/ethnic minority groups experienced superior outcomes compared to Black or White women receiving similar care. Further analysis is needed to determine what factors account for the longer OS for Hispanic/Latina and Asian women with EOC observed in our study. doi:10.1016/j.ygyno.2011.12.276

276 Abstract Withdrawn at Author Request.

doi:10.1016/j.ygyno.2011.12.277

Quality of Life

275 Racial and ethnic disparities in ovarian cancer at a multidisciplinary cancer center C. Brown1, A. Adekoya2, M. Steele1, S. Edwards-Bennett3. 1Memorial Sloan-Kettering Cancer Center, New York, NY, 2Mount Sinai School of Medicine, New York, NY, 3H. Lee Moffitt Cancer Center, Tampa, FL.

277 Factors associated with grade 3–4 treatment-related toxicity in women with advanced or recurrent cervical cancer: an exploratory analysis of Gynecologic Oncology Group trials (GOG) 179 and 204 D. Chase1, J. Kauderer2, L. Wenzel3, H. Long4, D. Cella5, B. Monk6. 1St. Joseph's Hospital and Medical Center, Phoenix, CA, 2Gynecologic Oncology Group, Buffalo, NY, 3University of California Irvine — Medical Center, Orange, CA, 4Mayo Clinic, Rochester, MN, 5Northwestern University/Prentice Women's Hospital, Chicago, IL, 6Creighton University School of Medicine, Omaha, NE.

Objective: To determine whether racial disparities in outcome between Black and White women with epithelial ovarian cancer in the United States reported in population-based studies persist when care is delivered at an NCI-designated comprehensive cancer center by a multidisciplinary team of gynecologic cancer specialists. Methods: We performed a retrospective review of women with epithelial ovarian cancer (EOC) who underwent primary surgical treatment at our institution from 1996 to 2008. Medical records were reviewed for demographics, tumor characteristics, socioeconomic status, primary surgical outcome, and overall survival (OS). Results for non-Hispanic Black, Hispanic/Latina, Asian and non-Hispanic White (NHW) women were included in this analysis. Chi-square tests were used to compare tumor characteristics and demographic variables. Overall survival (OS) was calculated using Kaplan–Meier (KM) estimates and compared between racial/ethnic groups with the log-rank test. Results: Of 992 women with ovarian cancer who had primary surgery at our institution, 22 (2.2%) were identified in their medical record as Black, 40 (4%) as Hispanic/Latina, 45 (4.5%) as Asian, and 755 (76%)

Objective: The ability to predict treatment-related toxicity could be useful at the initiation of treatment in advanced/recurrent cervical cancer patients. The objective of this study was to describe pretreatment patient characteristics and quality of life (QOL) scores as they relate to the development of moderate to severe toxicity in patients receiving chemotherapy for advanced/recurrent cervical cancer. Methods: The study sample was drawn from GOG protocols 179 and 204. Grade 3 or 4 toxicities were considered under four specified categories: peripheral neuropathy, fatigue, hematologic and gastrointestinal. The data variables explored included age, stage, pretreatment with radiation, performance status (PS), and the baseline FACTCX (Functional Assessment of Cancer Therapy—Cervix) QOL score. A logistic regression model was developed with the various adverse events as the binary [0/1] outcomes. Results: Six hundred seventy-three evaluable patients were used for the following analyses. Thirty-six patients did not complete the baseline QOL forms and six patients did not have toxicity information reported and were therefore excluded. Out of the 673 patients under considera-

doi:10.1016/j.ygyno.2011.12.275