Geographic Disparities in Cancer Survival and Access to Care: Ovarian Cancer in kentucky

International Journal of Applied Geospatial Research, 1(1), 67-79, January-March 2010 67

Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Globalis prohibited.

Keywords: Access to Care, Geographic Information System (GIS), Gynecologic Oncologist, Kaplan-Meier Survival, Kentucky, Ovarian Cancer

introdUCtion

In the United States more women die from ovarian cancer than any other gynecologic malignancy, and ovarian cancer ranks fifth as an overall cause of cancer death in women. In the year 2008 it is estimated that 21,650 new cases of ovarian cancer were diagnosed, and 15,520 women died of the disease

(Jemal, Siegel, Ward, Hao, Xu, Murray & Thun, 2008). Approximately 70% of the cases are advanced (Stage III or IV) when they are diagnosed, and no avail-able screening test, including CA-125 or ultrasound, has ever proven effective in reducing mortality. Optimizing treat-ment is thus the most effective way to improve survival.

Optimal treatment for ovarian can-cer involves a combination of surgery and chemotherapy. Surgery is almost

Geographic disparities in Cancer survival and access to Care:

ovarian Cancer in kentuckyMary E. Gordinier, Norton Cancer Institute, USA

Carol L. Hanchette, University of Louisville,USA

aBstraCtFrom 1995-2005, ovarian cancer accounted for 2.7% of new cancer cases diagnosed among women in Kentucky and was responsible for 4.7% of female cancer deaths in the state. The five-year survival rate for ovarian cancer is 45% for all stages combined. Multiple studies document a survival advantage for women with gynecologic malignancies when treated by a gynecologic oncologist. The authors used Kentucky Cancer Registry data for the years 1995-2005, geocoded to 5-digit ZIP code, to examine the hypothesis that ovarian cancer survival is higher among patients receiving treatment in areas where gynecologic oncologists practice. Their hypothesis was confirmed. A secondary goal of the study was to identify geographic areas of the state with lower overall access to care. Contrary to the expected pattern of low access to care in the Appalachian region of the state, their analysis indicated that access to successful treatment is a greater issue in the western portion of Kentucky.

DOI: 10.4018/jagr.2010071605

IGI PUBLISHING

This paper appears in the publication, International Journal of Applied Geospatial Research, Volume 1, Issue 1edited by Donald Patrick Albert © 2010, IGI Global

701 E. Chocolate Avenue, Hershey PA 17033-1240, USATel: 717/533-8845; Fax 717/533-8661; URL-http://www.igi-global.com

ITJ 5509

68 International Journal of Applied Geospatial Research, 1(1), 67-79, January-March 2010


always the first step, providing both a firm diagnosis as well as removing the tumor that causes symptoms. The quality of the initial surgery is a critical determinant of survival, because the amount of tumor remaining is predictive of outcome. A woman with no visible tumor at the completion of surgery will have a 5-year survival rate of approxi-mately 40%, in contrast to a woman with tumor implants 2 cm in size or more remaining, who will have a 5-year survival of 15% or less. (Ozols, Rubin, Thomas & Robboy, 2004).

Gynecologic Oncologists (Gyn Onc) are specialists specifically trained in both the surgical management and the chemotherapeutic treatment of gynecologic cancers. Involvement of a Gyn Onc in the care of patients with gynecologic malignancies con-fers a significant survival advantage to those patients. Survival of women with advanced ovarian cancer was 26 months when treated by a Gyn Onc, as opposed to 15 months when such a specialist was not involved, according to a study of the statewide, population-based Utah Cancer Registry (Carney, Lancaster, Ford, Tsodikov & Wiggins, 2002). Improved survival outcomes

under Gyn Onc treatment have been noted for other gynecologic cancers as well (MacDonald, Sause, Lee, Dodson, Zempolich & Gaffney, 2005).

Table 1 shows case distribution and 5-year survival rates of ovarian cancer for the U.S., by stage of diagnosis (American Cancer Society, 2008). In Kentucky, ovarian cancer accounted for 2.7% of new cancer cases diagnosed among women from 1995-2005 and was responsible for 4.7% of female cancer deaths in the state (Kentucky Cancer Registry, 2009). These percentages are overshadowed by the burden of lung cancer in Kentucky, which has the high-est rates in the nation. The Kentucky incidence rate of 12.5 per 100,000 is lower than the U.S. rate (13.1), but the state has a higher mortality rate: 9.4 vs. 9.0 (Jemal, Murray, Ward, Samuels, Tiwari & Ghafoor, 2005).

A preliminary study of Kentucky gynecologic cancers from 1995-2003 showed no relationships between stage of diagnosis and age, poverty, or per-cent of population rural (Hanchette & Gordinier, 2007). Using more detailed treatment data, this study examines the following hypothesis: ovarian can-cer survival is higher among patients

Table 1. Case distribution and 5-year survival rates by stage of diagnosis (Ameri-can Cancer Society, 2008)

Stage Distribution (%) 5-Year Survival Rate

Local 19 92.4

Regional 7 71.4

Distant 68 29.8



receiving treatment in areas where gynecologic oncologists practice. In the state of Kentucky, Gyn Onc special-ists practice almost exclusively in two counties: Jefferson and Fayette, where the cities of Louisville and Lexington are located. Six Gyn Oncs practice in Louisville, four in Lexington (Women’s Cancer Network, 2009). These cities house the state’s two largest universi-ties, the University of Louisville and the University of Kentucky, respectively. A small number of patients cross the state border to obtain care in Cincinnati (OH), Nashville (TN) and Huntington (WV). Figure 1 is a map of Kentucky counties with Gyn Onc locations and cross-border cities.

The concept of ‘access’ refers to the ability of patients to use the health

services they need, when and where they need them (Joseph & Phillips, 1984; McLafferty, 2003). Locational accessibility is a geographic concept that is often determined by the patient’s proximity to healthcare resources. Other measures of accessibility exist. ‘Effec-tive’ accessibility addresses such issues as financial and social access (Cromley & McLafferty, 2002). One expectation of this study is that we would find lower Gyn Onc access rates in the Appalachian region of the state. Clusters of high rates of cervical cancer, a gynecologic cancer whose rates often reflect lack of access to care, have been well documented for Appalachia (Centers for Disease Con-trol, 2002; Lengerich, Tucker, Powell, Colsher, Lehman, Ward, Siedlecki & Wyatt, 2005; Yabroff, Lawrence, King,

Figure 1. Kentucky counties with Louisville, Lexington and cities in adjacent states



Mangan, Washington, Yi, Kerner & Mandelblatt, 2005). A recent study of endometrial cancer in Kentucky re-ported that rural women often lacked commercial insurance, underwent less comprehensive surgical evaluation, had less multimodality treatment and were more likely to be treated in small hospi-tals (Modesitt, Huang, Shelton & Wyatt, 2006). However, the authors found few differences in outcomes in Appalachian vs. non-Appalachian counties.

CUriosity-driven and Client-driven researCh

Weller (2010, p. 2) has extolled the ben-efits of combining curiosity-driven and client-driven research, the latter often referred to as ‘applied.’ The journal, Applied Geography, defines the term as ‘research which uses geographical theory and methodology to resolve those human problems that have a geographi-cal dimension’ (Elsevier, 2009.) The research reported in this article is ap-plied. It analyzes a perceived problem, i.e. lack of access to the best ovarian cancer treatment, confirms that dispari-ties exist, and discusses approaches to ameliorating those disparities. The work began as curiosity-driven research, with the identification of specific research questions, many of which are common themes in the spatial analysis of cancer occurrence: are there geographic dispar-ities in cancer incidence and mortality in Kentucky? What are the relationships among geographic cancer clusters, stage of diagnosis, treatment mode,

survival rates, place characteristics and distance to Gyn Onc treatment? What emerged from the initial analysis was a strong interest in the project by Ovar-ian Awareness of Kentucky (OAK), an advocacy and education support group for ovarian cancer patients and fami-lies, and the realization that additional data provider processing by Kentucky Cancer Registry staff was needed. OAK provided a generous grant to underwrite the current research and fund the addi-tional data processing efforts (Doimer, 2007). Weller (2010) stated that “When we combine applied geography with client-driven research, however, things become relatively more focused in a hurry.” Our initial research project included cervical and uterine cancers, but ultimately focused on ovarian, the most serious of the three.

data and Methods

After approval from the University of Louisville Institutional Review Board (IRB) and the Kentucky Cancer Reg-istry (KCR), we obtained de-identified level II data for all Kentucky residents diagnosed with ovarian cancer from 1995 to 2005. Data abstracted included demographic data, stage, therapy type, patient county, (5-digit) ZIP code of residence and a code that provided information about access to a Gyn Onc specialist. This code was not included in the original KCR database; KCR analysts abstracted provider informa-tion with the funding provided by OAK. The reason for this is that specific in-



formation about treatment providers is highly sensitive as providers are wary about health outcomes being linked to their facilities. Therefore, the site of first treatment for each woman’s cancer was coded by location: (a) within a county where gynecologic oncology specialist care is available; (b) within a county where gynecologic oncology specialist care is not available; (c) out of state; or (d) unknown or no therapy. Thus, if a patient’s first treatment occurred in Jefferson or Fayette Counties, it was presumed that she had access to a Gyn Onc specialist. Data were reviewed to reconcile discrepancies between clini-cal and TNM staging. TNM staging is a globally recognized standard that describes the extent that cancer exists in a patient’s body. ‘T’ describes the size of the tumor and its level of inva-sion; ‘N’ describes the involvement of regional lymph nodes, and ‘M’ indicates whether there is distant metastasis. For this analysis, if the number of nodes examined was not documented for early stage cancers, stage was coded as NX rather than N0. This allowed us to separately analyze patients whose stage was based on proof that their cancer had not spread (by examination of lymph nodes) and patients whose stage was assumed to be early without the proof of lymph nodes. Up to 40% of patients who appear to be early stage will have a higher stage when lymph nodes are examined.

We analyzed the records of 3,105 women diagnosed with ovarian cancer during the study period. Table 2 shows

demographic, treatment and access information for all patients. We used spatial and non-spatial methods of analysis to examine geographic patterns of treatment access, stage of diagnosis and survival. ArcGIS 9.2 software (Environmental Systems Research Institute, Redlands, CA) was used for database management and query, map-ping and visual analysis. Address of residence for each patient was geocoded to 5-digit ZIP code centroids and the resulting latitude/longitude coordinates for each patient were provided by the Kentucky Cancer Registry. For analysis purposes, point mapping of patients by access code was carried out for Stage I & II and Stage III & IV patients, respec-tively. To protect patient confidentiality, ZIP code data were later aggregated by county. Point mapping was then carried out using the ArcGIS dot density func-tion. County-level choropleth mapping provided important information about geographic patterns of access. Finally, Kaplan-Meier survival curves were computed, by access code. The Ken-tucky Cancer Registry provides data on survival status and date of last contact, however, patients may live for a certain length of time after treatment and are lost to follow-up. The Kaplan-Meier method is a product limit estimator that is often used to measure the percent-age of patients that live a given length of time after treatment. Kaplan-Meier survival curves were computed by Kentucky Cancer Registry staff, after data interpretation and recoding by the authors, including the reconciliation



of discrepancies between clinical and TNM staging.

resUlts

Table 2 show the distribution of the 3,105 ovarian cancer cases by stage. Nearly 55% (n=1,702) of the cases were late stage (III & IV) diagnoses and about half that number were diagnosed during Stages I & II. Table 3 shows the distribu-tion of access codes by stage. Chi-square testing showed no significant difference between stage at diagnosis and access to treatment. However, geographic and survival analyses yielded important in-formation about geographic disparities in access to specialist treatment.

Geographic analysis

The maps in Figure 2 show the county-level distribution of all 3,105 cases by access code, using a random dot den-sity mapping function. As expected, the greatest numbers of patients with access lived in Jefferson and Fayette Counties. Patients who obtained treat-ment out-of-state are most numerous in the northern counties of Boone, Kenton and Campbell, which have easy access to providers in Cincinnati, OH. Patients in two eastern counties, Greenup and Boyd, appear to have obtained treatment in Huntington, WV. Patients from War-ren and Allen Counties (Bowling Green area) may have received treatment in Nashville, TN. The highest numbers of patients with no access appear to be located in western Kentucky, as a whole, and in the same northern and

Table 2. Demographic, treatment and access information for 3,105 cases

Characteristic N (%)

Age < 40 41-50 51-60 > 61

265 (8.5%) 435 (14.0%) 619 (19.9%) 1786 (57.6%)

Race White African American Other/Unknown

2930 (94.4%) 150 (4.8%) 25 (0.8%)

Stage of Diagnosis I & II III & IV N/A Unknown

846 (27.2%) 1702 (54.9%) 106 (3.4%) 451 (14.5%)

Access to Gyn Onc Yes No Out-of-state Unknown

1400 (45.1%) 832 (26.8%) 85 (2.7%) 788 (25.4%)



eastern counties where out-of-state ac-cess is highest. The expected pattern of lower access in the Appalachian region of eastern Kentucky is not indicated in Figure 2.

The Kentucky Cancer Program (KCP), created in 1982, is a state-funded, university-based program dedi-cated to reducing the cancer burden of Kentucky residents through education, research and community outreach. It is represented by two regions. The western counties (KCP West) are administered by the James Graham Brown Cancer Center at the University of Louisville and the Lucille Parker Markey Cancer

Center at the University of Kentucky administers the eastern counties (KCP East). Figure 3 shows the percentage of stage III & IV ovarian cancer pa-tients without access to a Gyn Onc, by county. Patients with out-of-state or unknown access were not included in the denominator. The Kentucky Cancer Program’s administrative boundaries are indicated by the heavy black line. Three geographic patterns exist. First, as expected, the counties surrounding Louisville and Lexington have high access rates, with fewer than 25% of patients lacking access. Second, as de-scribed in Figure 2, the counties with

Figure 2. County-level distribution of cases by access code. (Data source: Ken-tucky Cancer Registry)

Table 3. Distribution of diagnosis stage by access code

Stage Access No Access Out-of-State Unknown

I & II 471 281 19 75

III & IV 859 516 52 275



the highest numbers of out-of-state ac-cess in northern and eastern Kentucky have low access rates, in general, which indicates that patients are seeking care locally. What is unexpected is the third and most obvious pattern: the dispar-ity between Appalachian counties in eastern Kentucky and the counties in Kentucky’s far west. In every county in western Kentucky, including the Jackson Purchase and Western Ken-tucky Coalfield regions, the percentage of patients without access ranges from 50 to 100%. The percent of patients in Appalachia without access was much lower, mostly less than 50%.

These results are discussed in more detail later in the article as they are most likely due to the success of intervention

programs in Appalachia and the greater distances between Louisville and far western Kentucky counties.

kaplan-Meier survival analysis

Kaplan-Meier survival analysis was used to compare the outcomes of pa-tients first treated in counties where Gyn Onc specialist care was available against those first treated in counties without such care. It was carried out for two different node pathologies of stages I & II (NX and N0) and for stages III & IV.

For patients with late stage diag-noses, access to gynecologic oncology specialist care demonstrates a survival advantage as compared to those first treated in a county without specialist

Figure 3. Percent of stage III & IV ovarian cancer patients without access to a Gyn Onc, by county. Kentucky Cancer Program regional boundaries indicated by the heavy black line.



care. This is reflected in the Kaplan-Meier curve in Figure 4, where the top curve represents access and bottom curve reflects no access.

While the trend is not as clear, survival did differ according to access to gynecologic oncology care when patients were assigned a stage of I or II without evaluation of lymph nodes. Those treated in a county with specialist care had a survival equivalent to that of patients who were surgically staged. Those without access to specialist care fared worse overall. It should be noted that the number of cases in the non-surgically staged group is much lower than the other groups, so statistical

significance may be questioned. Despite low numbers, a trend to poorer survival in those without access is apparent.

disCUssion

Our hypothesis, that women with ovar-ian cancer treated in areas where gyne-cologic oncology specialists practice have better survival than those treated in areas without such specialists, was sup-ported by this study. The most striking finding, however, was the geographic distribution of access to Gyn Onc care. The eastern part of the state, which includes Appalachia, is predominantly mountainous and rural, containing the

Figure 4. Kaplan-Meier survival curve by access to Gyn Onc specialist



highest poverty rates in the state. In fact, the geographic distribution of poverty in Kentucky is nearly the inverse of the patterns shown in Figure 3. Histori-cally, access to quality health care in Appalachia has been poor. To reach the care of Gyn Oncs requires that patients travel to either Louisville or Lexington. We assumed that patients in eastern Kentucky would be more reluctant to leave their communities. They have less money for travel, and the roads in that part of the state are in poorer condi-tion than those in the western part of the state. After accounting for patients who were treated in neighboring states, the pattern remains striking: access to specialist care for ovarian cancer is worst in western Kentucky.

Appalachia has received a fair amount of attention due to its cervical cancer mortality, which in the 1960s was almost double the national rate. In response to this a number of initia-tives have been put in place to address the health care needs of Appalachia, including the Appalachian Cancer Network and the Kentucky Cancer Program. Within the Kentucky Cancer Program, the eastern and western halves of the state are administered separately. Community-based intervention projects in partnership with local churches and communities are especially active in eastern Kentucky counties. Federal and state funding supports these ef-forts. Now these historically distressed counties have seen a decline in cervical

cancer mortality, with rates more in line with levels of national improvement (Geary, n.d.). The findings of this study suggest that women with ovarian cancer have also benefited from this improved access to health care in Appalachia.

Another factor impacting access is geographic distance. The distance from Pikeville, a southeastern Kentucky out-lier, to Lexington is 140 miles, a drive of about 2.5 hours. Most other areas in Appalachia are closer to Lexington, although some remote locations are a 3-hour drive away. Many residents in western Kentucky live more than three hours from Louisville. Two of the larg-est towns – Paducah and Murray – both with good highway access, are more than 3 hours from Louisville.

The survival advantage seen in advanced ovarian cancer patients is likely due to the quality of the surgical debulking. Of the 723 patients whose surgery was coded as debulking, 495 (68.5%) had access to Gyn Onc care, as opposed to 154 (21.3%) treated in an area without access to providers with specialized training. It is well-known that ovarian cancer survival is inversely related to the size of the tumor implants that remain after the debulking surgery. The more completely the tumor is re-moved, the longer that person is likely to survive. Since chemotherapy regimens were fairly standard during the study period and those drugs are accessible to oncologists throughout the state, the quality of the first surgical effort prob-



ably accounts for the survival advantage of patients first treated in counties with gynecologic oncologists.

Although the majority of patients with ovarian cancer have advanced disease when they are diagnosed, some do have early-stage disease. For these patients, the presence of lymph nodes submitted for examination was used as a surrogate for a proper assessment for early ovarian cancer. Patients treated in an area with access to specialist care had a more thorough surgery, as assessed by number of nodes removed: in areas where specialist care was available only 4% had just one lymph node removed as part of their operation to prove whether spread had occurred, while in areas without gynecologic oncology care 23% were considered ‘node-negative’ when only one node was examined.

ConClUsion

This study examined whether ovarian cancer survival is higher among patients receiving treatment in areas where gynecologic oncologists practice and where geographic disparities in access to treatment existed. Our conclusions indicate that late-stage ovarian cancer survival is indeed higher in areas where gynecologic oncologists practice and that there are strong disparities in access to care among Kentucky regions. Our results defy current views about where these disparities exist. This research would not have been possible without the support of Ovarian Awareness of

Kentucky, an organization dedicated to the fight against ovarian and other gynecologic cancers through education and support.

The true value of this applied, client-driven research is that it points to a need for further study to better understand decision-making strategies in care-seeking behavior. From 2000-2005, one of the authors (Gordinier) participated in regional triage between small com-munities and a university hospital in southern New England. A system of guidelines was created to triage patients with gynecologic cancers and this system was successfully introduced to local community hospitals. Our long-term goal is to provide more effective cancer interventions in Kentucky areas with poor access to care and to improve treatment outcomes for those in greatest need of Gyn Onc services.

referenCes

American Cancer Society. (2008). Cancer facts and figures 2008. Atlanta, GA: Ameri-can Cancer Society.

Carney, M. E., Lancaster, J. M., Ford, C., Tsodikov, A., & Wiggins, C. L. (2002). A population-based study of patterns of care for ovarian cancer: Who is seen by a gynecologic oncologist and who is not? Gynecologic Oncology, 84(1), 36–42. doi:10.1006/gyno.2001.6460

Centers for Disease Control and Prevention. (2002). Cancer death rates – Appalachia, 1994-1998. Morbidity and Mortality Weekly Report, 51(24), 527–529.



Cromley, E. K., & McLafferty, S. L. (2002). GIS and public health. New York, NY: Guilford Press.

Doimer, J. (2007). News release: Local ovarian cancer group underwrites grant to study geographic disparities in cancer care. Retrieved April 2, 2009 from: http://www.ovarianawarenessofky.org/geograph-icStudy20070918.pdf.

Elsevier (2009). Applied Geogra-phy. Retrieved April 2, 2009 from: http://www.elsevier.com/wps/find/journaldescription.cws_home/30390/description#description

Geary, D. (n.d.). Cervical cancer mortality: exploring the disparity in Appalachia. East Tennessee State University, Office of Rural and Community Health, Rural Appalachian Cancer Demonstration Program. Program report to Centers for Disease Control and Prevention.

Hanchette, C., & Gordinier, M. (2007, Octo-ber). Geographic disparities in gynecologic cancer incidence and mortality in Kentucky. Presentation at the 30th Applied Geography Conference, Indianapolis, IN.

Jemal, A., Murray, T., Ward, E., Samuels, A., Tiwari, R. C., & Ghafoor, A. (2005). Cancer statistics, 2005. CA: a Cancer Jour-nal for Clinicians, 55, 10–30. doi:10.3322/canjclin.55.1.10

Jemal, A., Siegel, R., Ward, E., PhD, Hao,Y., Xu, J., Murray, T., & Thun, Cancer statistics, 2008. A Cancer Journal for Clinicians, 58, 71-96.

Joseph, A. E., & Phillips, D. R. (1984). Accessibility and utilization: Geographical perspectives on health care delivery. New York, NY: Harper & Row.

Kentucky Cancer Registry. (2009). Re-trieved March 29, 2009 from: http://www.kcr.uky.edu/.

Lengerich, E. J., Tucker, T. C., Pow-ell, R. K., Colsher, P., Lehman, E., & Ward, A. J. (2005). Cancer incidence in Kentucky, Pennsylvania, and West Virginia: Disparities in Appalachia. The Journal of Rural Health, 21(1), 39–47. doi:10.1111/j.1748-0361.2005.tb00060.x

MacDonald, O. K., Sause, W. T., Lee, R. J., Dodson, M. K., Zempolich, K., & Gaffney, D. K. (2005). Does oncologic specialization influence outcomes fol-lowing surgery in early stage adenocarci-noma of the endometrium? Gynecologic Oncology, 99(3), 730–735. doi:10.1016/j.ygyno.2005.07.111

McLafferty, S. L. (2003). GIS and health-care. Annual Review of Public Health, 24(1), 25–42. doi:10.1146/annurev.pub-lhealth.24.012902.141012

Modesitt, S. C., Huang, B., Shelton, B. J., & Wyatt, S. (2006). Endometrial cancer in Kentucky: The impact of age, smoking status, and rural residence. Gynecologic Oncology, 103(1), 300–306. doi:10.1016/j.ygyno.2006.03.009

Ozols, R. F., Rubin, S. C., Thomas, G. M., & Robboy, S. J. (2004). Epithelial ovarian cancer. In Hoskins, W.J., Perez, C.A., & Young, R.C. (Eds.), Principles and prac-tice of gynecologic oncology (4th ed.). New York: Lippincott, Williams &Wilkins.

Wellar, B. (2010). Significant advances in applied geography from combining client-driven and curiosity-driven research meth-odologies. International Journal of Applied Geospatial Research, 1(1), 1-25.



Women’s Cancer Network. (2009). Re-trieved May 1, 2009 from: http://www.wcn.org.

Yabroff, K. R., Lawrence, W. F., King, J. C., Mangan, P., Washington, K. S., & Yi, B. (2005). Geographic disparities in cervical cancer mortality: what are the roles of risk factor prevalence, screening, and use of recommended treatment? The Journal of Rural Health, 21(2), 149–157. doi:10.1111/j.1748-0361.2005.tb00075.x

Mary Gordinier, MD is a native of Kentucky, and received her MD from the University of Louisville. She completed her specialty training in obstetrics and gynecology at Brown University in Providence, RI, and subsequently completed a fellowship in Women’s Oncology at that institution. She completed her fellowship in Gynecologic Oncology at the M.D. Anderson Cancer Center in Houston, Texas. Since that time, she has served as faculty at Brown University and the University of Louisville. She currently is practicing at the Norton Cancer Institute in Louisville, KY, where in addition to active clinical work she is the primary investigator on numerous trials.

Carol Hanchette is an associate professor in the Department of Geography and Geosci-ences, University of Louisville, Louisville, Kentucky, USA. She also holds an Associate appointment in the University’s School of Public Health and Information Sciences. She received her MA and PhD in geography from the University of North Carolina, Chapel Hill. Hanchette’s teaching and research foci are on medical geography, globalization, spatial analysis and the use of geographic information science (GIS) for health analyses. Past research projects have included the spatial analysis of prostate cancer mortality (and preliminary testing of the Vitamin D hypothesis), geographic modeling and po-litical ecology of childhood lead poisoning in North Carolina, geographic analysis of CDC-funded HIV prevention services, and an NIH-funded study on validation of indirect pesticide exposure estimates. Her current research foci are geographic disparities in women's cancers in Kentucky, and asthma, air quality and environmental justice in Louisville, Kentucky.

Geographic Disparities in Cancer Survival and Access to Care: Ovarian Cancer in kentucky

Documents

Transcript of Geographic Disparities in Cancer Survival and Access to Care: Ovarian Cancer in kentucky