Rural health disparities in melanoma staging and prognostic outcomes in Iowa

Abstract

To the Editor: Melanoma is currently the fifth most common cancer in men and the sixth most common cancer in women, with 91,000 estimated new cases of melanoma and 9300 deaths in the United States in 2018. Presentation at a higher stage and higher mortality rates for multiple cancers1Hashibe M. Kirchhoff A.C. Kepka D. et al.Disparities in cancer survival and incidence by metropolitan versus rural residence in Utah.Cancer Med. 2018; 7: 1490-1497Crossref PubMed Scopus (26) Google Scholar have been found to be comparatively higher in numerous studies of rural populations. Geographic location strongly influences health care education, access to care, and outcomes.2Fennell K.M. Martin K. Wilson C.J. Trenerry C. Sharplin G. Dollman J. Barriers to seeking help for skin cancer detection in rural Australia.J Clin Med. 2017; 6: 19Crossref Scopus (9) Google Scholar Influential factors may include sociodemographic barriers such as access to finances, distance from providers, home environment, and patient cultural background. Of the US population, 14% to 19% (46.2-59 million people) live in rural areas,3Blake K.D. Moss J.L. Gaysynsky A. Srinivasan S. Croyle R.T. Making the case for investment in rural cancer control: an analysis of rural cancer incidence, mortality, and funding trends.Cancer Epidemiol Biomarkers Prev. 2017; 26: 992-997Crossref PubMed Scopus (110) Google Scholar representing a significant area for targeting and improving health care quality. We sought to assess whether disparities existed in stage at diagnosis and mortality between rural and metropolitan patients with melanoma and took advantage of our status as a state with 36% rural residents.4Iowa State Data CenterIowa quick facts.https://www.iowadatacenter.org/quickfactsDate: 2018Google Scholar We accessed the deidentified Surveillance, Epidemiology, and End Results Program of the National Cancer Institute public use data set for patients in the state of Iowa (Iowa Cancer Registry). Categories available included age, race, sex, marital status, stage at diagnosis (Breslow depth), treatment type, vital status, and site of melanoma. All patients with a first-time diagnosis of nonocular primary cutaneous melanoma between 1996 and 2015 were included. Location was categorized as rural or metropolitan by using the US Office of Management and Budget definition of metropolitan areas as urbanized areas with 50,000 or more people5Parker T. Rural-urban continuum codes. US Department of Agriculture Economic Research Service.https://www.ers.usda.gov/data-products/rural-urban-continuum-codes.aspxGoogle Scholar and all other areas as rural. Multivariate statistical analysis was performed to assess whether geographic location was associated with metastatic disease or increased mortality. We identified 12,991 diagnosed patients who met these criteria during the 20-year period, of whom 6829 (52.6%) were in metropolitan areas and 6162 (47.4%) were in rural areas. A total of 8550 (65.8%) patients were alive, and 4441 (34.2%) were deceased. The mean age in rural areas was 61 years, and in metropolitan areas it was 58 years. Living in rural areas was associated with decreased odds of being an ethnic minority (odds ratio, 0.36; P < .01; 95% CI, 0.21-0.62). Multivariate analysis of the full population indicated that patients with metastatic disease had increased odds (odds ratio, 1.13; P = .01; 95% CI, 1.03-1.24) of being located in rural areas of Iowa compared to those diagnosed with nonmetastatic disease (Table I). Those diagnosed in rural areas also had increased all-cause mortality when compared to those diagnosed in metropolitan areas (hazard ratio, 1.26; P < .01; 95% CI, 1.19-1.34) (Table II).Table IModeling the probability of a patient living in a rural area: Multivariate resultsCovariateLevelNRural/metropolitanOdds ratio95% CIP valueRaceOther4280.360.21-0.62<.01White12,563———SexFemale59260.940.88-1.01.11Male7065———StageMetastatic19671.131.03-1.24.01Nonmetastatic11,024———SiteNot specified4450.850.70-1.04.12Trunk/limbs96060.840.78-0.92<.01Head/neck2940———Marital statusDivorced/separated/widowed18461.121.01-1.24.03Single12770.730.65-0.82<.01Married7986———SurgeryYes12,2481.140.98-1.33.08No734———ChemotherapyYes150.980.35-2.69.96No12,726———RadiationYes4001.180.97-1.45.10No12,587———Hormone therapyYes150.970.35-2.68.95No12,975———Breslow depth (mm)Stage II (0.77-1.50)19851.040.93-1.15.52Stage III (1.51-2.25)7081.050.90-1.23.52Stage IV (2.26-3.0)4131.100.90-1.34.37Stage V (3.0 or more)9921.221.06-1.40<.01Stage I (0.75 or less)4527———Age12,9911.011.01-1.01<.01Bolded text indicates that statistical significance was noted.CI, Confidence interval. Open table in a new tab Table IIOverall survival on full population: Multivariate overall survivalCovariateLevelNSurvivalHazard ratio95% CIP valueRaceOther4270.080.05-0.14<.01White12,494———SexFemale58880.550.52-0.59<.01Male7033———SiteNot specified4452.882.55-3.25<.01Trunk/limbs95530.470.44-0.51<.01Head/neck2923———StageDistant/regional19623.993.74-4.26<.01Localized10,959———Rural/metropolitanRural61351.261.19-1.34<.01Metropolitan6786———Marital statusDivorced/separated/widowed18362.081.93-2.23<.01Single12700.770.68-0.86<.01Married7942———SurgeryYes12,1790.340.31-0.38<.01No733———ChemotherapyYes156.663.87-11.49<.01No12,656———RadiationYes4005.624.99-6.33<.01No12,517———HormoneYes153.341.80-6.21<.01No12,905———Breslow depth (mm)Stage II (0.77-1.50)19821.181.05-1.33<.01Stage III (1.51-2.25)7082.141.86-2.47<.01Stage IV (2.26-3.0)4123.232.76-3.77<.01Stage V (3.0 or more)9894.954.46-5.51<.01Stage I (0.75 or less)4517———Age12,9211.071.07-1.07<.01Bolded text indicates that statistical significance was noted.CI, Confidence interval. Open table in a new tab Bolded text indicates that statistical significance was noted. CI, Confidence interval. Bolded text indicates that statistical significance was noted. CI, Confidence interval. Here, we report the correlation between location of diagnosis and stage at diagnosis and all-cause mortality. Limitations include that the Surveillance, Epidemiology, and End Results data set reports only all-cause mortality. Furthermore, only location of diagnosis was reported; hence, it cannot be definitively said whether patients lived in metropolitan or rural areas. Early staging and treatment of melanoma are crucial to prevent further future morbidity and mortality and to decrease costs. Disparities seen in rural populations may indicate the need for greater focus in these areas.