Analysis_of_Data_Report

Individual Analysis of Research Report Danna Wood HCS/438 Statistical Applications October 31, 2011 Reagan Parks Individual Analysis of Research Report In 2011 approximately 280,000 women will be diagnosed with breast cancer and nearly 40,000 of these women will die from the disease. Breast cancer is the second leading cause of death, in relation to cancer, among women (American Cancer Society, 2011). Women of higher socioeconomic status (SES) are more likely to be diagnosed with breast cancer, yet women of lower socioeconomic status are more likely to die from their breast cancer diagnosis. A major factor in the difference of disease prognosis between women of lower SES and women of higher SES is the availability of healthcare and lack of insurance. Karliner (2007), “when US women who are living in poverty do develop breast cancer, they are more likely to be diagnosed at an advanced stage, are less likely to be treated with breast-conserving surgery and radiation when they have early-stage disease, and are less likely to survive their disease than more affluent women” (Ethnic Disparities in Breast Cancer: Socioeconomic Status). Due to the prevalence of breast cancer and the cost of healthcare among women in lower SES populations I will be analyzing a study which appeared in Journal of Women’s Health titled “Prevalence, Healthcare Utilization, and Costs of Breast Cancer in a State Medicaid Fee-for-Service Program”. The study was aimed at determining occurrence rate, medical intervention and the costs associated when Medicaid patients are diagnosed with breast cancer. A group of breast cancer patients enrolled in the West Virginia Medicaid Fee-for-Service (FFS) program was matched 1:1 with a control group of women enrolled in the same program but without breast cancer. Statistical procedures used in this study consisted of Chi-Square test, Kruskal-Wallis test and Wilcoxon Match-Pairs Signed Ranks test. The Chi-square test was used to investigate whether categorical distributions varied from one another. Kruskal-Wallis testing was used to compare the average breast cancer-related healthcare cost per recipient based on demographic characteristics. Wilcoxon test was used to compare healthcare utilization and costs between the control group and the breast cancer group (Khanna, R., Madhavan, S., Bhanegaonkar, A., & Remick, S. C., 2011). All three of the statistical procedure tests used to measure sample data in this report were non-parametric statistical tests. The Chi Square (X2) test is one of the most important tests in the nonparametric statistical category. Chi Square is used to test the difference between an actual sample and another distribution, typically hypothetical, such as one due to chance or probability. Chi Square test can also be used to test two or more actual samples, identifying differences (Key, 1997). Kruskal-Wallis one-way analysis of variance (ANOVA) is another nonparametric statistical test which is used to compare three or more independent groups of sample data. The Kruskal-Wallis test doesn’t make assumptions about data distribution instead it ranks the data to calculate statistics (TexaSoft, 2008). The Wilcoxon test compares two paired groups by calculating the difference between each set of pairs and analyzing the list of differences. The study validated that there is a high breast cancer incidence among the Medicaid FFS program, then went on to highlight prevalence rates according to demographic differences. The study confirmed that the highest prevalence of breast cancer related costs was seen amongst older white women who resided in non-metro rural counties. “The rate of breast cancer-related hospitalization among women recipients residing in non-metro urban counties was more than twice the rate of hospitalization among women recipients residing in metro counties or non-metro rural counties” (Khanna, Madhavan, Bhanegaonkar, & Remick, 2011, p. 742). The conclusions ascertained by the study are appropriate in relation to a diagnosis of breast cancer and the costs associated with healthcare in relation to this diagnosis. According to National Cancer Institute (2011), “the median age at diagnosis for cancer of the breast was 61 years of age” (SEER Incidence). Breast cancer diagnosis coupled with comorbidities in age group 60-64 years accounts for increased cost and hospitalization utilization in this demographic variable. I do, however, feel that the length of study was inappropriate to make a complete cost evaluation related to breast cancer and disease related expenses. Advanced stages of breast cancer can be treated for several years and the costs can accumulate rapidly over time. A 1 year study does not capture all of these variables. “The all-cause healthcare costs were significantly higher for women recipients with breast cancer compared to those without breast cancer ($16,345 vs. $13,027, p<0.001)” (Khanna, Madhavan, Bhanegaonkar, & Remick, 2011, p. 739). A normal p value would be considered significant if it was in the <0.01 to <0.05 range, meaning there was between a 1%-5% chance the findings were a statistical accident. The statistical significance (Alpha level) found in this study would be considered highly significant due to its p value of <0.001 or 0.1%. The researchers in this study had a very low threshold, leading to a higher confidence level. The study provided data that was statistically significant, thus accepting the null hypothesis: “breast cancer diagnosis among women recipients in the WV Medicaid FFS program was found to be associated with higher all-cause healthcare use and costs compared to women recipients in the matched control group” (Khanna, Madhavan, Bhanegaonkar, & Remick, 2011, p. 739). The costs associated with breast cancer diagnosis and their related medical interventions cause an increased burden to individuals and their insurance providers. With the rate of breast cancer diagnosis on the rise each year studies used to determine areas of cancer-related medical utilization and cost attributes can help provide better opportunities for cost containment and healthcare management, permitting underprivileged women opportunities to be diagnosed at an earlier stage of breast cancer, allowing them a greater chance of survival. References American Cancer Society. (2011). Learn about cancer. Retrieved from http://www.cancer.org/Cancer/BreastCancer/OverviewGuide/breast-cancer-overview-key-statistics Karliner, L.S. (2007). Medscape news today. Retrieved from http://www.medscape.com/viewarticle/566827_6 Key, J.P. (1997). Chi Square. Retrieved from http://www.okstate.edu/ag/agedcm4h/academic/aged5980a/5980/newpage28.htm Khanna, R., Madhavan, S., Bhanegaonkar, A., & Remick, S. C. (2011). Prevalence, Healthcare Utilization, and Costs of Breast Cancer in a State Medicaid Fee-for-Service Program. Journal of Women's Health (15409996), 20(5), 739-747. doi:10.1089/jwh.2010.2298. National Cancer Institute. (2011). Surveillance, epidemiology and end results (SEER) program. Retrieved from http://seer.cancer.gov/statfacts/html/breast.html TexaSoft. (2008). Winks statistics software. Retrieved from http://www.texasoft.com/winkkrus.html