Fatalism and Treatment Preferences of Lung and Colorectal Cancer Patients

Preferences and Beliefs of Lung and Colorectal Cancer Patients: The Role of Demographic and Socio-Economic Variables Julie Urmie, Ph.D. , Elizabeth Chrischilles, Ph.D., Jane Weeks, M.D., John Brooks, Ph.D., Jane Pendergast Ph.D., Dawn Provenzale, M.D., David Nerenz, Ph.D., Dee West, Ph.D. Academy Health Research Meeting, June 2006 1 Background    Increasing recognition of the need to include patient preference measures into research on cancer treatment and outcomes.1-4 Limited past research on patient preferences in cancer patients, particularly in quantitative studies. The CanCORS study includes several patient preference measures. 2 Background  A better understanding of the link between preferences and demographic/ socio-economic variables is necessary before including them as predictors of treatment choice and outcomes.  How much variation in preferences is already explained by demographic and socio-economic variables?  Past research provides some evidence of a link between demographic/socio-economic factors and preferences.   Older patients favor quality of life over quantity of life.5 High levels of fatalism in certain racial groups and low income elderly.6-9 3 Objectives & Research Questions  To examine the relationship between fatalism/treatment preferences and sociodemographic factors in lung and colorectal cancer patients.  Do low income or low education patients have different preferences than high income or high education patients?   Do preferences differ by racial/ethnic group? Do men have different preferences than women? 4 Data Source  Data were from the CanCORS baseline patient survey.  Data collection was not complete at the time of data analysis so results are preliminary.   Both colorectal and lung cancer patients were included. Three main preferences measures were used:    Fatalism Quantity of Life vs. Quality of life (QOL) Quantity of Life vs. Cost 5 Preference Variables  Fatalism: A four item scale adapted from the Powe Fatalism Index      “When bad things happen, we are not supposed to know why, we are just supposed to accept them.” “People die when it is their time to die and nothing can change that.” “Everything that happens is part of God’s plan.” “If bad things happen, it is because they were meant to be.” Four response categories: SD, D, A, SA  Scores were summed to create one fatalism score. 6 Preference Variables  Quantity of Life vs. Quality of Life Tradeoff   “If you had to make a choice now, would you prefer treatment that extends life as much as possible, even if it means having more pain and discomfort, or would you want treatment that focuses on relieving pain and discomfort as much as possible, even if it means not living as long.” 2 response categories 7 Preference Variables  Quantity of Life vs. Cost Tradeoff   “If you had to make a choice now, would you prefer treatment that extends life as much as possible, even if it means using up all of your financial resources, or would you want treatment that costs you less, even if it means not living as long?” 2 response categories 8 Statistical Models     OLS Regression predicting fatalism Logistic regression predicting quality of life vs. quantity of life tradeoff (quality of life = 0; quantity of life = 1). Logistic regression predicting cost vs. quantity of life tradeoff (cost = 0; quantity of life = 1) All three models used unweighted estimates. 9 Independent Variables         Age Gender (female = 0; male = 1) Cancer type (colon = 0; lung = 1) Education (less than HS grad = reference; HS grad; some 4year or trade; 4-year grad; some grad school or professional) Race (white,non-Hispanic = reference; white, Hispanic; Black; Asian; Other) Living Arrangement (married/living with partner = 1; else = 0) Income (less than $10,000 = reference; $10,000 to less than $20,000; $20,000 to less than $40,000; $40,000 to less than $60,000; $60,000 or more) Health insurance status (no insurance = 0; insurance = 1) 10 Distribution of Fatalism Scores Histogram of Fatalism Variable 1000 800 Count 600 400 200 0 4 5 6 7 8 9 10 11 12 13 14 15 16 N =4,129 N = 3,807 Fatalism •Mean = 10.65 (S.D. = 2.58). •Higher scores signify higher levels of Fatalism. •Cronbach’s Alpha = 0.801 11 Distribution of Quantity vs. QOL & Quantity vs. Cost 70 60.7 60 50 49.3 50.7 Percent 40 30 20 10 0 Quantity QOL N = 4,127 Quantity N = 3,948 39.3 Cost 12 Multivariate OLS Results for Fatalism  Higher levels of fatalism:     People who were married/living with partner (p < 0.01). African-Americans and Asians (p < 0.01). White, Hispanics (p = 0.06), Other Race (p = 0.07). Older adults (p = 0.08) Men (p <0.01). All higher education levels (p <0.01). The three highest income categories (p < 0.05).  Lower levels of fatalism:     Insurance coverage and type of cancer were not significant at p < 0.10.  Demographic/socio-economic variables explained 12.6 percent of variation in fatalism. 13 Multivariate Logistic Results for Quantity of Life vs. QOL Tradeoff  Significant odds ratios for the model predicting a preference for quantity of life over QOL.     Age OR = 0.97 (p < 0.01) Married/partner OR = 1.32 (p < 0.01) African-AmericanOR = 1.33 (p < 0.01) Male OR = 1.41 (p < 0.01) 14 Multivariate Logistic Results for Quantity of Life vs. Cost Tradeoff  Significant odds ratios for the model predicting a preference for quantity of life over conserving resources.      Married/partner OR = 0.73 (p < 0.01) Age OR = 0.97 (p < 0.01) Male OR = 1.14 (p = 0.08) Asian OR = 1.59 (p < 0.01) African-AmericanOR = 2.78 (p < 0.01) 15 Limitations  Preliminary Results.      Baseline survey data not complete at the time of analysis. Missing data. No sampling weights used. No interaction effects were examined. No comorbidity and staging information was included. 16 Conclusions   There are some important and complex differences in preferences among different socio-economic and demographic groups. This complexity and the limited amount of variation in preferences explained by sociodemographic variables show the importance of including explicit measures of preferences in models predicting treatment choice and outcomes. 17 References          1. Kassirer JP. Incorporating patients preferences into medical decisions. The New England Journal of Medicine. 1994;330(26):1895-96. 2. Silvestri G, Pritchard R, Welch G. Preferences for chemotherapy in patients with advanced non-small cell cancer: descriptive study based on scripted interviews. British Medical Journal. 1998;317:771-75. 3. Redmond K. Assessing patients’ needs and preferences in the management of advanced colorectal cancer. British Journal of Cancer. 1998;77(S2):5-7. 4. Earle CC, Vendetti LN, Neuman Pj et al. Who gets chemotherapy for metastatic lung cancer? Chest. 2000;117(5):1239-46. 5. Stiggelbout AM, De Haes JM, Kiebert GM, Kievet J, Leer JH. Tradeoffs between quality and quantity of life development of the QQ questionnaire for cancer patient attitudes. Medical Decision Making. 1996;16:184-192. 6. Powe BD. Fatalism among elderly African Americans: effects on colorectal screening. Cancer Nursing. 1995;18(5):385-392. 7. Powe BD. Cancer fatalism among elderly Caucasians and African Americans. Oncology Nursing Forum. 1995;22(9):1355-1359. 8. Powe BD and Weinrich S. An intervention to decrease cancer fatalism among rural elders. Oncology Nursing Forum, 1999;26(3):583-588. 9. Chavas LR, Hubbell FA, Mishra SI, Valdez RB. The influence of fatalism on selfreported use of papanicolaou smears. American Journal of Preventive Medicine. 1997;13(6):418-424. 18