Cancer Patient-Specific Prognostic Information

The goal of this project is to develop an interactive, Web-based Prognostigram program for adult patients with newly diagnosed cancer. The prognostic program creates individualized survival curves based on the Cox Proportional Hazards model of survival data from Barnes-Jewish Hospital (BJH) Oncology Data Services (ODS) and SEER*Stat (National Cancer Institute, Surveillance, Epidemiology, and End Results Software package). This program also includes important comorbid health information. Patient and physician focus groups were held to explore the utility of the program. Overall, both groups felt the program was informative and easy to use. Specific suggestions for improvement were made. Ultimately, the utility of the program needs to be assessed with Introduction Currently, prognostic estimates and decision-making in cancer care are primarily based on 5-year survival statistics published by the American Joint Committee on Cancer (1) and the American Cancer Society (2). These data relate mortality to site and TNM staging, but do not account for patient-specific factors such as patient age, gender, comorbidity, and cancer-related symptom severity. Since these factors are also independent predictors of survival (3-17), an index that considers these factors in addition to tumor site and TNM stage will be a better system for estimating survival. Information resources available to patients with newly diagnosed cancers are diverse. The physician and nursing staff serve as the primary providers of meaningful, appropriate information. Patients can supplement this information with general background information about cancer and cancer therapies from printed pamphlets, articles, and textbooks. Much of this didactic material is intended for non-medical audiences and is relatively easily understood. On the other hand, prognostic information, such as life survival or five-year survival rates, is more difficult for laymen to find and meaningfully interpret. At present, patients’ understanding of prognosis is generally limited to information provided to them by their physician. Physicians obtain prognostic information from the published literature and their personal experience with similarly affected patients. Interactive teaching and decision-making programs for patient use exist in varying medical specialties (18-20) One of the most successful programs is a videodisc series from the Foundation for Informed Medical Decision-Making (Hanover, NH). This series can help patients with back pain, breast cancer, ischemic heart disease, or benign prostatic hypertrophy understand their conditions and the various treatment options available to them. Other interactive teaching and decision-making resources for patients exist on the World Wide Web. For example, the American Heart Association "What’s My Risk" page (http://www.amhrt.org/risk/) uses baseline data from the Framingham study to determine the user’s ten-year risk of heart disease based on age, gender, race,