CHARACTERIZATION OF CANCER MANAGEMENT IN THE ELDERLY
Funded Grant
Overview
Affiliation
View All
Overview
description
We propose to utilize five data sets collected by our group to describe the clinical attributes of elderly cancer patients and compare these attributes and aspects of clinical management to those of younger patients. The data sets include 1) a study of over 3,700 cancer patients and 1,700 noncancer patients to establish the accuracy of death certification. This included information from one large referral hospital (Duke University), a Veterans Administration Hospital, and a number of community hospitals in three regions in North Carolina. 2) The Duke Comprehensive Cancer minimal data base including information on all patients with cancer seen at this institution, and 3) the Comprehensive Cancer Center specialized data bases for breast cancer, prostate cancer, and melanoma. Information is included in these data sets which will allow analyses of patient age, clinical presentation, time to presentation, type of initial treatment, place of initial treatment, extent and type of disease, other medical conditions, complications of illness, extent of treatment initiated, terminal events, place of death, cause of death, accuracy of death certification. The specialized data sets will allow assessment of relationships to cancer stage, location, histologic type, diagnostic and therapeutic choices, effective comparable treatment regimens for response, duration of response and time to relapse, effect on toxicity, time in hospital, wound healing, caretaker situations, relationship to prognosis. An interdisciplinary team including expertise in geriatrics, oncology, gerontology, epidemiology, and biostatistics will make assessments of the relationship of age to these various parameters. Both standard statistical methods such as Cox regression models, as well as models developed at this center including the Grade of Membership Analysis, specifically developed for simultaneously identifying clinical subgroups of patients and multivariant descriptions of the attributes of these patients, will be employed. Such analyses should provide information which can be utilized to formulate appropriate approaches to diagnosis, intervention and support for the elderly cancer patient. In the long run, we plan to use our interdisciplinary team to implement those followup programs suggested by the current study.
