CLAIMS DATA TO MEASURE CANCER SURVEILLANCE PROCEDURES

点击翻译按钮获取中文摘要

Follow-up testing to detect cancer recurrence and new primary tumors is commonly advocated for patients who receive potentially curative initial therapy, but there is little information on the actual use of testing, as well as variation in testing across subgroups of patients. Cancer registry systems such as the Surveillance, Epidemiology and End Results (SEER) Program typically collect data on therapy received for the first three to four months following diagnosis, and thus SEER cannot be used to study the use of procedures after the initial course of treatment. Therefore, investigations of follow-up testing generally are performed by review of paper medical records from individual providers. In Preliminary Studies, the Investigators have analyzed health plan and Medicare claims to determine patterns of follow-up care after initial therapy for colorectal cancer, and have used the electronic medical record to ascertain the indication for surveillance procedures. Because of reduced research costs and easier accessibility compared to the paper chart, both health claims and the electronic medical record represent potentially important sources of data to study surveillance procedures. The goal of the proposed study is to determine the utility, or completeness and accuracy of two alternative sources to the paper chart - the electronic medical record and health claims data, as measures of surveillance testing. Using data from a large, vertically integrated provider network that includes both fee-for-service and one of the country's largest managed care plans, a cohort of patients with localized or regionally invasive prostate, lung, colorectal, breast, and endometrial cancer, and non-Hodgkin's lymphoma will be studied. Based on the results of paper and electronic medical record review, a composite gold standard measure for the use of and indication for testing will be developed. The next phase of the study will evaluate the accuracy of health claims data for both fee for service and managed care enrollees in measuring the use of surveillance testing, compared to the composite gold standard. The feasibility of using claims data to categorize patterns of follow up as standard or intensive will be also ascertained. Because of the potentially enormous economic and societal implications of follow-up testing, and the paucity of data from other than paper chart review, this represents a vital area of investigation. The study will provide a methodological basis for future investigations that will use larger, nationally-based sources of electronic data to examine variations in practice among patient subgroups, as well as the clinical and economic impact of various surveillance strategies.