Detecting Early Tumors Enables Cancer Therapy (DETECT)

Background

DETECT is based on the notion that all women in HMOs have access to screening and care but some still have unfortunate outcomes. The question is why. Working with colleagues from nine HMOs around the United States, the DETECT team is evaluating whether late-stage breast and invasive cervical cancers occur because women are not screened, cancers are not detected when screening occurs, or follow-up does not occur when an abnormality is found. In addition, the team is assessing screening policies and implementation at the nine sites by surveying leaders, patients, and primary care providers.

Study Aims

1. Identify the proportion of late-stage breast (>3cm at the time of diagnosis) and invasive cervical cancers attributable to:
   • absence of screening
   • absence of detection
   • breakdown in follow-up
   • biologic characteristics of the tumor
2. Estimate the relative risk of late-stage breast cancer as a function of the three potential failures in screening implementation during the three years prior to the year of diagnosis.
3. Explore reasons for presence or absence of the screening services by interviewing key informants and women with positive screening examinations.
4. Determine whether rapid onset cervical cancer cases differ from others cases in terms of human papillomavirus.
5. Validate the automated data sources.

Methods

Multiple data sources are being used to elucidate the factors that contribute to breakdowns in the process of cancer screening, detection, and follow-up. The DETECT team is assessing screening policies and implementation at the nine sites by surveying leaders, patients, and primary care providers. The study is comprised of three distinct but related components: organization of cancer preventive care; occurrence of late-stage breast cancer; and occurrence of invasive cervical cancer. To address these components, multiple data collection modalities are being used, including patient surveys, provider surveys, key informant interviews, review of medical records, and review of organizational guidelines.

The DETECT study team is committed to making study findings and instruments available to the scientific community as a whole. The instrument used to abstract the breast cancer-related data, and the coding instructions for this abstraction instrument, are provided below in PDF format.

• Breast Cancer Chart Audit
• Coding Instructions for the Breast Cancer Chart Audit

Significance

The DETECT study is assessing factors that influence types of cancer care and the transitions between them, the factors at several levels (community, plan, and practice) that potentially impact access and quality, and the strategies groups and organizations can consider to reduce potential failures. Focusing on the steps and transitions in care where failures can occur can facilitate more organized systems and medical practices that improve care, establish meaningful measures of quality that promote improved outcomes, and enhance interdisciplinary research.

Project Status

This project was funded as part of the CRN's initial four-year funding cycle. The project leader is Stephen Taplin, MD, MPH. Findings from the main paper on the occurrence of late-stage breast cancer are available in the October 20, 2004 issue of the Journal of the National Cancer Institute. Other analyses are continuing. The CRN Publications section also includes results from the organizational assessment component of this project.

Return to top