Abstract
To determine patterns of mammography utilization in Arkansas, the Arkansas Mammography Data Collection Project (MDCP) was established. The project's objective was to compile into one database statewide information about mammograms performed. All mammography centers were invited to participate in the project. Many barriers were encountered that were center related, data related, or personnel related; different interventions were implemented for each barrier. At the conclusion of the project, 92 out of 112 centers (82%) participated in the project, creating a database of 157,976 mammography data sets. Identifying and overcoming many of the barriers were crucial steps in the project's success.
BREAST CANCER is the most common cancer diagnosed and the second cause of cancer death in women in the United States. In the year 2001, it was estimated that 193,700 new cases of breast cancer would be diagnosed in the United States, and 40,200 women would die from the disease.1 In the same year, 1,900 women would be diagnosed with breast cancer in Arkansas and 400 women would succumb to this disease. Early detection is critical because it offers the patient a better chance of cure. Mammography is currently the best early detection tool commonly available to diagnose breast cancer, and it is known to decrease breast cancer mortality among screened women.2-4 However, many women do not undergo screening mammography for a variety of reasons.5
The Breast and Cervical Cancer Mortality Prevention Act is aimed to increase the access of medically underserved women to breast and cervical cancer screening.6 The Arkansas Breast and Cervical Cancer Control program established BreastCare, which provides professional and public mammography education to women in Arkansas. In addition, BreastCare provides funding for programs that promote awareness for early detection, diagnosis, and treatment of breast cancer. The Arkansas Mammography Data Collection Project (MDCP) was funded by BreastCare to establish a statewide database describing the patterns of mammography utilization throughout the state's 75 counties. This database would be used to determine the actual mammography rates in each county, assess the needs of certain geographical area or population, and provide baseline data to measure the trends and evaluate the efficacy of public health interventions. The plan was to compile data by obtaining information about the mammograms performed at all state mammography centers. This method would eliminate many biases inherent in other data collection methods that are based on sampling and patient surveys or interviews, such as sampling bias and recall bias.7 The MDCP goal was to obtain the mammography data from 1997, since many of the projects funded by Breast Care would be launched in 1998. The data collected by the MDCP would serve as a baseline for evaluation of the efficacy of these interventions.
This article discusses the process of establishing collaborations with various mammography centers to create the statewide database. In addition, it presents the barriers that were encountered and the approaches followed to overcome many of these barriers. This information may be of value to public health workers and researchers considering establishing a network or collaboration among different health care entities that are geographically dispersed.