Technology for Optimizing Population Care in a Resource-Limited Environment

Using health information technology (IT) in the management of primary care populations will likely improve efficiencies; however ways to use technology in this area is not well understood. The use of an automated reminder system to prompt patients of the need for preventive maintenance and screening may improve efficiencies by occurring outside patient visits and bypassing clinicians. However, it is unknown whether having clinician input into these reminders is more efficient than purely automated systems.

The Technology for Optimizing Population Care in a Resource-limited Environment (TopCare) project designed, developed, implemented, and evaluated a system for automated reminders for breast, cervical, and colorectal cancer screening within the Massachusetts General Primary Care Practice Based Research Network (MGPC-PBRN). The goal of this project was to improve clinical decision support and enhance preventive cancer screening. It compared involving primary care physicians (PCPs) in the screening process to a fully automated process in order to determine whether screening rates differed when outreach was linked to the patient’s needs.

The specific aims of this project were to:

• Design, develop, and implement a novel cancer screening intervention program (TopCare) that facilitates the identification, individualized contact, and subsequent tracking of patients overdue for screening. 
• Conduct a practice-randomized trial of the TopCare program within the MGPC-PBRN assessing its impact on cancer screening rates in eligible patients. 
• Collect data prospectively throughout the randomized trial on costs, preferences, and clinical and process outcomes to inform a subsequent formal cost-benefit analysis. 

A randomized controlled trial of 18 practices was conducted. The screening program was integrated with electronic health record data to assess whether clinical decision support could efficiently enhance preventive care in a primary care setting. In the intervention practices, PCPs were given lists of their patients who were overdue for screening. They determined if screening was needed, and customized contact with patients by letter, practice delegate, or patient navigator. In the control practices, patients who were overdue for screening were automatically sent reminder letters and transferred to a practice delegate for followup. Patients in the intervention group whose PCPs failed to take action within 8 weeks were moved to the automated system.

Ninety percent of the intervention providers used the tool, reviewing nearly 8,000 patients who were overdue for at least one cancer screening. A total of 12,002 letters were sent to patients in intervention practices versus 16,378 letters in control practices. Screening rates did not differ among intervention and control practices for all cancers combined. A majority of system users reported that the system was easy to use and made time managing cancer screening more effective. Further analyses will explore whether implementation of the TopCare system resulted in higher screening rates.

This project demonstrated that population-based comprehensive screening for cancers is possible to automate with technology tools. Since the completion of the TopCare demonstration project, the system has been implemented as an operational system at Massachusetts General Hospital. The project team recommends similar research to identify preventive care and disease management scenarios that may be good candidates for automated, non-visit based models of care delivery.