Lessons from developing and running a clinical database for colorectal cancer.

Area of Science:

• Health Services Research
• Cancer Informatics
• Clinical Data Management

Background:

• UK policy mandates routine cancer service performance monitoring.
• Clinical databases are a key tool for this monitoring.
• Implementation of such databases has faced significant challenges.

Purpose of the Study:

• To detail the development of the Thames Cancer Registry clinical database for colorectal cancer.
• To present lessons learned over the first five years of its operation.

Main Methods:

• Database planning initiated in 1998, informed by national standards.
• Data variables derived from guidelines; structured proformas used for data abstraction.
• Pilot study conducted, followed by wider data collection across the North Thames region.

Main Results:

• Database records increased from 747 (1999) to 1107 (2002), reaching 8500 by 2004.
• Trust participation and data completeness varied; only 18/26 trusts submitted data, 12 annually.
• Completeness for key variables was high (80-100%), but lower for detailed diagnostic/treatment data (40-60%).
• Barriers included organizational (network changes, commitment, staffing), professional (priorities, team commitment), and data-related issues (format variety, delayed entry).

Conclusions:

• Establishing and maintaining clinical databases is resource-intensive and complex.
• Data completeness and quality are critical concerns.
• Clinician commitment and consistent trust support are essential for successful data collection and registry maintenance.