Automating Data Entry and Validation in Clinical Research
Clinical research often generates vast arrays of complex, interrelated demographic, patient exam, and laboratory measurement data. Researchers have long contended with the costly, fatiguing and inefficient processes of entering this data on a database and verifying its accuracy prior to report generation and statistical analysis. We have examined several new systems in which data entry forms are scanned, interpreted through Optical Character Recognition technology, and the research data automatically stored on a database.
The Berkeley Contact Lens Extended Wear Study (CLEWS), a large, randomized controlled clinical trial conducted at the University of California at Berkeley School of Optometry, was used as a testing ground for an automated data entry and validation system based on the Teleform software and an Epson scanner.
The system allowed design of data entry forms that satisfied the needs of our clinicians, biostatisticians, and administrative staff. The system drastically reduced the time required to enter patient exam, demographic, and laboratory measurement data onto the study database, and provided tools for verifying that the data were scanned accurately. The system improved both the quality of patient care and the integrity of clinical patient data, allowing clinicians to quickly and easily retrieve patient records, and permitted our biostatisticians to generate periodic recruitment monitoring, patient safety, protocol adherence, and data quality assurance reports in a timely fashion.
Graham, Andrew D.; Fusaro, Robert E.; and Polse, Kenneth A., "Automating Data Entry and Validation in Clinical Research" (April 1998). U.C. Berkeley Division of Biostatistics Working Paper Series. Working Paper 71.