Validity of an information and communication technology system for data capture in epidemiological studies.

The Belgian National Institute of Health Insurance is implementing an oral health data registration and surveillance system. This study aimed to develop and validate a system of electronic data capture for oral health surveys at a national level - Oral Survey-B - and to identify the advantages and disadvantages of the electronic system in comparison with the traditional handwritten data capture. Six series of full-mouth recordings simulating the clinical examination of 6 patients were set up in a Powerpoint presentation. The validation was undertaken by 52 general practitioners. A randomized one-period crossover design was used with two formats of data capture, i.e. electronic followed by handwritten or handwritten followed by electronic system. Further, 6 benchmarked handwritten forms were transferred to the electronic format. For the electronic data capture, 86.5% of the practitioners had a correct completion rate of ≥95%. The corresponding value for the handwritten data capture and transfer was 78.8% (p = 0.25, McNemar test). The overall accuracy of forms without any error was 73.4% for the electronic and 62.5% for the handwritten data capture (p < 0.001, signed-rank test). Significantly lower percentages of errors and less time were observed for the electronic data capture (p < 0.001, signed-rank test). Practitioners considered the electronic data capture as being much more difficult to carry out (p < 0.001). As information technology has turned into an ever more necessary working tool in epidemiology, there should be an important potential for uptake of further improvements in electronic data capture in the future.