Comparison of precision of a paperless electronic input method versus the conventional paper form in an andrology laboratory: a prospective study.

Abstract

Background: Manual counting for semen analysis is recommended by the World Health Organization. Technicians performing this usually record their results on a paper worksheet and then enter the data into an electronic laboratory information system. One disadvantage of this approach is the chance of post-analytical transcription errors, which can be reduced by checking the computer entries before reporting by another technician. Such practice inevitably increases the running cost and delays the reporting time. The present study was to establish a paperless electronic data entry system for semen analysis and compare its precision with the conventional paper method. During semen analysis, readings on the cell counter were video recorded. The precision of the paper record entries was determined by comparing them with the corresponding video records. Patient characteristics and semen analysis results were input directly into an in-house developed data entry system via a tablet computer immediately after analysis. The same set of data was also handwritten on a paper form and was subsequently input into a standard computerized database according to routine practice. The agreement of the data entries between the two systems was then compared.

Results: A total of 787 semen analyses were included in the study, involving 201 samples in Phase I and 586 samples in Phase II of the study. Phase I was the initial learning period. The overall rate of transcription error of the paper form was 0.07%, whereas that of the paperless system was 0.17%. In phase II, the paperless system was modified according to users' comments. The transcription error rate of the paper form was 0.05%, while that of the paperless system was substantially reduced to 0.01% (p = 0.008).

Conclusion: The paperless system is a reliable tool for recording data from semen analysis compared with the conventional paper form. However, training is needed to reduce the error rate of the paperless system.