Comparison of keyboard, light pen and voice recognition as methods of data input.

There is an increasing requirement for doctors and nurses to interact with computers for clinical and administrative reasons. It is important to assess different methods of computer input to provide the most suitable type of input device for any particular application. A study was undertaken to compare three different input devices--standard 'QWERTY' keyboard, light pen, and voice recognition--for speed, errors and number of corrections during entry of patient admission data. Twenty intensive care nurses took part in the study and were asked to express their assessment of speed, accuracy, ease of use and preference. The average speed of data entry by keyboard (37.70 s) was significantly faster than light pen (58.09 s; p less than 0.001); while light pen entry was significantly faster than voice recognition (111.65 s; p less than 0.001). There was no difference between the average number of errors for keyboard (0.10) and light pen (0), but significantly more errors occurred with voice recognition (1.05) than light pen (p less than 0.001). Significantly more corrections were required with the light pen (0.50) compared with the keyboard (0; p less than 0.001), and between voice recognition (11.95) and light pen (p less than 0.001). Keyboard input was considered by the intensive care unit staff as the quickest, most accurate, easiest and most preferred method of entering data into a computer.