Table 1: Scores of the questionnaire for visual discomfort for yoga and control groups at baseline [BL] and day 60.

*** P < .001, post-hoc test for multiple comparisons

Visual discomfort in professional computer users is con- tributed to by various factors such as lighting, glare, display quality, ergonomic positioning of the monitor and regularity of work breaks [6]. The symptom which largely contributes to subjectively rated visual discomfort is 'dry eye'. Dry eye is itself contributed to by various factors, including certain diseases (e.g., Sjögren's syndrome, use of certain medication (e.g., anti-histaminics), gender (being more common in females)), and individual factors [17]. Individual factors include blink rate and completeness of blinking which significantly affect tear film dynamics and ocular surface health [18,19]. Blink rate especially has been shown to vary with the task performed [20]. The mean (± S.D.) rate of blinking was 22 (± 9) per minute under relaxed conditions, 10 (± 6) per minute while the subjects were reading a book at table level, and 7 (± 7) per minute while working at a video display terminal. Hence the frequency of blinking reduces while mentally alert and with gaze focused.

Specific yoga practices have been found to bring about physiological changes suggestive of 'alertful rest' [21]. This description was based on a simultaneous decrease in heart rate and oxygen consumption along with a reduction in peripheral cutaneous blood flow. Also the visual cleansing practices used in the present trial have been shown to facilitate visual perceptual sensitivity in terms of a decrease in optical illusion [22]. A reduction in anxiety has been found to be associated with better visual perceptual sensitivity [23]. A relaxed state (as described above) is associated with a higher frequency of blinking. Yoga practice has been associated with better self rated relaxation [24] as well as with physiological relaxation [25]. Hence the reduction in visual discomfort in the yoga group in the present study may be attributed to an improvement in the ability to focus while remaining relaxed which may have increased the blink rate.

In contrast to the yoga group the control group showed an increase in self rated visual discomfort. These differences between the groups could be due to psychological benefits that are reported with 'additional care' [26]. In the present study the frequent meetings which the yoga group had with the instructor could serve as additional care and may have contributed to the benefits seen in the yoga group. The absence of this psychological support and the yoga practice in the control group may have contributed to increased visual discomfort at follow-up.

A main limitation of the study is that well recognized objective indicators of visual discomfort (especially dryness) were not measured. It would have been ideal to have carried out a semi-quantitative estimation of the superficial lipid layer or have measured the tear breakup time [27]. However another variable which is an objective indi- cator of VDT related fatigue was measured in these subjects, and the results were reported elsewhere [28]. This is the critical flicker fusion frequency (CFF), which is the flicker frequency rate beyond which one can no longer perceive the flicker. Flicker related changes in the visual system from working at a cathode ray tube (CRT) compu- ter screen have also been measured [29]. A group of subjects worked for 3 hours at simulated CRT displays with different flicker rates. The CFF was found to decrease. A similar result was obtained when the effect of performing the same task on a CRT was compared with the perform- ance on a back slide projection system (BPS) [30]. The CFF of the group decreased after working on the CRT com- puter screen while it did not change when working on the BPS. These studies suggest that the visual system possibly gets fatigued as a result of viewing supra-threshold flicker.

Table 2: Analysis of variance for scores in the 'Dry Eye Questionnaire'


Greenhouse-Geisser epsilon = 1.000, hence Sphericity Assumed