Is critical flicker-fusion frequency a valid measure of visual fatigue? A post-hoc analysis of a double-masked...

Critical flicker-fusion frequency (CFF) has been used in clinical studies as a measure of visual fatigue. We examine the correlation between CFF and subjective reports of visual fatigue in a group of symptomatic computer users, to consider whether CFF may be used as a surrogate measure of visual fatigue symptoms.

We analysed data from a previous randomised controlled trial. One hundred and twenty adults, diagnosed with computer vision syndrome, had CFF and visual fatigue symptoms quantified before and after a visually demanding 2-h computer task. Symptoms were assessed using a questionnaire with nine subcomponents that summed to a total score of 900. CFF was measured using a two-interval forced-choice method, with the flicker rate altered by a computer-controlled staircase procedure. For our primary analysis, we determined Spearman correlation coefficients between post-task symptom scores and CFF, and between change from baseline symptom scores and CFF. We also used a bootstrap procedure to consider whether symptom score subcomponents were significantly (Bonferroni-corrected) different from overall scores with regard to their correlations with CFF.

Although visual fatigue symptom scores altered significantly post-task (mean change: 92 units; 95% confidence interval [CI]: 11 to 122), CFF did not (mean change -0.7 Hz; 95% CI: -1.7 to 0.3). There was no significant correlation between overall symptom scores and CFF, either for the post-task (r = -0.13; 95% CI: -0.31 to 0.05) or the change from baseline (r = -0.18; 95% CI: -0.35 to 0.01) analysis. Subcomponents of the symptom questionnaire did not show a significant correlation with CFF, either for the post-task or the change from baseline analysis.

We find that CFF is not a useful surrogate for symptoms of visual fatigue, given its low correlation with scores on a visual fatigue symptom questionnaire.