Purpose
Although dry eye is often considered a variable condition, little is known about the relationship between habitual symptoms (measured from recall periods of days, weeks or months) and symptoms measured at the time of testing. The aim of this study was to validate the Current Symptoms Questionnaire (CSQ) and examine its relationship to habitual symptoms through a meta-analytical approach.
Methods
Baseline CSQ data (no intervention) from normal and dry eye subjects in 12 studies that took place between 2002 and 2011 were compared to habitual symptoms concurrently measured by the Dry Eye Questionnaire (DEQ). Receiver-Operating Characteristic (ROC) curve analysis was used to compare CSQ scores against the DEQ-5, a validated short form of the DEQ comprising a subset of five DEQ questions, and a prior clinical diagnosis of dry eye. Rasch analysis was used to determine some psychometric properties of the CSQ.
Results
There were correlations between current and habitual symptoms of discomfort and dryness with respect to their frequency and intensity (Spearman r ranged 0.40-0.57, all p<0.05). The areas under the ROC were similar between CSQ and the DEQ-5 (0.84 and 0.81 for CSQ and DEQ-5, respectively). Rasch item classification curves showed that the responses to certain questions (e.g., dryness & discomfort) were ordered while others (e.g., light sensitivity) were answered in a more disorganized way. This disorder was reduced when categories were reduced (3 categories providing consistent responses). Rasch infit & outfit statistics pointed to questions about blur, light sensitivity and itching being the least compatible with the CSQ testing along a single latent dimension (all mean square fit, p<0.05).
Conclusions
Symptoms of discomfort and dryness measured at the time of testing are associated with habitual symptoms from a longer recall period. Similar to the DEQ-5, CSQ scores effectively segregated subjects with and without a prior diagnosis of dry eye. In addition, the uni-dimensionality of CSQ to certain discomfort related symptoms suggests that CSQ is a valid instrument to be use for the assessment of ocular dryness and discomfort.
Keywords: 468 clinical research methodology •
486 cornea: tears/tear film/dry eye