Psychometric validation of the visual function questionnaire-25 in patients with diabetic macular edema

BACKGROUND: Diabetic Macular Edema (DME) is a common cause of impaired vision and blindness amongst diabetics. If not detected and treated early, the resulting vision loss can lead to considerable health costs and decreased health-related quality of life (HRQoL). The aim of this study was to provide evidence of the psychometric properties of the National Eye Institute - Visual Functioning Questionnaire (VFQ-25) for use in a cohort of DME patients who participated in a clinical efficacy and safety trial of pegaptinib sodium (Macugen). METHODS: A phase 2/3 randomised, double masked trial evaluated pegaptanib injection versus sham injection in patients with DME. The analysis was conducted using baseline HRQoL data of the VFQ-25 and the EQ-5D, on a modified intent-to-treat sample of 235 patients. These measures were administered by a trained interviewer by telephone in all but one of the study countries, where face-to-face interviews were conducted in the clinic. The measures were completed in the week prior to baseline, and after 54 weeks of treatment. Distance visual acuity, measured according to the Early Treatment Diabetic Retinopathy Study (ETDRS), was assessed at all time points. Psychometric properties of the VFQ-25 assessed included domain structure, reliability, concurrent and construct validity, responsiveness. RESULTS: The VFQ-25 was found to consist of 11 domains slightly different than those proposed. Nevertheless, none of the eight established multi-item scales met the criterion for further splitting and the VFQ-25 was scored as in the developers’ instructions. Internal consistency reliability was demonstrated for six out of the eight original multi-item scales, with Cronbach's alpha ranging from 0.58 (Distance Activities) to 0.85 (Vision Specific: Dependency). The VFQ-25 domains generally showed a low to moderate correlation with EQ-5D visual analogue scale (range 0.16-0.43) and with the visual acuity score (range 0.10-0.41). Construct validity was upheld with higher VFQ-25 scores for patients who saw more letters according to the ETDRS. Almost all scales were shown to be responsive with Guyatt's statistic ranging from 0.10 to 0.56 at 54 weeks. CONCLUSIONS: The VFQ-25 has evidence to support its validity and reliability for measuring HRQoL in DME. However, some operating characteristics of the instrument need further consideration and discussion in the case of DME patients. Further research is therefore warranted in this indication.