Purchase this article with an account.
Aachal Kotecha, William J Feuer, Keith Barton, Steven Gedde; Minimally important Differences in the NEI-VFQ 25: Data from the TVT study. Invest. Ophthalmol. Vis. Sci. 2016;57(12):3910.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To report the minimally important difference (MID) for the National Eye Institute-Visual Function Questionnaire (NEIVFQ-25) for the TVT study cohort.
The Tube Versus Trabeculectomy (TVT) study is a multicenter randomized clinical trial comparing tube-shunt surgery to trabeculectomy with mitomycin C (MMC) in patients with previous ocular surgery. Patient reported outcomes were assessed using the vision-specific NEIVFQ-25. 202 of 207 patients enrolled in the study had NEIVFQ measured at baseline. Clinical measures included the mean deviation (MD; Humphrey 24-2 visual field) of the surgical eye, and logMAR visual acuity (VA) measurements. Approximately half these patients had surgery performed in their better acuity eye; thus data were stratified to ‘better’/ ‘worse’ eye MD. Distribution and clinical anchor based methods were used to determine MID ranges. For anchor-based calculations, clinically significant changes in anchor were defined as > 2dB MD and > 0.2 logMAR. MIDs were only calculated for if the correlation between anchor and NEIVFQ score exceeded 0.3.
Correlations between composite score and clinical anchors exceeded 0.3 for surgical eye MD, better eye MD and better eye acuity. Mean [SD; range] values of anchors at baseline were: surgical eye MD = -16.6 [9.3; -32 to -0.5] dB; better eye MD = -15.0 [9.3; -32 to -0.9] dB; better eye acuity = 0.2 [0.3; -0.1 to 1.3] logMAR. For anchor-based cross sectional analysis, composite score MID [95%CI] was 6.3 [4.6 to 7.9] for better eye VA, 1.0 [0.6 to 1.4] for surgical eye MD and to 1.8 [1.2 to 2.5] for better eye MD. Distribution-based MIDs were 5.8 using 1/3 standard deviation of the composite score. MIDs for subscores were also calculated.
In this cohort of patients with advanced glaucoma, a range of MIDs for the NEIVFQ were found, some of which are comparable to results reported by other studies using the NEIVFQ. Distribution based MIDs corresponded well with anchor-based MIDs based on acuity measures.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.
This PDF is available to Subscribers Only