Psychometric evaluation of the near activity visual questionnaire presbyopia (NAVQ-P) and additional patient-reported outcome items

Journal of Patient-Reported Outcomes

Table 10 Correlation-weighted average estimates of meaningful improvement for NAVQ-P total score, NVCI, and NVS using change from baseline to Month 3

Meaningful change	Anchor	Estimates	Correlation-weighted estimate	Recommended thresholds
NAVQ-P Total Score
Between-group difference	PGI-S (1 grade)	-8.50	11.2-point change	11.2-point improvement
	PGI-S (2 grade)	-14.80
	PGI-C	-10.20
Within-group change	PGI-S (1 grade)	-12.10	14.8-point change	14.8-point improvement
	PGI-S (2 grade)	-19.80
	PGI-C	-12.00
Within-patient change (responder definition)*	PGI-S (1 grade)	-8.0	9.5-point change (ROC only) 12.1-point change (ROC and within-group change)	Range of 8 to 15-point improvement (10-point change recommended as proposed threshold)
	PGI-S (2 grade)	-14.0
	PGI-C	-8.0
NVCI
Between-group difference	PGI-S (1 grade)	-0.6	0.9-point change	0.9-point improvement
	PGI-S (2 grade)	-1.1
	PGI-C	-1.0
Within-group change	PGI-S (1 grade)	-0.8	1.0-point change	1.0-point improvement
	PGI-S (2 grade)	-1.3
	PGI-C	-0.9
Within-patient change (responder definition)*	PGI-S (1 grade)	-1.0	1.0-point change (ROC only) 1.0-point change (ROC and within-group change)	1.0-point improvement
NVS
Between-group difference	PGI-S (1 grade)	0.8	1.1-point change	1.1-point improvement
	PGI-S (2 grade)	1.4
	PGI-C	1.2
Within-group change	PGI-S (1 grade)	1.3	1.6-point change	1.6-point improvement
	PGI-S (2 grade)	2.0
	PGI-C	1.4
Within-patient change (responder definition)*	PGI-S (1 grade)	1.0	1.3-point change (ROC only) 1.4-point change (ROC and within-group change)	1.0-point improvement
	PGI-S (2 grade)	2.0
	PGI-C	1.0

*Only ROC estimates using Method 4 (minimising sum of squares) were included in the correlation-weighted estimate of meaningful within-patient change since this method is proven mathematically to be closest to the top-left corner of the ROC curve. Correlation-weighted average estimates of meaningful within-patient change (responder definition) were calculated using the ROC curve-based estimates and separately with both the ROC-based and within-group change estimates included.