Optimising assessment of dark adaptation data using time to event analysis

Abstract

In age-related macular degeneration (AMD) research, dark adaptation has been found to be a promising functional measurement. In more severe cases of AMD, dark adaptation cannot always be recorded within a maximum allowed time for the test (~ 20–30 min). These data are recorded either as censored data-points (data capped at the maximum test time) or as an estimated recovery time based on the trend observed from the data recorded within the maximum recording time. Therefore, dark adaptation data can have unusual attributes that may not be handled by standard statistical techniques. Here we show time-to-event analysis is a more powerful method for analysis of rod-intercept time data in measuring dark adaptation. For example, at 80% power (at α = 0.05) sample sizes were estimated to be 20 and 61 with uncapped (uncensored) and capped (censored) data using a standard t-test; these values improved to 12 and 38 when using the proposed time-to-event analysis. Our method can accommodate both skewed data and censored data points and offers the advantage of significantly reducing sample sizes when planning studies where this functional test is an outcome measure. The latter is important because designing trials and studies more efficiently equates to newer treatments likely being examined more efficiently.

Assessment of the Classification of Age-Related Macular Degeneration Severity from the Northern Ireland Sensory Ageing Study Using a Measure of Dark Adaptation

“…In fact, the authors explicitly state that they assigned a value of 20 minutes to all the observations not recovering within the maximum allocated time. In our previous paper 36 , we have shown this to introduce important distortions in the estimates of RIT. Our approach retains all the advantages of allowing for correction of covariates while addressing fundamental properties of the specific nature of the data.…”

“…Our primary analysis focused on how average RMDA (the RIT parameter) differed between groupings assigned using the different classification techniques. Here we specifically used a time-to-event analysis as described in our previous work 36 .…”

“…In short, the time-to-event analysis can be conveniently used to model the time taken to recover from bleaching while accounting for predictors of interest, such as differences between groups, and correcting for covariates such as age or other attributes 36 . Kaplan-Meier curves were plotted to visually represent the comparisons of the results from the models.…”

“…We also utilised a time-to-event model (sometimes referred to as a survival model) to assess the magnitude of measurement differences in RMDA. This model is a statistically correct method for this data and offers advantages over alternative methods like ttests and non-parametric tests, described previously 36 . While we were not the first to identify age as a possible confounder in RIT analyses, another strength of our methodology was that we compensated for age-effects.…”

Functional evaluation in inherited retinal disease

“…78 Higgins et al have recently proposed a novel 'time-to-event' analysis method that can be applied to this data, providing better statistical power. 79…”