Reference Intervals
RET
eval
Device User Manual 70
Some subjects (n=118) were tested after artificial dilation, while others were tested with natural
pupils and constant Troland stimuli that compensate for pupil size (n=233). Dilated subjects who
did not dilate to at least 6 mm were excluded from tests that did not compensate for pupil size.
Far outliers (defined as 3 interquartile ranges away from the 25th and 75th percentiles) were
removed after age correction that used a robust (bisquare) linear least squares fit. Replicates
were averaged. Percentiles were computed from their rank (Schoonjans, De Bacquer, and
Schmid 2011). No underlying distribution was assumed. A bootstrap method was used to
compute the 90% confidence intervals of the 5% and 95% reference limits.
The tables below show the 5% and 95% reference limits, along with their 90% confidence
intervals (CI). In addition, the median (50%) value in the reference data is shown. The data has
been age adjusted to 0 years of age. The slope of the age correction factor is also shown. To
convert the reference limits in the table below to a particular age, multiply the patient’s age (in
years) at the time of testing by the slope, and then add that value to reference limits.
DR Score. Flash: 4, 16, and 32 Td·s white, Background: 0 Td white