Raw IOP measurements in the right (OD) and left (OS) eye over the last few days
Smoothed IOP curves for both eyes
Right minus left eye IOP difference
19.11 zamieniono oczami (1) NodoFreeCombi i (2) TimoComod + Azopt
2.12 OS bez zmian NodoFree Combi, w OD zmiana na Vizitrav + NodoFree
Combi
9.12 OD bez zmian, w OS włączono Vizitrav
Treatment crossed over between eyes on Nov. 19
Dec 2nd OD stays on NodoFree Combi, OS switched to Vizitrav + NodoFree
Combi
Dec 9th Vizitrav added in OS
Polynomial model with hour of the day, treatment and eye as predictors
Poziom odniesienia to oko prawe (OD), NodoFree Combi. Krople działają multiplikatywnie raczej niż addytywnie, stąd log(IOP) jako zmienna objaśniana.
Model parameters with right eye and NodoFree Combi as reference levels. IOP-lowering drops work in a multiplicative rather than additive fashion, thus we predict log(IOP).
##
## Call:
## lm(formula = log(IOP) ~ Eye + Treatment + poly(Hour, 5), data = IOP[IOP$Date >
## from, ])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.56109 -0.14674 0.00623 0.14845 0.47530
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.82220 0.02018 139.865 < 2e-16 ***
## EyeOS -0.11618 0.01813 -6.408 2.77e-10 ***
## TreatmentTimo+Azopt 0.03654 0.02364 1.546 0.1227
## TreatmentTravo+NodoCombi -0.27120 0.02014 -13.466 < 2e-16 ***
## poly(Hour, 5)1 -1.19215 0.20904 -5.703 1.76e-08 ***
## poly(Hour, 5)2 0.17975 0.20873 0.861 0.3895
## poly(Hour, 5)3 -0.49078 0.20908 -2.347 0.0192 *
## poly(Hour, 5)4 -2.60410 0.20901 -12.459 < 2e-16 ***
## poly(Hour, 5)5 0.94617 0.20876 4.532 6.90e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2087 on 677 degrees of freedom
## Multiple R-squared: 0.4551, Adjusted R-squared: 0.4487
## F-statistic: 70.68 on 8 and 677 DF, p-value: < 2.2e-16
Współczynniki powyższe są wykładnikami logarytmu naturalnego: Po ich
“odlogarytmowaniu” dostajemy poniższe współczynniki. Mówią one, że
ciśnienie śródgałkowe dla wartości odniesienia (= oko prawe, NodoFree
Combi) wynosi 16.8 mm Hg, w oku lewym ciśnienie jest niższe
o -11%, natomiast dodanie travoprostu (Vizitrav) do
NodoFree Combi daje dodatkową redukcję IOP o -24%. Wpływ
pory doby zilustrowano na wykresie efektu godziny poniżej.
Estimates are natural logs, after exponentiating we obtain
coefficients as below. The model thus predicts an IOP of
16.8 mm Hg in the right eye using NodoFree Combi, left eye
pressure being lower by -11%, whereas adding travoprost
drops (Vizitrav) to NodoFree Combi results in an additional IOP
reduction of -24%. For hour-of-the-day effect, see the plot
below.
## IOP
## (Intercept) 16.8
## EyeOS -11%
## TreatmentTimo+Azopt 4%
## TreatmentTravo+NodoCombi -24%
Effect plots
Daily maximum, minimum, and mid-range
Models for daily values
Poziom odniesienia to oko prawe (OD), NodoFree Combi.
Reference levels are for right eye and NodoFree Combi.
Daily mid-range values (between max and min)
##
## Call:
## lm(formula = log(IOP.day.mid) ~ Eye + Treatment, data = daily.s)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.33599 -0.05368 0.01954 0.05552 0.18653
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.85799 0.02269 125.945 < 2e-16 ***
## EyeOS -0.07965 0.02130 -3.740 0.00035 ***
## TreatmentTimo+Azopt 0.01285 0.02651 0.485 0.62928
## TreatmentTravo+NodoCombi -0.23936 0.02456 -9.747 3.87e-15 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.09017 on 78 degrees of freedom
## Multiple R-squared: 0.6397, Adjusted R-squared: 0.6258
## F-statistic: 46.15 on 3 and 78 DF, p-value: < 2.2e-16
## IOP
## (Intercept) 17.4
## EyeOS -8%
## TreatmentTimo+Azopt 1%
## TreatmentTravo+NodoCombi -21%
Daily minimum
##
## Call:
## lm(formula = log(IOP.day.min) ~ Eye + Treatment, data = daily.s)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.273517 -0.102376 -0.008518 0.096842 0.309184
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.46999 0.03216 76.798 < 2e-16 ***
## EyeOS -0.11703 0.03019 -3.877 0.000219 ***
## TreatmentTimo+Azopt 0.05200 0.03757 1.384 0.170229
## TreatmentTravo+NodoCombi -0.26425 0.03480 -7.593 5.77e-11 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1278 on 78 degrees of freedom
## Multiple R-squared: 0.5663, Adjusted R-squared: 0.5496
## F-statistic: 33.94 on 3 and 78 DF, p-value: 3.846e-14
## IOP
## (Intercept) 11.8
## EyeOS -11%
## TreatmentTimo+Azopt 5%
## TreatmentTravo+NodoCombi -23%
Daily maximum
##
## Call:
## lm(formula = log(IOP.day.max) ~ Eye + Treatment, data = daily.s)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.43309 -0.07145 0.01890 0.06335 0.19859
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.133618 0.027573 113.650 < 2e-16 ***
## EyeOS -0.061472 0.025878 -2.375 0.020 *
## TreatmentTimo+Azopt -0.007034 0.032206 -0.218 0.828
## TreatmentTravo+NodoCombi -0.228111 0.029837 -7.645 4.57e-11 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1096 on 78 degrees of freedom
## Multiple R-squared: 0.4977, Adjusted R-squared: 0.4784
## F-statistic: 25.76 on 3 and 78 DF, p-value: 1.108e-11
## IOP
## (Intercept) 23
## EyeOS -6%
## TreatmentTimo+Azopt -1%
## TreatmentTravo+NodoCombi -20%
IOP drop and return to baseline following 10-minute moderate cycling