Skip to main content

Advertisement

Table 2 Bivariate quantitative analyses of all investigated parameters in dependence of year of publication. All given data are calculated from the equations of the regression lines. For P-values see table 3

From: The fading of reported effectiveness. A meta-analysis of randomised controlled trials

   Pravastatin Atorvastatin Timolol Latanoprost
Reported effect size* Change in 5 years -3.22 +0.31 -0.56 -1.78
  95% CI limits (-4.50/-1.93) (-3.29/+3.91) (-0.79/-0.34) (-3.04/-0.51)
Baseline Change in 5 years -41.80 -14.63 -0.70 -1.82
  95% CI limits (-55.74/-27.86) (-48.38/+19.11) (-1.08/-0.32) (-3.69/+0.05)
Study size Change in 5 years +533.54 +233.63 +80.55 -16.94
  95% CI limits (-3.94/+1071.01) (-429.97/+897.23) (+43.38/+117.71) (-199.23/+165.35)
Treatment group§ Change in 5 years -0.20 -0.28 -0.12 -0.23
  95% CI limits (-0.31/-0.08) (-0.58/+0.01) (-0.17/-0.06) (-0.50/+0.03)
  1. Abbreviations: CI, confidence interval; LDL-C, low-density lipoprotein cholesterol.
  2. * Unit of measurement for reported effect size: Change of intraocular pressure measured in mmHg (Timolol, Latanoprost), change in low-density lipoprotein cholesterol measured in % (Pravastatin, Atorvastatin).
  3. Unit of measurement for baseline: Intraocular pressure measured in mmHg (Timolol, Latanoprost), low-density lipoprotein cholesterol measured in mg/dl (Pravastatin, Atorvastatin). To convert low-density lipoprotein cholesterol from milligrams per deciliter to millimoles per liter, multiply milligrams per deciliter by 0.0259.
  4. Unit of measurement for study size: Number of patients included in final analysis.
  5. § The parameter treatment group has two possibilities: control group = 0, experimental group = 1. Point biserial correlation was used to obtain the equation of the regression line and to calculate the given data.