Skip to main content
. Author manuscript; available in PMC: 2017 Mar 1.
Published in final edited form as: Res Social Adm Pharm. 2015 Jun 15;12(2):218–246. doi: 10.1016/j.sapharm.2015.06.001

Table 3.

Random effects medication adherence estimates and tests

k Effect size p (ES) 95% Confidence interval Standard error I2 Q p(Q)
Treatment vs. control comparisonsa 40 0.301 <.001 0.186, 0.415 0.058 68.472 123.701 <.001
Treatment subjects pre-post comparisonsb 36 0.533 <.001 0.404, 0.663 0.066 70.189 117.403 <.001
Control subjects pre-post comparisons 24 0.011 .861 −0.115, 0.138 0.064 59.535   56.840 <.001

k denotes number of comparisons. Q is a conventional heterogeneity statistic. I2 is the percentage of total variation among studies’ observed effect size due to heterogeneity.

a

One comparison excluded as an outlier. The overall effect size with inclusion of outlier was 0.423 (SE = 0.079, 95% CI: 0.277, 0.587).

b

Two comparisons excluded as outliers. The overall effect size with inclusion of outliers was 0.618 (SE = 0.084, 95% CI: 0.454, 0.783).