Data Synthesis and Analysis
For each study outcome that was reported by more than one study, the
results from individual studies were combined statistically using the
random-effects meta-analysis model, stratified by the level of exposure
(ACEIs/ARBs, ACEIs, ARBs); whereas for outcomes which were reported by
only one study, narrative synthesis was used. For studies which did not
report the summary statistics and measure of effects, we firstly used
the reported primary statistics (number of patients with/without the
outcomes in both exposed/unexposed group) to calculate the corresponding
measure of effects (Odds ratios) and their 95% confidence interval
(35), and subsequently used these measure of effects in the
random-effects meta-analysis. Several sub-group analyses were also
undertaken to explore the effect of potential confounders on the
robustness and sensitivity of combined pooled estimates and included
sub-group analyses based on whether the reported measure of effects was
crude or adjusted, the study was peer-reviewed or not, the study’s
methodological quality as per the risk of bias assessment was performed
as well as the continent where the study was conducted. Meta-analyses
pooled estimated were presented as odds ratios and 95%CI and
graphically as forest plots. Heterogeneity between the studies was
evaluated using I2 statistic (36), indicating whether
variability is more likely due to study heterogeneity or chance.
Negative I2 values were set to zero, hence
I2 values ranged between 0%-100% with 0% indicating
lack of heterogeneity, whereas 25%, 50%, and 75% indicating low,
moderate and high heterogeneity, respectively (36). Data were analysed
using STATA 12.
Role of the Funding
Source
None