Figure 2 displays a forest plot of 11 studies comparing asthma patients using long-acting beta agonist (LABA)/inhaled corticosteroid (ICS) vs. ICS only. Mean difference of pulmonary function in L/min is presented for each study as well as the mean, standard deviation, sample size and year of publication for each intervention group from each study. The studies labeled Strand and SAM40036 are missing their SD and the missing values of SD are imputed using direct substitution. The random effects combined mean difference is 18.59 (95% CI 15.51 to 21.68) indicating the LABA/ICS group has higher values for pulmonary function. There was no statistically significant heterogeneity across the 11 studies. The purpose of this plot is to show the results of meta-analysis when the missing SDs are imputed using the direct substitution method. The two studies with imputed SDs are indicated in the box.

Figure 2Results of meta-analysis of pulmonary function, imputing missed data using direct substitution.*

From: Handling Continuous Outcomes in Quantitative Synthesis

Cover of Methods Guide for Effectiveness and Comparative Effectiveness Reviews
Methods Guide for Effectiveness and Comparative Effectiveness Reviews [Internet].

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.