Skip to contents

Corrected Covered Area (CCA)

CCA.Rmd

Introduction

The Corrected Covered Area (CCA) quantifies the overlap of studies included in multiple systematic reviews. A high CCA value indicates a high degree of overlap, suggesting potential redundancy in evidence synthesis.

This vignette explains:

  • The formula for CCA

  • How to compute CCA in R

  • How to visualize study overlap using a heatmap

CCA Formula

CCA is calculated as:

CCA=N−rr(c−1)×100 CCA = \frac{N - r}{r (c - 1)} \times 100

Where: - NN = total number of study occurrences across reviews - rr = number of unique included studies - cc = number of systematic reviews

Example Data

We use an example dataset with 5 studies across 5 systematic reviews:

library(tidyverse)
library(junoR)  

# Example dataset
data <- tibble(
  study = rep(paste0("Study_", 1:5), times = 5),
  review = rep(paste0("Review_", 1:5), each = 5),
  status = c("included", "included", "ineligible", NA, NA, 
             NA, "included", NA, "excluded", NA, 
             "included", NA, "included", "excluded", "ineligible", 
             "included", "excluded", "included", "excluded", NA, 
             NA, "included", "ineligible", NA, "excluded")
)

# View the data
print(data)
## # A tibble: 25 × 3
##    study   review   status    
##    <chr>   <chr>    <chr>     
##  1 Study_1 Review_1 included  
##  2 Study_2 Review_1 included  
##  3 Study_3 Review_1 ineligible
##  4 Study_4 Review_1 NA        
##  5 Study_5 Review_1 NA        
##  6 Study_1 Review_2 NA        
##  7 Study_2 Review_2 included  
##  8 Study_3 Review_2 NA        
##  9 Study_4 Review_2 excluded  
## 10 Study_5 Review_2 NA        
## # ℹ 15 more rows

Compute CCA

We use the cca_heatmap() function to compute CCA:

# Run the function
result <- cca_heatmap(data)

# Print CCA result
result$CCA
## [1] 41.66667

This means that 42% of the included studies are redundant.

Visualising Study Overlap

The function also generates a heatmap showing which studies are included in which reviews:

# Generate heatmap
result$plot

This produces a heatmap where:

Green = included studies

Dark grey = excluded studies

Orange = ineligible studies

Light grey = missing data

Interpretation of Results

CCA (%) Overlap Level Interpretation
0-20% Low Minimal study redundancy across reviews
21-50% Moderate Some overlap, but different studies are included
51-100% High High redundancy, reviews rely on similar studies
values greater than 100% Extreme Indicates a possible error in study categorisation

References

Pieper, D., Antoine, S. L., Mathes, T., Neugebauer, E. A., & Eikermann, M. (2014). Systematic review finds overlapping reviews were not mentioned in every other overview. Journal of Clinical Epidemiology, 67(4), 368–375. DOI: 10.1016/j.jclinepi.2013.11.007

Hedges, L. V., & Olkin, I. (1988). Statistical Methods for Meta-Analysis. Academic Press.