impact: IMPACT data
In metamisc: Meta-Analysis of Diagnosis and Prognosis Research Studies

Description Usage Format Details Source References Examples

The IMPACT dataset comprises 15 studies of patients suffering from traumatic brain injury, including individual patient data from 11 randomized controlled trials and four observational studies.

1	data("impact")

A data frame with 11022 observations on the following 11 variables.

name: Name of the study
type: Type of study, RCT: randomized controlled trial,OBS: observational cohort
age: Age of the patient
motor_score: Glasgow Coma Scale motor score
pupil: Pupillary reactivity
ct: Marshall Computerized Tomography classification
hypox: Hypoxia (0=no, 1=yes)
hypots: Hypotension (0=no, 1=yes)
tsah: Traumatic subarachnoid hemorrhage (0=no, 1=yes)
edh: Epidural hematoma (0=no, 1=yes)
mort: 6-month mortality (0=alive, 1=dead)

The included studies were part of the IMPACT project, where a total of 25 prognostic factors were considered for prediction of 6-month mortality. Missing values were imputed using the study as a fixed effect in the imputation model (Steyerberg et al, 2008).

Steyerberg EW, Nieboer D, Debray TPA, Van Houwelingen JC. Assessment of heterogeneity in an individual participant data meta-analysis of prediction models: An overview and illustration. Stat Med. 2019;38(22):4290–309.

Murray GD, Butcher I, McHugh GS, et al. Multivariable prognostic analysis in traumatic brain injury: results from the IMPACT study. J Neurotrauma. 2007;24(2):329–337.

Steyerberg EW, Mushkudiani N, Perel P, et al. Predicting outcome after traumatic brain injury: development and international validation of prognostic scores based on admission characteristics. PLOS Med. 2008;5(8):e165.

data(impact)

by(impact, impact$name, summary)

# Plot the distribution of age by study
library(ggplot2)
e <- ggplot(impact, aes(x = name, y = age))
e + geom_violin(aes(fill = type), trim = FALSE)  + 
    theme(axis.text.x = element_text(angle = 45)) + 
    xlab("Study")