Analysis pipeline for modeling of illness perception one year after COVID-19
We investigate severity and quality of illness perception (IP) in a cross-sectional observational cohort of COVID-19 convalescents (CovILD study) suffering from persistent somatic symptoms or heart/lung abnormalities at one year after diagnosis. By regularized regression and unspervised clustering we could discover a strong association of persistent somatic symptoms (PSS) such as fatigue, physical performance loss or sleep problems with the total IP severity and its emotion/concern/consequences component. For details, please refer to our paper.
Please cite the repository and the peer-reviewed publication of the analysis results (DOI: 10.1016/j.jpsychores.2023.111234). The raw data files will be made upon request to the senior study author, Prof. Judith Löffler-Ragg.
The following development packages are required to run the pipeline:
devtools::install_github('PiotrTymoszuk/soucer') ## script sourcing
devtools::install_github('PiotrTymoszuk/ExDA') ## exploratory data analysis and staristical hypothesis testing
devtools::install_github('PiotrTymoszuk/clustTools') ## factor analysis and unsupervised clustering
devtools::install_github('PiotrTymoszuk/caretExtra') ## fit statistics and quality control for the Caret models
devtools::install_github('PiotrTymoszuk/lmqc') ## fit statistics and quality control for linear models
devtools::install_github('PiotrTymoszuk/figur') ## management of figures and tables in Rmd documents
devtools::install_github('PiotrTymoszuk/trafo') ## handling of tabular data
Source 'exec.R' to launch the entire pipeline:
source('exec.R')
The repository maintainer is Piotr Tymoszuk. Data requests should be addressed to Prof. Judith Löffler-Ragg.