Burnout Syndrome Among ICU Nurses in Riyadh Hospitals
Cross-sectional study of 412 ICU nurses across 14 hospitals in the Riyadh region. Binary logistic regression and structural equation modeling identified key organizational predictors of burnout for a PhD dissertation at a major Riyadh university.
Project highlights
Quick summary
When you study a complex psychological or organizational phenomenon such as burnout and need measurement tool validation before modeling predictors and mediation.
We validated the measurement scale first, then modeled the drivers of burnout.
- Results chapter with APA 7th edition tables, model-fit indices, and path diagrams.
- Validated measurement model with full CFA/SEM output and reliability evidence.
Methods & Tools
Identifying the organizational drivers of nurse burnout demanded both rigorous measurement validity and defensible modeling across 14 hospital sites. The MBI-HSS instrument had to be psychometrically confirmed before any predictor model could be trusted.
Cleaned and screened 412 multi-site responses, assessed the missingness mechanism and assumptions, and established internal consistency and convergent/discriminant validity of the MBI-HSS subscales.
Confirmed the three-factor MBI-HSS structure with confirmatory factor analysis and structural equation modeling (CFI 0.96, RMSEA within thresholds) in R (lavaan) and AMOS.
Modeled burnout predictors with multivariable logistic regression and tested mediation pathways, reporting standardized effects with bias-corrected bootstrap confidence intervals.
Results chapter with APA 7th edition tables, model-fit indices, and path diagrams.
Validated measurement model with full CFA/SEM output and reliability evidence.
Prioritized, evidence-based recommendations for workforce and retention interventions.
Measurement validity (reliability, convergent/discriminant) established before any predictor modeling.
Model fit judged against pre-set thresholds (CFI, RMSEA, SRMR).
Mediation effects verified with bias-corrected bootstrap resampling.
Fully reproducible R Markdown workflow with documented assumptions.
This was my PhD on burnout among ICU nurses, and the viva was my biggest fear. They sat with me and explained every test and every p-value until I understood my own data. On the day of the defense I was calm, and the committee could see that I really knew my analysis.
Have a project to register?
Submit your project in the dashboard and we’ll map the analysis plan, deliverables, and timeline.