All projects
BI

Patient Satisfaction Score Analysis

Patient satisfaction scores drive quality ratings and reimbursement decisions. This project analysed HCAHPS survey data to identify the strongest drivers of overall satisfaction.

3 key satisfaction drivers identified
2025
PythonPandasScikit-learnPower BI

Dataset

CMS: HCAHPS Patient Experience

View source

Key Questions

  • Which survey dimensions most strongly correlate with overall hospital rating?
  • Are there regional or hospital-size patterns in satisfaction scores?
  • Can we predict a hospital's overall rating from its individual dimension scores?

Methods

  • Correlation and regression analysis across HCAHPS dimensions
  • Principal component analysis to reduce survey dimensionality
  • Random forest feature importance for rating prediction
  • Benchmarking analysis against national and regional averages

Results

The three strongest predictors of overall satisfaction were: nurse communication (r=0.82), responsiveness of hospital staff (r=0.78), and care transition communication (r=0.75). Smaller hospitals (<100 beds) scored significantly higher on communication dimensions.

Satisfaction Score Correlation with Overall Rating

Recommendations

  • Prioritise nurse communication training as the highest-impact quality improvement initiative
  • Implement structured handoff protocols to improve care transition communication
  • Set up real-time patient feedback mechanisms rather than relying solely on post-discharge surveys
  • Benchmark against peer hospitals of similar size for more meaningful comparisons

Limitations

HCAHPS data represents respondents only (typically 25-30% response rate), which may introduce selection bias. Scores reflect patient perception, which may not always align with clinical quality measures.