Assessing Nursing Homes Quality Indicators' Between-Provider Variability and Reliability: A Cross-Sectional Study...

Area of Science:

• Gerontology
• Health Services Research
• Quality Improvement

Background:

• Public reporting of nursing home quality relies on indicators to reflect care levels.
• Switzerland developed six national quality indicators across polypharmacy, pain, physical restraints, and weight loss.
• Assessing between-provider variability and reliability is crucial for meaningful quality comparisons.

Purpose of the Study:

• To evaluate the between-provider variability and reliability of six Swiss national nursing home quality indicators.
• To utilize intraclass correlations (ICC) and rankability for assessing indicator performance.
• To determine the suitability of these indicators for public reporting and quality comparisons.

Main Methods:

• Cross-sectional, multicenter study involving 11,412 residents from 152 Swiss nursing homes.
• Calculated Intraclass Correlation 1 (ICC1) to measure between-provider variability.
• Assessed indicator rankability and reliability using Intraclass Correlation 2 (ICC2).
• Employed empirical Bayes estimate-based caterpillar plots to visualize variability.

Main Results:

• High ICC1 values indicated substantial between-provider variability for most indicators, particularly physical restraints (0.396).
• Rankability results were mixed; polypharmacy (0.144) and self-reported pain (0.471) showed lower rankability.
• All six indicators demonstrated good reliability with high ICC2 values, ranging from 0.896 to 0.990.

Conclusions:

• Intraclass correlations and rankability are effective methods for assessing nursing home quality indicators.
• The six selected indicators reliably differentiate care quality between nursing homes.
• Polypharmacy and self-reported pain indicators require careful interpretation due to chance-driven variability, potentially limiting their comparative utility.