Variation in Low-Value Service Use Across Veterans Affairs Facilities

Area of Science:

• Health Services Research
• Healthcare Quality Improvement
• Health Economics

Background:

• The extent of variation in low-value healthcare services within integrated systems like the Veterans Health Administration (VA) is not well understood.
• Understanding this variation is crucial for optimizing healthcare resource allocation and improving patient outcomes.

Purpose of the Study:

• To quantify the variation in the utilization of low-value services across different VA facilities.
• To investigate the relationship between facility characteristics and the use of these services.

Main Methods:

• Retrospective cross-sectional study using VA administrative data from 139 facilities and over 5 million patients.
• Facility-level rates of low-value service use were calculated based on 29 services across six domains.
• Ordinary least squares regression was used to adjust rates for patient characteristics and examine associations with facility attributes.

Main Results:

• A mean of 20.0 low-value services per 100 patients per year was observed, with significant variation across facilities (90th/10th percentile ratio of 2.0).
• After adjusting for patient factors, facility variation narrowed, but remained substantial (90th/10th percentile ratio of 1.8).
• The percentage of patients using VA Community Care was the only facility characteristic associated with increased low-value service use; no other characteristics were significantly associated with total low-value service use.

Conclusions:

• Significant variation in low-value service use exists across VA facilities, alongside substantial overall utilization.
• These findings suggest that system-wide interventions targeting low-value services may be more effective than facility-specific approaches.
• Further research into the drivers of variation and the impact of community care utilization is warranted.