Does Performance Vary Within the Same Hospital When Separately Examining Different Patient Subgroups?

Area of Science:

• Healthcare Quality Improvement
• Surgical Outcomes Research
• Health Services Research

Background:

• Surgical quality programs typically aggregate patient data, potentially masking performance variations within hospitals.
• Current reporting methods may not identify specific patient subgroups where hospital performance differs.
• Opportunities for targeted quality improvement may be missed due to generalized reporting.

Purpose of the Study:

• To determine if hospital performance varies across six distinct patient subgroups.
• To identify the percentage of hospitals exhibiting performance differences exceeding chance expectations within subgroups.
• To inform quality programs about subgroup-specific performance variations.

Main Methods:

• Utilized data from the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) from 2005-2012.
• Analyzed adult patients undergoing major gastrointestinal and thoracic resections, divided into subgroups (e.g., elderly vs. nonelderly, cancer vs. noncancer).
• Employed hierarchical models to calculate observed-to-expected ratios and statistical tests to compare performance differences between subgroups within hospitals.

Main Results:

• Significant differences in overall quality were observed between elderly vs. nonelderly, renal insufficiency vs. normal renal function, cancer vs. noncancer, and emergency vs. nonemergency patient groups.
• Within-hospital performance variations exceeded chance expectations in a substantial proportion of hospitals: 31.1% for renal function, 40.8% for cancer status, and 55.4% for emergency status.
• These findings indicate significant heterogeneity in hospital performance based on patient characteristics.

Conclusions:

• Hospital performance is not uniform and varies considerably depending on the patient subgroup.
• Quality improvement initiatives should consider subgroup-specific analyses to identify and address performance disparities.
• Tailored reporting by patient subgroups can enhance the effectiveness of surgical quality programs.