Patient Factors Predictive of Hospital Readmissions Within 30 Days

Area of Science:

• Health Services Research
• Health Economics
• Public Health Policy

Background:

• The Affordable Care Act mandates reduced Medicare payments for hospitals with excess readmissions.
• A novel readmission measure was developed using hospital discharge abstract data.
• This measure allows hospitals to monitor inpatient populations against national benchmarks.

Purpose of the Study:

• To develop and validate a dynamic database for tracking risk-adjusted hospital readmission rates.
• To identify patient-specific factors influencing 30-day readmission risk.

Main Methods:

• Analysis of deidentified discharge abstract data from approximately 15 million inpatient discharges across 611 diverse acute care hospitals.
• Application of multivariate logistic regressions to assess the impact of various patient factors on readmission odds.
• Data spanned a 2-year period (2008 Q4-2010 Q3), encompassing urban academic and rural community hospitals.

Main Results:

• Meaningful risk-adjusted readmission rates can be dynamically tracked.
• Index admission clinical conditions are the strongest predictors of readmission.
• Patient age, comorbid conditions, race, income, and payer status significantly influence readmission risk.

Conclusions:

• Stratified payment models can promote equity and transparency in healthcare disparities.
• Current models cannot differentiate between modifiable and non-modifiable readmissions.
• The optimal readmission rate for specific conditions remains undetermined.