Using Present-On-Admission Coding to Improve Exclusion Rules for Quality Metrics

Entrance to a children's Hospital.

Beginning in June 2010, the Centers for Medicare and Medicaid Services began collecting data on hospital failure-to-rescue rates.

The Issue:
As part of their ongoing examination on quality of care and performance, failure-to-rescue measured death among patients with treatable complications. However, there is concern that this type of administrative data is not efficient enough to differentiate between a hospital-acquired complication or a present-on-admission (POA) comorbidity.

Key Findings

  • Using a POA indicator along with three rule-based exclusion rules, the authors demonstrated a substantial improvement in the failure-to-rescue measure over the standard Agency for Healthcare Research and Quality (AHRQ) failure-to-rescue measure which excludes complications presumed to be POA.

  • A higher percentage of registered nurses and higher levels of staffing were associated with a lower risk of death in the study sample.

The accuracy of failure-to-rescue exclusion rules can be improved using data with good POA coding to identify POA comorbidities compared with the current AHRQ rule. Of note, modifying the rules is unlikely to compromise the nursing-sensitive nature of the measure.

About the Study:
The final validation dataset for this study included 82,871 discharges from California hospitals (from 2000–2002) with good POA coding practices. The three new alternative exclusion rules were developed from POA-coded data from a large academic medical center.