GUS SLOTMAN, MD, FCCM
Director of Clinical Research
Department of Surgery, Inspira Health Network
VINELAND, New Jersey, United States
Disclosure information not submitted.
Title: Interaction of Pre-Discharge Laboratory Tests with 30-Day Re-Admission for Pneumonia and COPD
Introduction: CMS penalizes hospitals for 30-day readmission (RA30) after discharge for pneumonia and COPD. RA30 prediction specific to pneumonia and COPD is not standardized. While physician judgment is not prognostic of RA30, whether clinical practice differentiates RA30 from noRA30 is unknown.
Hypothesis: Physician ordering during COPD and pneumonia hospitalizations may reflect assessment of illness acuity and vary RA30/noRA30.
Methods: De-identified data from 3350 hospitalized pneumonia patients and 2294 with COPD were studied by %RA30 in 2 groups: routine lab tests ordered within the first 48 hours of admission and/or at 96- and 48-hours pre-hospital discharge (PD) (Non-Missing), and tests not ordered at those intervals (Missing). Lab tests included blood glucose, BMP (serum BUN, creatinine, sodium, chloride, potassium, CO2, anion gap), osmolality, GFR calculated, calcium), and CBC (WBC, hemoglobin, % neutrophils). Statistics: Chi-squared equation.
Results: Among COPD patients, RA30 did not vary with Missing and Non-Missing routine hospital labs ordered within the first 48 hours of admission and/or 96hrs and 48hrs PD (p >0.05). Pneumonia Missing/Non-Missing RA30: 1st48hrs – BMP 2.11%/7.22% (p=0.055); WBC and Hb 1.56%/7.18% (p=0.083); %PMN 6.06%/7.14% (p=0.566); 96hrs PD – glucose 2.38%/7.39% (p=0.006); BMP 3.92%/7.33% (p=0.041); Hb 2.62%/7.34% (p=0.013); %PMN 8.39%/6.45% (p=0.04); 48hrs PD – BUN 5.36%/7.75% (p=0.015); Hb 6.14%/7.41% (p=0.206); %PMN 7.61%/6.25% (p=0.133); glucose 4.35%/7.81% (p=0.001)
Conclusions: In hospitalized COPD patients, RA30 does not vary by Missing versus Non-Missing physician ordering of routine hospital laboratory tests, possibly reflecting the more chronic and recurrent nature of COPD exacerbations. However, among hospitalized pneumonia patients in the Missing group RA30 was significantly reduced compared with the Non-Missing group, for whom physicians ordered BMP, CBC, and/or glucose at 96- and 48-hrs PD. Such interaction between RA30 and physician ordering during the last four days of pneumonia hospitalizations suggests that bedside assessment of illness acuity in this more episodic illness may rule out the perceived need for lab confirmation of clinical judgement as discharge approaches. Nevertheless, translating these findings into clinical practice is not clear from the data.