Download Full Text (802 KB)



Training Level

Resident PGY 4


Henry Ford Hospital


Background: Readmissions are an increasingly important focus for improvement regarding quality, value and patient burden in our surgical patient population. Predictive strategies to-date have focused on patient demographics and comorbidities with some hospital data added to the mix. We hypothesized that inpatient harm events would increase the likelihood of readmission in surgical patients.

Methods: We created a system-wide inpatient registry within the Henry Ford Health System, Detroit, MI, with 30-day readmission and harm events labeled. The surgical subset was created using a DRG table utilized for this work. The study was approved by our IRB.

Results: Between 2015-2017, 37,566 surgical patients met inclusion criterion. A total of 2,887 patients (7.69%) were readmitted. Fifteen out of 29 harm events were statistically significant (P3.5) on harm events were: Mucosal Pressure Ulcer, C. Difficile Toxin, catheter-associated urinary tract infection (CAUTI), and blood clot international normalized ratio (INR) >=5. By multiple logistic regression of the significant harm events, we found 12 harm events remained statistically significant (P400 having the highest odds ratios (>1.9). CAUTI, INR>=5, and Deep Tissue Injury (DTI) Pressure were no longer significant.

Discussion:This study demonstrated that inpatient hospital-based harm events were potential predictors of 30-day readmission. The multiple harm events identified independently associated with readmission suggesting that patients with these harm events would have a higher risk of readmission. Further work by including other metrics is warranted.

Presentation Date


The Efects of Harm Events on the Rate of 30-day Readmissions in Surgical Patients