Background
Surgical site complications (SSCs) contribute to increased healthcare costs. Predictive analytics can aid in identifying high-risk patients and implementing optimization strategies. This study aimed to develop and validate a risk-assessment score for SSC-associated readmissions (SSC-ARs) in patients undergoing open spine surgery.
Methods
The Premier Healthcare Database (PHD) of adult patients (n=157,664; 3,182 SSC-ARs) between January 2019 and September 2020 was used for retrospective data analysis to create an SSC risk score using mixed effects logistic regression modeling. Full and reduced models were developed using patient-, facility-, or procedure-related predictors. The full model used 37 predictors and the reduced used 19.
Results
The reduced model exhibited fair discriminatory capability (C-statistic =74.12%) and demonstrated better model fit [Pearson chi-square/degrees of freedom (DF) =0.93] compared to the full model (C-statistic =74.56%; Pearson chi-square/DF =0.92). The risk scoring system, based on the reduced model, comprised the following factors: female (1 point), blood disorder [2], congestive heart failure [2], dementia [3], chronic pulmonary disease [2], rheumatic disease [3], hypertension [2], obesity [2], severe comorbidity [2], nicotine dependence [1], liver disease [2], paraplegia and hemiplegia [3], peripheral vascular disease [2], renal disease [2], cancer [1], diabetes [2], revision surgery [2], operative hours ≥5 [4], emergency/urgent surgery [2]. A final risk score (sum of the points for each surgery; range, 0–40) was validated using a 1,000-surgery random hold-out sample (C-statistic =85.16%).
Conclusions
The resulting SSC-AR risk score, composed of readily obtainable clinical information, could serve as a robust predictive tool for unplanned readmissions related to wound complications in the preoperative setting of open spine surgery.