Utilization of network graph connectivity to evaluate complications of inpatient medical and surgical populations as a method to prioritize quality improvement efforts

Abstract

Network graphs can provide a quantitative framework for identifying complications with significant volumes and strong relationships to other complications, as a method for prioritization of quality improvement work. Here we examine the application of network graphing techniques to acute care inpatient complications on acute care medical-surgical units of a quaternary care center. The 3M PPC software identified 66 complications among 106 unique patients with two or more complications during an inpatient hospital stay. The network graph highlighted renal failure without dialysis and septicemia and severe infections as highly connected complications in this population.