OBJECTIVES: An increase in prevalence of antimicrobial resistance among gram-negative pathogens in complicated intra-abdominal infections (cIAI) has been noted recently (Sartelli et al 2013). A challenge in empiric treatment of cIAI is identifying initial appropriate antibiotic therapy (IAAT), which is associated with reduced length of stay and mortality compared with inappropriate therapy. Using local antibiogram data to select empiric therapy may increase the likelihood of IAAT. The objective of this research is to demonstrate how an economic model can assist medical decision-making by identifying threshold differences in antibiotic resistance at which ceftolozane/tazobactam+metronidazole is cost-effective/cost-saving compared with piperacillin/tazobactam as IAAT. METHODS: We used a decision analytic Monte Carlo simulation model (Kauf et al 2015) to compare cost and QALYs of persons infected with nosocomial gram-negative cIAI and treated empirically with either ceftolozane/tazobactam+metronidazole or piperacillin/tazobactam. We ran the model for baseline prevalence of resistance in the Program to Assess Ceftolozane/Tazobactam Susceptibility (PACTS) in-vitro surveillance database. We limited threshold analysis to three commonly isolated gram-negative pathogens: Escherichia coli, Klebsiella pneumoniae, and Pseudomonas aeruginosaRESULTS: At baseline resistance levels from PACTS, ceftolozane/tazobactam+metronidazole was cost-saving compared with piperacillin/tazobactam. Ceftolozane/tazobactam was cost-effective (cost-saving) when resistance rates for the three bacteria in the piperacillin/tazobactam were > 2% (7%) relative to ceftolozane/tazobactam. When resistance rates for E. coli alone were changed, ceftolozane/tazobactam was cost-effective (cost-saving) at differences > 3% (13%). Results were sensitive to gram-positive/-negative pathogen prevalence, drug/hospitalization cost.  CONCLUSIONS: Economic models can be used to identify IAAT based on in-vitro resistance data. Once threshold differences in antibiotic resistance that make a comparator drug cost-effective are established, local antibiograms can help identify optimal IAAT.