PurposeClinicians and stewardship programs are challenged with positioning of novel, higher priced antibiotic agents for the treatment of clinical infections. We developed a decision-analytic model to describe costs, including drug, total treatment costs, and health care outcomes, associated with telavancin (TLV) compared with vancomycin (VAN) for patients with Staphylococcus aureus (SA) hospital-acquired bacterial pneumonia (HABP). MethodsThis decision-analytic model assessed the treatment of SA-HABP with TLV versus VAN. Data were obtained from the ATTAIN (Assessment of Telavancin for Treatment of Hospital-Acquired Pneumonia) clinical trials on the following: the probability of clinical cure; probability of nephrotoxicity; and prevalence of polymicrobial infection (30%), methicillin-resistant Staphylococcus aureus (MRSA) (68%), and SA with VAN MIC ≥1 µg/mL (85%). Data on length of stay for cure (10 days), failure (10 additional days), and nephrotoxicity (3.5 days) were based on literature. Cost per treated patient and incremental cost-effectiveness ratio (ICER) per additional cure were calculated for SA-HABP and for monomicrobial SA-HABP. One-way sensitivity analyses were performed. FindingsPatients with SA-HABP were sub-grouped by methicillin susceptibility (n = 140, 32%) or resistance (n = 293, 68%), and occurrence of polymicrobial (n = 128, 30%) vs monomicrobial (n = 305, 70%) infections. Under the base case, hospital cost for patients with HABP treated with TLV was $42,564 and with VAN, it was $42,296. Telavancin was associated with higher drug ($2082) and nephrotoxicity ($467) costs and lower intensive care unit (−$1738) and ventilator (−$114) costs. ICER was $4156 per additional cure. ICER was sensitive to probabilities of cure, length of treatment in cures, intensive care unit cost, TLV cost, and additional length of stay due to failure. For monomicrobial SA-HABP, TLV was associated with a net cost savings of $907 per patient and yielded economic dominance. ImplicationsOur decision-analytic model suggests that TLV for monomicrobial SA-HABP is associated with higher drug acquisition costs but a favorable ICER relative to VAN, provided that effective antimicrobial stewardship limits therapy to 7 days. Sensitivity analyses suggest a potential economic benefit of TLV treatment with appropriate patient selection. Antimicrobial stewardship programs may be able to reduce total costs through judicious use of novel antimicrobial agents. ClinicalTrials.gov identifiers: NCT00107952 and NCT00124020.