Introduction: The Center for Medicare and Medicaid Services recently declared that central venous catheter-associated bloodstream infections (CLABs) are preventable and no longer reimbursable. The new penalty paradigm creates substantial economic incentives for hospitals to eliminate infections. Modeling exercises offer the opportunity to justify expenditures for the prevention of rare patient safety events.
Methods: We constructed a decision analytic model of the theoretical impact of an educational intervention to improve the safety of central venous catheter insertion. This methodology allows for decision nodes representing procedure urgency, procedure locations, and effects of education on both infectious and mechanical complications. We performed deterministic sensitivity analyses to estimate effects of changes in complication rates, educational impact, and cost.
Results: In an already safety system (CLAB rate of 1:1000 line days or 0.7%), the initial sensitivity results suggest that if education results in a 50% reduction in CLAB and a 25% reduction in mechanical complications, survival is equal, and cost is increased by $92,400 in a large hospital system annually. If all CLAB and mechanical complications are eliminated, survival improves slightly, and cost is reduced by $64,800 annually.
Conclusions: These results suggest that if the educational intervention is effective, a small increase in cost can reduce complications. Our analysis does not consider increased revenue generated by virtual bed capacity increases or dynamic changes in practice. This model serves as a template for other health care institutions to estimate the costs and benefits of their own proposed educational interventions.
From the *Department of Emergency Medicine and Division of Education, Lehigh Valley Health Network, Allentown, Pennsylvania; and the †Department of Economics, Lehigh University, Bethlehem, Pennsylvania.
Correspondence: William F. Bond, MD, MS, Lehigh Valley Health Network, Division of Education, Suite 202, 1247 S Cedar Crest Blvd, Allentown, PA 18103 (e-mail: firstname.lastname@example.org).
Sources of Support: This study has no financial support.