Methods: We used a mathematical model of C. difficile transmission in an ICU to estimate the effects of a co-occurring antibiotic and PPI stewardship program. This approach captured any synergistic dynamics between the two interventions (e.g. patients taking both PPIs and antibiotics) while being able to independently estimate their effects. This model simulated for five years and 5000 iterations, with the reduction in antibiotic and PPI use independently varied independently between 0% and 40%. The rates of C. difficile were then estimated using Poisson regression models accounting for admission volume.
Results: Both antibiotic and PPI stewardship reduced the number of incident C. difficile cases within the simulated ICU. A 30% decrease in fluoroquinolone use corresponded with a 21.9% decrease in incident C. difficile cases (p < 0.001), while a 30% decrease in PPI use corresponded with a 9.1% reduction (p < 0.001) in incident cases. There was no evidence of a synergistic effect between the two interventions (p=0.60). PPI stewardship also decreased length of stay, resulting in a 7% increase in admissions in the simulated ICUs (p < 0.001).
Conclusion: PPI stewardship might prove a valuable adjunct to existing antibiotic stewardship programs. The reductions in C. difficile transmission were more modest for PPI stewardship as compared to programs targeting fluoroquinolones. PPI stewardship, however, may reach different patient populations, and may represent an additional area for substantial improvement even in facilities that have made substantial gains in reducing fluoroquinolone use.