The association between U.S. Poison Center assistance and length of stay and hospital charges

Abstract Context. Poison centers (PCs) play an important role in poison prevention and treatment. Studies show that PCs reduce system-wide cost by reducing the number of unnecessary visits to emergency departments and by providing improved patient management. However, there remains a debate regarding the impact of PCs on patient outcomes at the hospital level. Objective. To evaluate the impact of PC involvement on length of hospitalization and total hospital charges. Materials and methods. We conducted a retrospective analysis of inpatient cases treated in Illinois hospitals in 2010. We linked the Illinois Poison Center database with an Illinois hospital billing dataset and controlled for important patient-level and facility-level covariates. Results. In the multivariable model, length of hospitalization among PC-assisted patients was 0.58 days shorter than that of patients without PC assistance (p < 0.001). Hospital charges for PC-assisted patients in the lower quintiles were significantly higher than patients without PC assistance (+$953; p < 0.001), but were substantially lower in the most costly quintile of patients (−$4852; p < 0.001). Balancing the higher charges for treating patients with PC assistance in the lower quintiles with the savings in the highest quintile, among inpatients there is a potential cumulative decrease of $2,078 in hospital charges per 10 patients. Discussion. Among the inpatient cases, PC assistance was associated with lower total charges only among the most expensive to treat. However, this outlier group is very important when discussing medical costs. It has been repeatedly shown that the majority of treatment costs are attributable to a small fraction of patients as seen in this study.