Respiratory Depression on General Care Floors Increases Cost of Care: Results from the Prodigy Trial

Document Type


Publication Date



BACKGROUND: Opioid induced respiratory depression (OIRD) causes increased morbidity and mortality.1 We hypothesized that OIRD also increases cost of care, possibly via increased length of stay and treatment of associated complications. The PRediction of Opioid-induced respiratory Depression In patients monitored by capnoGraphY (PRODIGY) trial (NCT02811302) was performed to derive a risk prediction tool identifying patients at risk of OIRD, identified using continuous capnography and pulse oximetry monitoring. A prespecified secondary endpoint of PRODIGY was to examine the cost associated with OIRD. METHODS: Following the conclusion of the PRODIGY trial, individual patients’ hospital cost data was collected for 964 patients (64% of total enrolled), including 866 patients who started parenteral opioids and continuous capnography and pulse oximetry monitoring (Capnostream™ respiratory monitor, Medtronic) on the general care floor. Cost data included study sites in the United States (US), Japan, Singapore, Netherlands, and France. Cost comparison between patients with and without ≥1 OIRD episode was performed for each country. To examine the impact of OIRD on overall hospital cost per patient, a propensity weighted generalized linear model was used, considering all significant demographic and clinical variables. Due to the limited sample size for France, a multi-variable generalized linear model, controlling for significantly different covariates in univariate analysis, was generated. For the US market, the interaction between length of stay and OIRD on overall hospital cost was examined. RESULTS: The average hospital cost for 148 US patients with ≥1 OIRD episode was significantly higher compared to 272 US patients without OIRD episodes ($23,619 ± $16,868 vs $19,173 ± $13,549, p=0.0001). The mean cost for US patients at high risk for OIRD episodes, determined using the PRODIGY risk prediction tool, remained significantly higher for patients that experienced ≥1 OIRD episode during continuous monitoring ($25,057 ± $19,490 vs $18,609 ± $9,714 for patients without OIRD episodes, p=0.0087). Propensity weighting of patients with or without ≥1 OIRD episode found that the cost for patients who experience ≥1 OIRD episodes is 15% higher in the United States (exponentiated estimate 1.15, p=0.0013) and 18% higher in France (exponentiated estimate 1.18, p=0. 0268), compared to patients without OIRD episodes (Table 1). Modeling of US cost data indicated that both length of stay and OIRD episodes impacted total cost, with the cost increasing exponentially for patients with ≥1 OIRD episode as length of stay increases (Figure 1). CONCLUSIONS: Utilizing the highly granular PRODIGY trial data for this prespecified secondary analysis, we were able to demonstrate a significant relationship between OIRD and cost of care. OIRD increases both direct and indirect costs (hospital length of stay). Due to different reimbursement systems, it is challenging to extrapolate US cost data to European and Asian countries. Early identification of patients at risk for OIRD holds the promise to not only improve patient safety and outcome, but also reduce cost of care.


American Society of Anesthesiologists, The Annual Anesthesiology Meeting, October 2019, Orlando FL. Abstract Number: A2242

This document is currently not available here.