Modelling costs of interventional pulmonary embolism treatment: implications of US trends for a European healthcare system

Abstract

Aims: Catheter-directed treatment (CDT) of acute pulmonary embolism (PE) is entering a growth phase in Europe following a steady increase in the USA in the past decade, but the potential economic impact on European healthcare systems remains unknown.

Methods and results: We built two statistical models for the monthly trend of proportion of CDT among patients with severe (intermediate- or high-risk) PE in the USA. The conservative model was based on admission data from the National Inpatient Sample (NIS) 2016-20 and the model reflecting increasing access to advanced treatment from the PERT™ national quality assurance database registry 2018-21. By applying these models to the forecast of annual PE-related hospitalizations in Germany, we calculated the annual number of severe PE cases and the expected increase in CDT use for the period 2025-30. The NIS-based model yielded a slow increase, reaching 3.1% (95% confidence interval 3.0-3.2%) among all hospitalizations with PE in 2030; in the PERT-based model, increase would be steeper, reaching 8.7% (8.3-9.2%). Based on current reimbursement rates, we estimated an increase of annual costs for PE-related hospitalizations in Germany ranging from 15.3 to 49.8 million euros by 2030. This calculation does not account for potential cost savings, including those from reduced length of hospital stay.

Conclusion: Our approach and results, which may be adapted to other European healthcare systems, provide a benchmark for healthcare costs expected to result from CDT. Data from ongoing trials on clinical benefits and cost savings are needed to determine cost-effectiveness and inform reimbursement decisions.

Keywords: Catheter-directed treatment; Cost of illness; Economic impact; Hospitalization costs; Pulmonary embolism.

Figure 1

Observed and modelled monthly proportion of use of catheter-directed treatment in the National Inpatient Sample (2016–20) and the Pulmonary Embolism Response Teams (PERT) registry (2018–21). Observed monthly proportion of the use of catheter-directed treatment (CDT; black dots and lines) and modelled monthly proportion based on an individual patient-level binomial (logistic) model (green lines) among patients with severe pulmonary embolism (PE) in the National Inpatient Sample (left panel) and the PERT™ registry (right panel).

Figure 2

Number of patients with pulmonary embolism (PE) estimated to receive catheter-directed treatment (CDT) in Germany, 2025–30. The total number of patients with PE and the predicted proportion of those with severe PE for the time period 2025–30 were estimated based on (i) a forecast for the German population and (ii) observed trends of the proportion of patients with severe PE between 2016 and 2020, both obtained from the German Federal Statistical Office. The 95% confidence interval for the predicted CDT penetration in the German PE population for this future period was modelled separately from the US Nationwide Inpatient Sample (NIS; left panel) and the Pulmonary Embolism Response Teams (PERT)™ national quality assurance database registry (right panel) using binomial (logistic) regression as shown in Figure 1. The NIS model was trained on 2016–20 data, the PERT model was trained on 2018–21 data, and the calculated penetration in the years 2016–20 in each model was applied to the German population for the time period 2025–30.

Figure 3

Percent increase in total estimated costs of hospitalizations for pulmonary embolism (PE) in Germany until 2030, compared with the reference year 2025. Costs were calculated based on the estimated annual numbers of patients to be hospitalized with PE as depicted in Figure 2 and on current German reimbursement for PE hospitalization with vs. without catheter-directed treatment (CDT). Predicted increase of CDT costs was modelled separately from the US Nationwide Inpatient Sample (NIS; left panel) and the US Pulmonary Embolism Response Teams (PERT)™ national quality assurance database registry (right panel). The bars display the relative increase of costs relative to 2025 of each limit of the 95% confidence interval (CI) for the estimated number of patients using CDT. In the PERT-based model, the upper limit of the 95% CI, while always reflecting higher absolute numbers (as visible in Figure 2) and costs increases less quickly than the lower limit over time, thus displaying a slower increase relative to its baseline in 2025.