Drivers and recent trends of hospitalisation costs related to acute pulmonary embolism

Abstract

Background and aims: The socio-economic burden imposed by acute pulmonary embolism (PE) on European healthcare systems is largely unknown. We sought to determine temporal trends and identify cost drivers of hospitalisation for PE in Germany.

Methods and results: We analysed the totality of reimbursed hospitalisation costs in Germany (G-DRG system) in the years 2016-2020. Overall, 484 884 PE hospitalisations were coded in this period. Direct hospital costs amounted to a median of 3572 (IQR, 2804 to 5869) euros, resulting in average total reimbursements of 710 million euros annually. Age, PE severity, comorbidities and in-hospital (particularly bleeding) complications were identified by multivariable logistic regression as significant cost drivers. Use of catheter-directed therapy (CDT) constantly increased (annual change in the absolute proportion of hospitalisations with CDT + 0.40% [95% CI + 0.32% to + 0.47%]; P < 0.001), and it more than doubled in the group of patients with severe PE (28% of the entire population) over time. Although CDT use was overall associated with increased hospitalisation costs, this association was no longer present (adjusted OR 1.02 [0.80-1.31]) in patients with severe PE and shock; this was related, at least in part, to a reduction in the median length of hospital stay (for 14.0 to 8.0 days).

Conclusions: We identified current and emerging cost drivers of hospitalisation for PE, focusing on severe disease and intermediate/high risk of an adverse early outcome. The present study may inform reimbursement decisions by policymakers and help to guide future health economic analysis of advanced treatment options for patients with PE.

Keywords: Catheter-directed treatment; Cost of illness; Economic burden; Hospitalisation costs; Pulmonary embolism.

Fig. 1

Patient characteristics as potential cost drivers during hospitalisation. For every influencing factor, median hospitalisation costs with interquartile range for presence (upper forest plot with red medians) or absence (lower forest plots with blue medians) of these factors was calculated. COPD, chronic obstructive pulmonary disease; COVID-19, coronavirus disease 19

Fig. 2

Association, in patients with shock, of different reperfusion treatment procedures with in-hospital mortality (A), major bleeding (B), reimbursed hospitalisation costs (C) and length of hospital stay (D). Results of univariate and multivariable logistic regression analysis are presented as odds ratios with corresponding 95% confidence intervals (CI), representing the use versus non-use of the respective treatment. The multivariable model adjusted for the following variables: age, sex, obesity, diabetes mellitus, cancer, coronary artery disease, heart failure, chronic obstructive pulmonary disease, essential arterial hypertension, acute/chronic kidney failure, surgery, chronic anaemia and atrial fibrillation/flutter

Fig. 3

Annual trends of hospitalisations (A), in-hospital case fatality (B) and reimbursed costs (C), in euros, of patients with pulmonary embolism in Germany during the study period. In C, the continuous black line denotes median costs; the dashed lines, the corresponding upper and lower quartile values. Note, the line of the lower quartile is almost superimposed on the solid line of the median costs

Fig. 4

The changing landscape of pulmonary embolism management in Germany. Annual trends revealed progressive decline of the proportion of reperfusion treatment with systemic thrombolysis (β for annual absolute change, − 0.04% [95%CI − 0.06 to − 0.02%]) and a constantly very low rate of surgical embolectomy in the entire population of hospitalised PE patients (A). Frequency of use of systemic thrombolysis and surgical embolectomy was relatively high (only) among very young patients, but it fell with growing age (B). It is however in older patients, after the sixth decade of life, that absolute numbers of hospitalisations and case fatality rates of PE increased dramatically (C). Annual trends further showed consistently large numbers of hospitalisations with severe pulmonary embolism (D). The proportion of use of catheter-directed treatment hospitalisations of patients with severe pulmonary embolism increased constantly since 2017 (E). Finally, linear regression analysis showed a decrease in case fatality of severe pulmonary embolism, with a (possibly temporary) rebound in the pandemic year 2020 (F)