Multiple myeloma (MM) has an inherent high risk of thromboembolic events associated with patient as well as disease- and treatment-related factors. Previous studies have assessed the association of MM-related thromboembolism using “traditional” Kaplan–Meier (KM) and/or Cox proportional hazard (PH) regression. In the presence of high incidence of death, as would be the case in cancer patients with advanced age, these statistical models will produce bias estimates. Instead, a competing risk framework should be used. This study assessed the baseline patient demographic and clinical characteristics associated with MM-related thromboembolism and compared the cumulative incidence and the measures of association obtained using each statistical approach. The cumulative incidence of thromboembolism was 9.2% using the competing risk framework and nearly 12% using the KM approach. Bias in the measures of covariate risk associations was highest for factors related to risk of death such as increased age (75% bias) and severe liver disease (50%) for the Cox PH model compared to the competing risk model. These results show that correct specification of statistical techniques can have a large impact on the results obtained.

Document Type


Publication Date


Notes/Citation Information

Published in Healthcare, v. 4, issue 1, 16, p. 1-11.

© 2016 by the authors; licensee MDPI, Basel, Switzerland.

This article is an open access article distributed under the terms and conditions of the Creative Commons by Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).

Digital Object Identifier (DOI)


Funding Information

The project described was supported by the National Center for Advancing Translational Sciences, National Institutes of Health, through grant number UL1TR000117.