Application of Multiplicative and Additive Hazards Models to Injury Prevention among Healthcare Workers
Sabuj Sarker
Department of Community Health and Epidemiology, College of Medicine, University of Saskatchewan, 107 Wiggins Road, Saskatoon, SK S7N 5E5, Canada
Dae- Kee Min
Department of Information Statistics, Duksung Women’s University, 33 Ssangmun-Dong Dobong-Gu Seoul, South Korea
Timothy R. Black
Department of Human Resources, University of Saskatchewan, 105 Administration Place Saskatoon, Saskatchewan, Canada
Hyun J. Lim *
Department of Community Health and Epidemiology, College of Medicine, University of Saskatchewan, 107 Wiggins Road, Saskatoon, SK S7N 5E5, Canada
*Author to whom correspondence should be addressed.
Abstract
The Cox multiplicative model is used widely in survival analysis, where the covariates act multiplicatively on unknown baseline hazards. However, the Cox model requires the proportionality assumption, which limits its applications. The additive hazards model has been used as an alternative to the Cox model, where the covariates act additively on unknown baseline hazards. In this study, the performance of the Cox multiplicative hazards model and the additive hazards model has been demonstrated in an injury prevention study. Both the multiplicative and additive hazards models show similar results in selecting significant covariates in the final model in our study. The coefficient of the covariates in the additive hazards model is easy to interpret in an additive manner and should be considered when the proportionality assumption of the Cox model is doubtful. The multiplicative and additive hazards models describe different features of the association between the risk factors and the study outcomes. They may be used each other as supplementary approach for further understanding of the data.
Keywords: Survival analysis, Cox model, Aalen’s model, lin and Ying’s model, injury study, occupational health