A Reliability Analysis of Occupational Exposure Data Using a Family of Proportional and Additive Hazard Functions

Jonathan F. Bard

DOI: 10.2190/643R-GM5C-VC73-XLX6


The growing debate on regulation in the work place calls for an improvement in the analytic techniques used to assess occupational risks and mortality. The intent of this article is to investigate a family of hazard models that may be employed in this effort. The underlying problem is first placed in context with a discussion of risk attitudes and the dangers of exposure. Next, proportional and additive hazard models are introduced which combine a standard (but unknown) hazard function with a set of explanatory variables. The family is developed with the aid of a parameterized link function, while maximum likelihood is used to obtain coefficient estimates. After describing the methodology an application is given for the case where only two samples of grouped data are available. A related example is then worked out using survival data collected on persons employed in the asbestos processing industry. The results confirm both the flexibility and sensitivity of the approach while leaving open the possibility of further refinements.

Creative Commons License This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License.