Study of social loss or gain from a particular cause: scope for the developing countries - a review with reference to cancer

Abstract

Incidence rates have long been used to assess the burden of different diseases in a population, whereas loss due to occurrence of diseases is studied using the death rates. Death rates however, are based on and therefore describe, only number of lives lost. There have been two approaches to arrive at the actual loss or gain from a particular cause viz. Person years of life lost (PYLL) approach and cause elimination life table (CELT) approach. This review covers these approaches and the competing risk theory and models focusing on the methodological developments. A summary of the conceptual and methodological developments on these concepts has also been presented. There are eight possible approaches in dealing with the loss in the presence or gain in the absence of a particular cause of death depending upon the preferences related to PYLL/CELT approach, modeling/descriptive approach, considering or without considering competing causes. A close look at the two basic approaches reveals that PYLL and cause elimination are just different terminologies used to address the same quantity, loss in the presence or gain in the absence. As far as descriptive vs. modeling approaches are concerned, all the descriptive procedures can be put in the form of models and all the models can be presented in a descriptive way. Regarding results using different models, no practical difference exists in the results based on different models for competing risks. However, exclusion of the competing risks may result in a considerable bias in the developing countries where general mortality is relatively higher. This review study suggests freedom in the selection of a modeling or a descriptive approach without any considerable loss of accuracy but at the same time emphasizes the consideration of the competing risks. An empirical study may be recommended to confirm the conclusions of this study.