P J Veugelers1, A L Kim, J R Guernsey. 1. Department of Community Health and Epidemiology, Faculty of Medicine, Dalhousie University, 5849 University Avenue, Halifax, Nova Scotia, B3H 4H7, Canada.
Abstract
STUDY OBJECTIVE: Simple measures of inequalities in health are proposed to facilitate the work of health policy makers and to build on the understanding of health differences between populations. In addition, it is aimed to make these measures applicable for comparisons of small populations and subgroups. METHODS: Inequalities in health or health deficiencies were quantified as the difference between the life expectancy of the subgroup of interest and that of the national population. Health deficiencies were divided into disease specific components by partial application of cause eliminated life table methods. To manage small numbers and to depict time trends, locally weighted regression smoothing was applied. Confidence intervals were constructed through Monte Carlo simulations. APPLICATIONS AND COMPARISONS: The proposed approaches were applied to the health situation in Cape Breton County, Nova Scotia, Canada, and disclosed the significance of different diseases and distinct patterns between communities. The proposed measures were also compared with the traditionally used standardised mortality rates and ratios. Here, the proposed measures appeared beneficial in that they are easier to comprehend and that they provide time trends and more robust estimates. CONCLUSIONS: The above advantages make the proposed approaches beneficial to health policy makers and epidemiologists. The approaches may also be incorporated in economic evaluations as well as in more sophisticated public health models.
STUDY OBJECTIVE: Simple measures of inequalities in health are proposed to facilitate the work of health policy makers and to build on the understanding of health differences between populations. In addition, it is aimed to make these measures applicable for comparisons of small populations and subgroups. METHODS: Inequalities in health or health deficiencies were quantified as the difference between the life expectancy of the subgroup of interest and that of the national population. Health deficiencies were divided into disease specific components by partial application of cause eliminated life table methods. To manage small numbers and to depict time trends, locally weighted regression smoothing was applied. Confidence intervals were constructed through Monte Carlo simulations. APPLICATIONS AND COMPARISONS: The proposed approaches were applied to the health situation in Cape Breton County, Nova Scotia, Canada, and disclosed the significance of different diseases and distinct patterns between communities. The proposed measures were also compared with the traditionally used standardised mortality rates and ratios. Here, the proposed measures appeared beneficial in that they are easier to comprehend and that they provide time trends and more robust estimates. CONCLUSIONS: The above advantages make the proposed approaches beneficial to health policy makers and epidemiologists. The approaches may also be incorporated in economic evaluations as well as in more sophisticated public health models.
Authors: P J Veugelers; P G Cornelisse; K J Craib; S A Marion; R S Hogg; S A Strathdee; J S Montaner; M V O'Shaughnessy; M T Schechter Journal: Am J Epidemiol Date: 1998-09-01 Impact factor: 4.897
Authors: P J Veugelers; S A Strathdee; A R Moss; K A Page; B Tindall; M T Schechter; R A Coutinho; G J van Griensven Journal: Epidemiology Date: 1995-07 Impact factor: 4.822