On the measurement of relative and absolute income-related health inequality
Introduction
The measurement of income-related inequalities in health has become an important field in health economics (Wagstaff, Paci, & van Doorslaer, 1991). In a recent contribution, van Doorslaer et al. (1997) for instance used concentration indices to estimate income-related inequalities for nine industrialized countries using a measure of morbidity based on self assessed health. In a subsequent contribution, Gerdtham and Johannesson (2000) estimated concentration indices with respect to both measures of morbidity and health based on quality adjusted life years in different age groups in Sweden. They noted that the age pattern differed depending on whether the concentration indices were estimated with respect to morbidity or health. Recently, van Doorslaer and Koolman (2000) have found that ranking of relative inequality measures across countries depends on whether health or morbidity is used. To avoid this problem, they also report absolute inequality measures for each country. The main purpose of this methodological note is to illustrate why the ranking of relative income-related health inequality between countries and within countries across dimensions of health depends on whether inequality is measured with respect to health or morbidity. We also examine why this problem does not arise when absolute inequality measures are employed and exemplify these issues using comparisons of scales from the Short Form 36 (SF-36) (Ware & Gandek, 1998) included in population health surveys conducted in Sweden and Australia.
Section snippets
Methods
The last few years has witnessed the development of standardized surveys for measuring health status and these are likely to be increasingly used in the measurement and comparisons of health inequalities in different populations. A prime example is the SF-36 health survey that has been translated and studied in over 40 countries (Ware & Gandek, 1998). The SF-36 is a 36 item questionnaire that measures health in terms of eight multi-item scales: physical functioning (PF), role limitations due to
Results
Table 1 provides summary statistics for each scale of the SF-36 surveys by country. With the exception of BP the mean scores in the Swedish sample were significantly higher than those reported in the Australian sample. The concentration indexes for health scores by scale are reported in the second and third column of Table 2. In both Australia and Sweden, the highest degree of inequality was in RP with 0.033 and 0.031, respectively. The relative concentration indexes in terms of morbidity
Discussion
The concentration index has been used in comparisons of the income-related health inequality between countries (van Doorslaer et al., 1997). Our results show that whether the concentration index is defined with respect to morbidity or health can affect the ranking between countries and across dimensions of health. However, this issue is not confined to the concentration index as all relative health inequality measures (e.g. comparing the relative inequality of the highest and lowest income
Acknowledgements
We would like to thank Mark Dusheiko for helpful comments on a previous draft of this paper.
References (14)
- et al.
More or less equal? Comparing Australian income-related inequality in self-reported health with other industrialised countries
Australian and New Zealand Journal of Public Health
(2000) - et al.
Income-related inequality in life-years and quality-adjusted life-years
Journal of Health Economics
(2000) - et al.
A note on validating Wagstaff and van Doorslaer's health measure in the analysis of inequalities in health
Journal of Health Economics
(1999) - et al.
Socio-economic inequalities in healthmeasurement, computation and statistical inference
Journal of Econometrics
(1997) - et al.
Income-related inequalities in health some international comparisons
Journal of Health Economics
(1997) - et al.
On the measurement of inequalities in health
Social Science and Medicine
(1991) - et al.
Overview of the SF-36 Health Survey and the International Quality of Life Assessment (IQOLA) Project
Journal of Clinical Epidemiology
(1998)