Elsevier

Social Science & Medicine

Volume 59, Issue 7, October 2004, Pages 1423-1433
Social Science & Medicine

Analysis of health-related quality of life (HRQL), its distribution, and its distribution by income in Japan, 1989 and 1998

https://doi.org/10.1016/j.socscimed.2004.01.025Get rights and content

Abstract

Research conducted over the past decade has increasingly focused on health inequality. The goals of improving the health of a population are now often expressed as the increase in the average level of health and the decrease in health inequality as clearly specified in the Healthy People 2010 and the World Health Report 2000. Using a nationally representative sample of the 1989 and 1998 waves of the Japanese Comprehensive Survey of Living Conditions of the People on Health and Welfare (CSLC), this research examines the average health-related quality of life (HRQL), its distribution, and its distribution by income share. This study departs from previous health inequality analyses in the following two ways: (1) construction of a measure of HRQL in the CSLC and its application to health inequality analysis, and (2) inclusion of the dead in health inequality analysis using a cross-sectional survey. This study found that between 1989 and 1998 the average HRQL in Japan slightly reduced (0.005 reduction), its inequality by income slightly reduced (0.002 reduction in the difference between the top 20% and bottom 20% income share groups), and its inequality measured by the Gini coefficient slightly increased (0.002 increase). Women’s HRQL was almost always lower than men’s, except in earlier ages younger than 10 years old. HRQL was more unequally distributed among women than men and in older ages. This analysis shows that the success in the improvement in the length of life in Japan did not always coincide with the improvement in HRQL and provides a basis for the future population health research.

Introduction

Research conducted over the past decade has increasingly focused on health inequality. The World Health Organization (WHO), for example, claims that a traditional average health of a population does not provide enough information as a population health measure, and investigation of the distribution of health within a population is necessary (World Health Organization, 2000). The goals of improving the health of a population are thus often expressed as the increase in the average level of health and the decrease in health inequality as clearly specified in the Healthy People 2010 (US Department of Health and Human Services, 2000) and the World Health Report 2000 (World Health Organization, 2000).

In this keen interest in health inequality, Japan may attract particular attention. Japanese people’s average health attainment is often ranked the best in the world: life expectancy was 77.1 years for men and 84.0 years for women in 1999 (Statistics and Information Department, 1999), and the Disability-adjusted Life Expectancy or DALE for Japanese people in 1997–1999 was estimated as 74.5 years (World Health Organization, 2000). It is interesting to investigate how health is distributed in the country with the world’s highest average level of health. Moreover, explanation of exactly what brought this high health attainment remains unsatisfactory beyond such general speculations as the public health system, work ethic, diet, and economy in Japan (Marmot & Smith, 1989). There is a general perception of the Japanese society as “egalitarian,” and some suspect that this egalitarian characteristic might relate to the high health attainment (Wilkinson, 1996). This leads to interests in health and its distribution in relation to such other goods as income.

Analysis of health inequality in Japan has just begun. At the overall country level, the WHO ranks equality of child survival in Japan as the third among its 191 member countries (World Health Organization, 2000). Within the country, studies have reported differences in health by income (Kondo, 2000; Shibuya, Hashimoto, & Yano, 2002), geographic location (Hasegawa, 2001; Takano, 1998), and occupation (Hasegawa, 2001). More studies need to follow, to fully explore a unique potential that Japan can offer in health inequality research.

We investigate a trend of health inequality in Japan between 1989 and 1998 using a nationally representative sample of the Comprehensive Survey of Living Conditions of the People on Health and Welfare (CSLC). Our analysis departs from previous health inequality analyses in Japan in the following two ways: (1) construction of a measure of health-related quality of life (HRQL) in the CSLC and its application to health inequality analysis, and (2) inclusion of the dead in health inequality analysis using a cross-sectional survey.

In our appreciation of health, we value both “living long” and “living well” (Mullahy, 2001). Such traditional measures as mortality and life expectancy correspond to our value of the former, while measures of HRQL attempt to capture the latter. Various measures of HRQL have been developed by different research groups, including the EQ-5D, the Health Utilities Index (HUI), and the Health and Activity Limitation Index (HALex) (for an excellent, comprehensive guide, see McDowell & Newll, 1996). None of these measures is perfect, but the development is rapid, their use expanding.

Japan is slow in introducing HRQL measures. The EQ-5D (Committee for the Development of the Japanese Version EuroQol, 1998) and the HUI (Hisashige, Mikasa, & Katayama, 1996) were, for example, translated into Japanese, but their application has been limited to small, exploratory samples and never been applied to a nationally representative sample. By applying the HALex with modification, this research is the first to use a measure of HRQL in analysis of the health of a nationally representative sample of Japanese people. We are fully aware of the cultural sensitivity of health valuation and difficulties in cross-cultural application of a HRQL measure. Nonetheless, health inequality cannot be measured without measuring health; the development of health inequality analysis must go hand in hand with the development of health measurement. The application of the US-born HALex to Japanese data must be validated, but we leave this issue aside for the future research.

The second feature of this research is the inclusion of the dead in the analysis of health inequality despite the use of a cross-sectional survey. A cross-sectional health survey usually only collects health information on the living and neglects the dead of the target population. Recognizing death as a health outcome, one might argue that analysis of the health of the living only provides partial information concerning health inequality in a population. However healthy the living population may be, if there were also a great proportion of deaths in the population, we may not be able to capture the true picture of health inequality merely by looking at the health of the living.

Moreover, HRQL information is generally obtained through cross-sectional surveys. HRQL measures define death (as zero), but data on death are missing in cross-sectional surveys. HRQL can incorporate death information by combining life years or expectancy and provide quality-adjusted life years or QALYs (or quality-adjusted life expectancy, QALEs). But if we were interested in a distribution of HRQL rather than that of QALYs or QALEs, would there be any way we can incorporate death? Our dead imputation method provides a full assessment of a HRQL distribution at a point of time, ranging from death to perfect health.

We set the following three objectives in this research:

  • (1)

    What was the average HRQL in Japan, overall, by sex and by age group in 1989 and 1998?

  • (2)

    How was HRQL distributed by income share in Japan, overall, by sex and by age group in 1989 and 1998?

  • (3)

    How was HRQL, from death to the perfect health, distributed in Japan, overall, by sex and by age group in 1989 and 1998?

This paper proceeds as follows. First, we explain data sources, construction of the HRQL measure, and health inequality measures used in this study. Second, we present results in the order of the three questions above. Finally, we discuss issues raised in this analysis.

In this report, “health distribution” is a way in which health is spread among individuals or groups of people in a population of concern, “health equality” suggests the health distribution in which health is spread equally to every party in a population of concern, and “health inequality” means all health distributions that are otherwise.

Section snippets

The living

Data about the living come from the CSLC conducted by the Ministry of Health, Labor and Welfare, Japan (Statistics and Information Department (1989a), Statistics and Information Department (1998a)). This cross-sectional survey of a nationally representative sample of non-institutionalized individuals consists of four parts, household, health, income, and assets. It has been collected every 3 years since 1986. Our analysis used household, health, and income data in 1989 and 1998, the earliest

The average HRQL

Between 1989 and 1998, the average HRQL for the living Japanese population slightly declined from 0.857 to 0.852 (both sexes combined). Table 4 shows life expectancies (Statistics and Information Department (1989b), Statistics and Information Department (1998b)) and the average HRQL for the living population by sex in 1989 and 1998. For both men and women, the improvement of the length of life during this period (1.25-year increase for men, 2.24-year increase for women) did not coincide with

Discussion

This study examined the average HRQL, its distribution by income, and its distribution itself among Japanese people in 1989 and 1998. This analysis showed that during this period in Japan the average HRQL slightly reduced (0.005 reduction), its inequality by income slightly reduced (0.002 reduction in the difference between the top 20% and bottom 20% income share groups), and its inequality measured by the Gini coefficient slightly increased (0.002 increase).

Since Japan is generally ranked as

Acknowledgements

We would like to acknowledge helpful suggestions and comments from Profs. John Mullahy, David Kindig, Patrick Remington, Dennis Fryback, and Barbara Wolfe in the Department of Population Health Sciences in the University of Wisconsin-Madison and Prof. Alberto Palloni in the Center for Demography and Ecology in the University of Wisconsin-Madison. We would also like to thank anonymous reviewers for their constructive comments. We claim responsibility for all remaining errors. The access and use

References (27)

  • W. Groot

    Adaptation and scale of reference bias in self-assessments of quality of life

    Journal of Health Economics

    (2000)
  • J. Le Grand

    Inequality in healthSome international comparison

    European Economic Review

    (1987)
  • Committee for the Development of the Japanese Version EuroQol. (1998). Development of the Japanese version EuroQol (in...
  • Cutler, D. M., & Richardson, E. (1997). Measuring the health of the US population. Brookings papers: Microeconomics,...
  • P. Erickson

    Evaluation of a population-based measure of quality of life—The Health and Activity Limitation Index (Halex)

    Quality of Life Research

    (1998)
  • Erickson, P., Wilson, R., & Shannon, I. (1995). Years of health life. Healthy people 2000. Statistical notes, Centers...
  • T. Hasegawa

    JapanHistorical and current dimensions of health and health equity

  • Hisashige, A., Mikasa, H., & Katayama, T. (1996). Quality of life of community residents from the multi-attribute...
  • Honda, C., & Ohkusa, Y. (2001). International comparison of subjective health evaluation—USA, UK and Japan. Discussion...
  • R. Illsley et al.

    Measurement of inequality in health

  • K. Kondo

    Why are there many elderly who need care from others in the low-income group?

    Shakai hoken jun ho

    (2000)
  • Luxembourg Income Study. (2003). Luxembourg income study key figures: Income inequality measures....
  • M.G. Marmot et al.

    Why are the Japanese living longer?

    British Medical Journal

    (1989)
  • Cited by (0)

    View full text