Abstract-This paper offers a critical appraisal of the various methods employed to date to measure inequalities in health. It suggests that only two of these--the slope index of inequality and the concentration index-are likely to present an accurate picture of socioeconomic inequalities in health. The paper also presents several empirical examples lo illustrate of the dangers of using other measures such as the range, the Lorenz curve and the index of dissimilarity.

1 We arc grateful to Frans Ruttcn for support throughout the research leading up to this paper, to the Centro Europa Riccrche (CER) in Rome-and in particular to Stefania Gabriel-for providing data from the CER Health Care Consumption Survey, to Ugo Ercolani and George France for help in obtaining data on Italian health care expenditure; and to Jaap van den Berg of the Dutch Central Bureau of Statistics (CBS) for providing pre-rclcasc versions of recently published tables from the CBS Health Interview Survey. 2 Lc Grand finds that 'the top socioeconomic group (professionals, employers and managers) receives 40 per cent more NHS expenditure per person reporting illness than the bottom one (semi-skilled and unskilled manual workers)' (Le Grand, 1982, p. 46).

This paper offers a critical appraisal of the various methods used to date to investigate inequity in the delivery of health care. It concludes that none of the methods used to date is particularly well equipped to provide unbiassed estimates of the extent of inequity. It also concludes that Le Grand's (1978) approach is likely to point towards inequity favouring the rich even when none exists. The paper offers an alternative approach, which builds on the approaches to date but seeks to overcome their deficiencies.

This paper exploits the Workplace Industrial Relations Survey from 1990 (WIRS3) to examine the determinants of workplace injuries for a sample of manufacturing establishments in the UK. A key focus of this paper is an assessment of the role played by union‐appointed safety representatives and joint health and safety consultative committees in reducing the frequency of workplace accidents. We find that joint consultative committees, with all employee representatives appointed by unions, significantly reduce workplace injuries relative to those establishments where the management alone determine health and safety arrangements. However, an important role is also found for those joint consultative commitees where no employee representatives are appointed by unions.

No abstract