Religious and ethnic differences in health: evidence from the Health Surveys for England 1999 and 2004

Abstract

Objectives. To examine the role of religion in the patterning of health inequities, and how this is related to ethnicity and socioeconomic status.

Design. Multivariate analyses using nationally representative data on self-assessed fair or poor health, longstanding limiting illness, diagnosed diabetes, diagnosed hypertension, waist–hip ratio, body mass index, current tobacco use and participating in no regular physical activity from 14,924 Christians, 4337 Muslims, 656 Sikhs, 1197 Hindus and 2682 people reporting not identifying with any religion with different ethnic backgrounds, who were interviewed as part of the Health Survey for England in either 1999 or 2004, adjusted for age, gender and socioeconomic status and periodicity.

Results. Odds ratios for general health, hypertension, diabetes, waist–hip ratio, tobacco use and physical activity speak to the importance of ethnicity in the patterning of health inequalities. But there is also evidence of an important, independent role for religion, with risks for the different health indicators varying between people with the same ethnic, but different religious, identifications. Adjusting for socioeconomic status attenuated the ethnic/religious patterning of, particularly, self-assessed health, longstanding activity-limiting illness, waist–hip ratio, body mass index and tobacco use.

Conclusions. This evidence enables greater understanding of the complexities of the relationship between ethnicity, religion and health, recognising the need to understand the heterogeneity underlying both ethnic and religious group membership and the processes producing the structural disadvantage facing certain religious and ethnic groups in the mediation of the relationship between health and ethnicity/religion.

Keywords:

• religion
• ethnic
• health
• socioeconomic
• class
• gender
• minority
• Muslim
• Hindu
• Sikh
• Christian
• obesity
• diabetes
• hypertension

Acknowledgements

This work was funded by the Economic and Social Research Council (RES-163-25-0009).

Notes

1. People responding ‘Roman Catholic’, ‘Church of England’, ‘Church of Wales’, ‘Church of Scotland’, ‘Church of Ireland’, ‘Jehovah's Witness’, ‘Baptist’, ‘Methodist’, ‘Pentecostal’, ‘Church of God’, ‘Church of Christ’, ‘Seventh Day Adventist’, ‘United Reformed’, ‘Presbyterian’, ‘Congregational’ or ‘other Christian’ were coded ‘Christian’.