Jump to comment:

• Do cardiometabolic, behavioural and socioeconomic factors explain the ‘healthy migrant effect’ in the UK? Linked mortality follow-up of South Asians compared with white Europeans in the Newcastle Heart Project
  Raj S Bhopal, Louise Hayes and Richard J Q McNally
  Published on: 5 September 2017
• Evidence for a healthy migrant effect on mortality in England and Wales
  Ian M Timæus, Anne P Scott
  Published on: 16 August 2017

• Published on: 5 September 2017

  Do cardiometabolic, behavioural and socioeconomic factors explain the ‘healthy migrant effect’ in the UK? Linked mortality follow-up of South Asians compared with white Europeans in the Newcastle Heart Project

  • Raj S Bhopal, Professor of Public health Usher Institute of Population Health Sciences and Informatics, The University of Edinburgh
  • Other Contributors:
    • Louise Hayes, Research methodologist
    • Richard J Q McNally, Reader in epidemiology

  We thank Timaeus and Scott for drawing readers' attention to our interpretation(1) of their data which differs from their own(2) (rapid response 28/7/2017). We are glad to explain our thinking especially as the issues go beyond their data and to the concepts and the UK quantitative evidence. We agree that in their paper after adjustment for three socio-economic and an area of residence variables the mortality rate ratios are lower in South Asian groups than in the White group.(2) The explanation for our different interpretation is that we placed emphasis on their model adjusting mortality for age, sex and period while they emphasised the results of models further adjusting for socio-economic status and residence.(2)

  Generally the ‘healthy migrant effect’ is considered as unexpected and hence a paradox because immigrant populations sometimes have better health, most usually mortality, despite their socio-economic and other disadvantages.(3, 4) It is not generally understood as an effect that arises after adjustments for socio-economic and other related factors. In Timaeus and Scott’s model 1 the rate ratios for Indian, Pakistani and Bangladeshi populations born abroad and participating in the Longitudinal Study in England and Wales are shown in their table 5 and were 0.91, 0.95 and 1.01 with the 95% confidence intervals all including the reference value of 1. In model 1, the point estimates of the rate ratios for the same ethnic groups born in the UK were simil...

  Show More

  We thank Timaeus and Scott for drawing readers' attention to our interpretation(1) of their data which differs from their own(2) (rapid response 28/7/2017). We are glad to explain our thinking especially as the issues go beyond their data and to the concepts and the UK quantitative evidence. We agree that in their paper after adjustment for three socio-economic and an area of residence variables the mortality rate ratios are lower in South Asian groups than in the White group.(2) The explanation for our different interpretation is that we placed emphasis on their model adjusting mortality for age, sex and period while they emphasised the results of models further adjusting for socio-economic status and residence.(2)

  Generally the ‘healthy migrant effect’ is considered as unexpected and hence a paradox because immigrant populations sometimes have better health, most usually mortality, despite their socio-economic and other disadvantages.(3, 4) It is not generally understood as an effect that arises after adjustments for socio-economic and other related factors. In Timaeus and Scott’s model 1 the rate ratios for Indian, Pakistani and Bangladeshi populations born abroad and participating in the Longitudinal Study in England and Wales are shown in their table 5 and were 0.91, 0.95 and 1.01 with the 95% confidence intervals all including the reference value of 1. In model 1, the point estimates of the rate ratios for the same ethnic groups born in the UK were similar to the above for Indian (0.88) and Pakistani (0.99) populations but much lower for Bangladeshis (0.55, with wide confidence intervals).

  The interpretation of results following adjustment for socio-economic variables across different ethnic groups is difficult as reflected in a literature with contrary and unexpected findings,(5, 6) stimulating conceptual and methodological explorations.(7) Timaeus and Scott have, in addition to three socio-economic variables, also included one relating to residence in Metropolitan and non-metropolitan areas. Adjustments made little difference to the rate ratio for Indians (0.88), but reduced it for Pakistani (0.78) and Bangladeshi (0.59) populations born abroad (model 2, table 5). It is not straightforward to interpret the data, especially without examining the associations between each variable and the outcome by ethnic group as Fischbacher et al explain.(7)

  Given the lack of good evidence for a healthy migrant effect in South Asian/Indian Subcontinent populations in England and Wales, (5, 8-10) though it was clear cut in Scotland, (11, 12) we interpreted Timaeus and Scott's paper as providing little evidence in favour of one. An alternative interpretation that we now offer on further reflection would have been somewhat closer to the authors’ own interpretation i.e. in Indian, Pakistani and Bangladeshi populations there was little evidence of a healthy migrant effect in age and sex adjusted data though adjustment for four additional factors indicated that it may have been suppressed by unfavourable socio-economic circumstances and living in certain kinds of areas.

  Variable interpretations of data are sometimes genuine and sometimes mistakes; ours is in the former category. On this theme, Timaeus and Scott stated women born in the West Africa had lower all-cause mortality than those born in England and Wales in Wild et al’s work. But the SMR was 121(9) They also said that immigrants to England and Wales from most places of origin have lower all-cause mortality than the UK born population-that is not true for many groups, particularly those from the Indian subcontinent.(5, 9, 10, 13)

  Timaeus and Scott say that UK-born minority ethnic populations do not have lower mortality than UK born Whites(2) but, as shown above, for the South Asian groups the point estimates of the rate ratios were similar to those for South Asians born abroad, and in the case of Bangladeshis were considerably lower. (That the confidence intervals straddled the reference value is surely a reflection of the smaller numbers of outcomes.)

  We welcome this debate, which will hopefully stimulate further work, especially on linked mortality data sets, which are producing results(1, 2, 12, 14) which are somewhat different, especially in South Asians, from those in cross-sectional analysis of unlinked mortality and population size data.(5, 9, 10, 13) The totality of the evidence merits careful examination, possibly with reanalysis for better comparison, to judge whether there is truly a healthy migrant effect in South Asians in the UK.

  1. Hayes L, White M, McNally RJQ, Unwin N, Tran A, Bhopal R. Do cardiometabolic, behavioural and socioeconomic factors explain the ‘healthy migrant effect’ in the UK? Linked mortality follow-up of South Asians compared with white Europeans in the Newcastle Heart Project. Journal of Epidemiology and Community Health. 2017.
  2. Scott AP, Timaeus IM. Mortality differentials 1991-2005 by self-reported ethnicity: findings from the ONS Longitudinal Study. J Epidemiol Community Health. 2013;67(9):743-50.
  3. Roura M. Unravelling migrants' health paradoxes: a transdisciplinary research agenda. J Epidemiol Community Health. 2017;71(9):870-3.
  4. Bhopal RS. Migration, Ethnicity, Race and Health in Multicultural Societies. 2 ed. Oxford: Oxford University Press; 2014.
  5. Marmot MG, Adelstein AM, Bulusu L. Immigrant mortality in England and Wales 1970 -78. Causes of death by country of birth. London: HMSO; 1984.
  6. Harding S, Balarajan R. Longitudinal study of socio-economic differences in mortality among South Asian and West Indian migrants. EthnHealth. 2001;6:121-8.
  7. Fischbacher CM, Cezard G, Bhopal RS, Pearce J, Bansal N. Measures of socioeconomic position are not consistently associated with ethnic differences in cardiovascular disease in Scotland: methods from the Scottish Health and Ethnicity Linkage Study (SHELS). International Journal of Epidemiology. 2014;43(1):129-39.
  8. Balarajan R, Bulusu L, Adelstein AM, Shukla V. Patterns of mortality among migrants to England and Wales from the Indian subcontinent. BMedJ. 1984;289:1185-7.
  9. Wild SH, Fischbacher C, Brock A, Griffiths C, Bhopal R. Mortality from all causes and circulatory disease by country of birth in England and Wales 2001-2003. J Public Health (Oxf). 2007;29(2):191-8.
  10. Balarajan R, Bulusu L. Mortality among immigrants in England and Wales, 1979 - 83. In: Britton M, editor. Mortality and Geography: A review in the mid 1980's. London: HMSO; 1990. p. 103-21.
  11. Fischbacher C, Steiner M, Bhopal R, Chalmers J, Jamieson J, Knowles D, et al. Variations in all cause and cardiovascular mortality by country of birth in Scotland, 1997-2003. Scottish Medical Journal. 2007;52(4):5-10.
  12. Gruer L, Cezard G, Clark E, Douglas A, Steiner M, Millard A, et al. Life expectancy of different ethnic groups using death records linked to population census data for 4.62 million people in Scotland. Journal of Epidemiology and Community Health. 2016;70:1251-4.
  13. Wild S, McKeigue P. Cross sectional analysis of mortality by country of birth in England and Wales, 1970-92. BMJ. 1997;314(7082):705-10.
  14. Wallace M, Hill K. Mortality Among Immigrants in England and Wales: A Longitudinal Study. 2014.

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top
• Published on: 16 August 2017

  Evidence for a healthy migrant effect on mortality in England and Wales

  • Ian M Timæus, Professor of Demography London School of Hygiene & Tropical Medicine
  • Other Contributors:
    • Anne P Scott, Demographer

  Hayes et al. [1] repeatedly cite a 2013 article by Scott and Timæus [2], also published in this journal, as having ‘not found a healthy migrant effect in South Asians’ and as providing ‘little evidence of a South Asian mortality advantage’. This contradicts our own interpretation of the results that we presented in that paper. We concluded that ‘Immigrants are selected for good health’. Moreover, with specific reference to South Asians, we stated that: ‘adjusted for SES and residence, … Indian, Pakistani, [and] Bangladeshi … immigrants all had lower mortality than UK-born Whites who were living in similar circumstances to them … This suggests that immigrants from the Indian subcontinent … are … selected for health’.

  We think it regrettable that Hayes et al. do not indicate to readers of their paper that their interpretation of the results in our paper is almost diametrically opposed to our own. Moreover, they provide no explanation whatsoever of why they came to the view that we had misinterpreted our results.

  Our study investigated all-cause mortality at ages 1−79 in 1991−2005 by self-reported ethnicity and country of birth. The data were from the Office for National Statistics Longitudinal Study of England and Wales for the cohort aged 0−64 in 1991. Poisson regression was used to adjust the estimates for metropolitan residence and three indicators of socioeconomic status. In the fully-adjusted model, but not the model that adjusted only for age, sex and per...

  Show More

  Hayes et al. [1] repeatedly cite a 2013 article by Scott and Timæus [2], also published in this journal, as having ‘not found a healthy migrant effect in South Asians’ and as providing ‘little evidence of a South Asian mortality advantage’. This contradicts our own interpretation of the results that we presented in that paper. We concluded that ‘Immigrants are selected for good health’. Moreover, with specific reference to South Asians, we stated that: ‘adjusted for SES and residence, … Indian, Pakistani, [and] Bangladeshi … immigrants all had lower mortality than UK-born Whites who were living in similar circumstances to them … This suggests that immigrants from the Indian subcontinent … are … selected for health’.

  We think it regrettable that Hayes et al. do not indicate to readers of their paper that their interpretation of the results in our paper is almost diametrically opposed to our own. Moreover, they provide no explanation whatsoever of why they came to the view that we had misinterpreted our results.

  Our study investigated all-cause mortality at ages 1−79 in 1991−2005 by self-reported ethnicity and country of birth. The data were from the Office for National Statistics Longitudinal Study of England and Wales for the cohort aged 0−64 in 1991. Poisson regression was used to adjust the estimates for metropolitan residence and three indicators of socioeconomic status. In the fully-adjusted model, but not the model that adjusted only for age, sex and period, immigrants of every ethnicity other than African and other non-Caribbean Black immigrants had significantly lower mortality than the UK-born White population. Most of our estimates of mortality by ethnic group in the UK-born population had wide confidence intervals. However, the UK-born Black Caribbean population had significantly higher mortality than the White population when we adjusted only for age, sex and period. This association disappeared after we also adjusted for socioeconomic status and place of residence.

  Our own interpretation of these results remains that they demonstrate that immigrants to England and Wales, including those from South Asia, are selected for good health. However, the potential reduction in mortality arising from this healthy migrant effect is largely offset by the adverse consequences of the low socioeconomic status of most immigrants compared with the population as a whole (and equally vice versa). Our results provide no evidence to suggest that selection for health in one generation results in a persistent benefit to the mortality of subsequent generations.

  References

  [1] Hayes L, White M, McNally RJQ, et al. Do cardiometabolic, behavioural and socioeconomic factors explain the ‘healthy migrant effect’ in the UK? Linked mortality follow-up of South Asians compared with white Europeans in the Newcastle Heart Project. Journal of Epidemiology and Community Health 2017. doi:10.1136/jech-2017-209348

  [2] Scott AP, Timæus IM. Mortality differentials 1991-2005 by self-reported ethnicity: findings from the ONS Longitudinal Study. Journal of Epidemiology and Community Health 2013; 67:743-50. doi:10.1136/jech-2012-202265

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top