The heavier the better? Birthweight and perinatal mortality in different ethnic groups

Int J Epidemiol. 2002 Jun;31(3):654-60. doi: 10.1093/ije/31.3.654.

Abstract

Background: Mother's ethnicity is associated with her baby's birthweight and risk of perinatal mortality. Given the close relation between birthweight and perinatal mortality, we explored whether ethnic differences in birthweight explain ethnic differences in perinatal mortality.

Methods: Data on all births to mothers born in Norway (808 658), Pakistan (6854), Vietnam (3283) and North Africa (1461) from 1980 to 1995 were obtained from the Medical Birth Registry of Norway. The associations between birthweight and perinatal mortality among ethnic groups were analysed using univariate and multivariate methods.

Results: Mean birthweights were low for Vietnamese and Pakistani mothers (3202 g, 3244 g) and high for Norwegian and North African mothers (3530 g, 3559 g). Mean birthweights were largely unrelated to perinatal mortality, which was lowest for Vietnamese (8.2/1000, 95% CI: 5.1-11.3) and highest for Pakistanis (14.9/1000, 95% CI: 12.0-17.7). Intermediate perinatal mortality rates were found among Norwegians (9.5/1000, 95% CI: 9.3-9.7) and North Africans (9.6/1000, 95% CI: 4.6-14.6). Further comparison of weight-specific mortality rates between the two largest ethnic groups showed the low birthweight paradox, where among low-weight births, perinatal mortality was lower among Pakistani than among Norwegian babies. However, adjustment to a relative birthweight scale (units of standard deviations from population-specific mean value) revealed higher rates of weight-specific mortality among Pakistanis across the entire range of birthweights. Multivariate adjustment for relative birthweight and other factors did not change these results.

Conclusions: Differences in perinatal mortality between the ethnic groups were not explained by differences in mean birthweight. Paradoxical differences in birthweight-specific mortality rates could be resolved by adjustment to a relative scale.

Publication types

  • Comparative Study

MeSH terms

  • Adult
  • Africa, Northern / ethnology
  • Analysis of Variance
  • Asia / ethnology
  • Birth Weight*
  • Ethnicity / statistics & numerical data*
  • Female
  • Humans
  • Infant Mortality*
  • Infant, Newborn
  • Multivariate Analysis
  • Norway / epidemiology
  • Pregnancy
  • Risk