Papers

Why are babies getting heavier? Comparison of Scottish births from 1980 to 1992

BMJ 1997; 315 doi: https://doi.org/10.1136/bmj.315.7117.1205 (Published 08 November 1997) Cite this as: BMJ 1997;315:1205
  1. S R Bonellie, lecturera,
  2. G M Raab, professora
  1. a Department of Mathematics, Napier University, Edinburgh EH10 5DT
  1. Correspondence to: Dr Bonellie
  • Accepted 11 March 1997

Introduction

Over the past decade the weights of babies born in the United Kingdom have been increasing,1 which may have implications for the pattern of adult disease.2 From 1980 to 1992 the mean birth weight of live singleton births in Scotland increased steadily from 3326 g to 3382 g. We investigated factors that may explain this trend.

The distribution of gestational age has changed noticeably over the past decade because of the gradual introduction of its assessment by ultrasonography.3 In 1980 around 42% of all live births occurred at 40 weeks' gestation; by 1992 this had fallen to 32%. Thus trends in birth weights adjusted for gestation may be misleading and are not considered here.

Subjects, methods, and results

We assessed data on live, singleton births in Scotland as entered on SMR2 forms (discharge sheets for maternity admissions), to develop a predictive model for birth weight. The data covered about 94% of live singleton births from 1980 to 1992. Regression models for boys and girls were fitted separately using maternal age (linear and quadratic terms), height (linear and quadratic terms), and parity (null parous v other). A separate term was fitted for each calendar year. Each of these factors had a significant effect on birth weight.

From 1980 to 1992 mean maternal height increased from 160.0 cm to 161.9 cm, mean maternal age increased from 25.8 to 27.1 years, and the percentage of first births increased by about 2.5%. Changes in both mean maternal height and age were reflected by an overall upward shift in distribution.

Table 1 shows the effects of changes in maternal height, age, and parity on birth weight. For both sexes the difference in mean birth weight was substantially reduced when the increase in maternal height was taken into account. The increase in maternal age also reduced the difference, but the change in parity had the opposite effect as first babies have lower birth weights on average. Almost half of the difference in mean birth weight between 1980 and 1992 can be explained by changes in these three factors, the increase in maternal height having the largest effect. When all three variables were included in the model the effect of maternal age on birth weight was smaller since younger women are much more likely to be having their first babies.

Increases in mean birth weight between 1980 and 1992 for boys and girls born in Scotland.

View this table:

Comment

A decrease of 7% in the proportion of induced births over the past decade may explain about 10 g of the increase in birth weight. Other factors associated with low birth weight are maternal smoking and social deprivation.4 The proportion of women who smoke in Scotland fell from 42% in 1980 to 34% in 1992.5 The SMR2 form cannot be used to establish whether the same is true for mothers since information on smoking was not collected before 1992, but a decrease in maternal smoking may explain some of the remaining trend to higher birth weights.

The relation with socioeconomic status is more complex. Other factors which are potential causes of low birth weight, such as maternal diet and lifestyle, may also be associated with deprivation. Similarly, dietary and lifestyle factors in the last generation may explain the increase in maternal height. Our results suggest that improvements in socioeconomic status may not be seen fully as increased birth weight for two generations. A more comprehensive study would be required to ascertain the effect of this increase on the health of the adult population.

Acknowledgments

We thank the staff of the Information and Statistics Division of the NHS in Scotland for access to SMR2 data. Further details of this study are available from the authors in the form of a technical report.

Funding: None

Conflict of interest: None

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract