Stumped by the Hump: The Curious Rise and Fall of Norwegian Birthweights, 1991-2007

Abstract

Background: There was a distinct rise in mean birthweights in Norway starting in 1991 that plateaued in 1996-2002 and then declined to previous levels. We investigated whether these changes corresponded to trends in neonatal mortality or other birthweight-associated pregnancy outcomes. We also explored known predictors of birthweight and examined whether these could explain the birthweight trends.

Methods: We calculated mean birthweight for all live births in Norway in each year from 1982 to 2016, together with annual neonatal mortality and proportion of infants born preterm, or with low Apgar score. We stratified mean birthweight over time by factors including parity, gestational age, and Scandinavian versus non-Scandinavian origin of mother, to test robustness of the pattern. In addition, we used multivariable linear regression to obtain adjusted estimates for mean birthweight per year.

Results: A 50-g rise and fall of mean birthweights during a 25-year period was not accompanied by corresponding changes in neonatal mortality, preterm births, or Apgar scores. The distinct hump pattern was restricted to term births and was not apparent among infants of mothers born outside Scandinavia. We saw a similar pattern for Sweden but not Finland. Known predictors of birthweight (such as parity, mode of onset of delivery, and marital status) did not explain the hump.

Conclusions: A distinct temporal hump in mean birthweight among Norwegian term births had no obvious explanations. Furthermore, these fluctuations in birthweight were not associated indirectly with adverse outcomes in measures of infant health.

FIGURE 1.

Mean birthweight for all live-born children in Norway by year of birth from 1982 to 2016. Vertical lines show the 95% confidence intervals.

FIGURE 2.

Proportion (% of all live-born children) with (A) neonatal mortality, (B) preterm births, and (C) newborns with Apgar score less than seven at five min, in Norway in the years 1982 to 2016. Note the use of log-scale for the y-axes.

FIGURE 3.

Mean birthweight by year of birth in Norway from 1982 to 2016, for preterm births (<37 gestational weeks) and term live-born children by year of birth.

FIGURE 4.

Mean birthweight (in grams) in live-born term births by year of birth by following groups. (A), Parity; (B) season of birth (Summer: June, July, August; Winter: December, January, February; Spring: March, April, May; Fall: September, October, November); and (C) mother born in Scandinavia or other country.

FIGURE 5.

Crude and adjusted change in mean birthweight in live-born children born at term to Scandinavian-born mothers by year of birth (1982 as reference). Adjusted estimates obtained by linear regression were adjusted for parity (primipara vs. multipara), maternal age continuous and squared, gestational age continuos and squared, onset of delivery (spontaneous vs. medically induced vs. cesarean section), marital status (married, cohabiting, nonmarried/single, divorced/separated/widowed and other/unknown), and congenital anomalies (no vs. yes). The shaded band shows 95% confidence intervals for adjusted estimates.

FIGURE 6.

Mean birthweight for live-born children in Norway (full line) and Finland (long dashes), all births in Sweden (short dashes), by year of birth. Data from Finland are available from 1987 onward (The Finnish Medical Birth Registry was established in 1987).