Neonatal Mortality Disparities by Gestational Age in European Countries

Abstract

Importance: There are wide disparities in neonatal mortality rates (NMRs, deaths <28 days of life after live birth per 1000 live births) between countries in Europe, indicating potential for improvement. Comparing country-specific patterns of births and deaths with countries with low mortality rates can facilitate the development of effective intervention strategies.

Objective: To investigate how these disparities are associated with the distribution of gestational age (GA) and GA-specific mortality rates.

Design, setting, and participants: This was a cross-sectional study of all live births in 14 participating European countries using routine data compiled by the Euro-Peristat Network. Live births with a GA of 22 weeks or higher from 2015 to 2020 were included. Data were analyzed from May to October 2023.

Exposures: GA at birth.

Main outcomes and measures: The study investigated excess neonatal mortality, defined as a rate difference relative to the pooled rate in the 3 countries with the lowest NMRs (Norway, Sweden, and Finland; hereafter termed the top 3). The Kitagawa method was used to divide this excess into the proportion explained by the GA distribution of births and by GA-specific mortality rates. A sensitivity analysis was conducted among births 24 weeks' GA or greater.

Results: There were 35 094 neonatal deaths among 15 123 428 live births for an overall NMR of 2.32 per 1000. The pooled NMR in the top 3 was 1.44 per 1000 (1937 of 1 342 528). Excess neonatal mortality compared with the top 3 ranged from 0.17 per 1000 in the Czech Republic to 1.82 per 1000 in Romania. Excess deaths were predominantly concentrated among births less than 28 weeks' GA (57.6% overall). Full-term births represented 22.7% of the excess deaths in Belgium, 17.8% in France, 40.6% in Romania and 17.3% in the United Kingdom. Heterogeneous patterns were observed when partitioning excess mortality into the proportion associated with the GA distribution vs GA-specific mortality. For example, these proportions were 9.2% and 90.8% in France, 58.4% and 41.6% in the United Kingdom, and 92.9% and 7.1% in Austria, respectively. These associations remained stable after removing births under 24 weeks' GA in most, but not all, countries.

Conclusions and relevance: This cohort study of 14 European countries found wide NMR disparities with varying patterns by GA. This knowledge is important for developing effective strategies to reduce neonatal mortality.

Figure.. Number of Neonatal Deaths Considered in Excess by Gestational Age in Comparison With Sweden, Norway, and Finland

The shaded area indicates the total number of neonatal deaths considered in excess, in each country and at each gestational age compared with Sweden, Norway, and Finland. The total number of excess deaths is the sum of the number of excess deaths attributed to gestational age distribution and the number of excess deaths attributed to gestational age-specific mortality, represented respectively by the light blue and dark blue bars. If a bar indicates a negative value, it means that the proportion of births at that gestational age is favorable compared with Sweden, Norway, and Finland (light blue bar), or that the mortality rate at that gestational age is favorable compared with Sweden, Norway, and Finland (dark blue bar). This figure displays the 4 countries with the highest number of births during the study period. The other countries included in the study can be found in eFigure 1 and eFigure 2 in Supplement 1.