ANALYSIS

Kazakhstan's newborn deaths fell by a third since 2010, but boys still die more

A Global Burden of Disease analysis found early-neonatal mortality in Kazakhstan dropped about 33% from 2010 to 2023 — yet boys kept dying at higher rates than girls, a gap that only recently began to close.

Early-neonatal (0–6 days) all-cause mortality rate per 100,000 person-years, Kazakhstan, by sex, 2010 vs 2023Male 2010: 32705; Female 2010: 23486; Male 2023: 19524; Female 2023: 1810502000040000Male 201032705Female 201023486Male 202319524Female 202318105
Early-neonatal (0–6 days) all-cause mortality rate per 100,000 person-years, Kazakhstan, by sex, 2010 vs 2023
GroupValue (value)
Male 201032705 (27966 to 38180)
Female 201023486 (20001 to 27303)
Male 202319524 (15357 to 24388)
Female 202318105 (14526 to 22641)
Early-neonatal (0–6 days) all-cause mortality rate per 100,000 person-years, Kazakhstan, by sex, 2010 vs 2023 GBD 2023 age-specific rates are expressed per 100,000 person-years within the 0–6 day interval and are not mortality probabilities per live birth. Bars show point estimates with 95% uncertainty intervals. Source: PLOS One

Newborn survival in Kazakhstan improved sharply over the 2010s, but the improvement did not fall evenly on boys and girls. A population-based analysis of Global Burden of Disease 2023 estimates, published in PLOS One on 8 October 2026, found that early-neonatal all-cause mortality — deaths in the first six days of life — dropped by roughly a third between 2010 and 2023, while boys continued to die at higher rates than girls throughout the period [s1].

The study, by researchers at Astana Medical University and Georgia Southern University, is a secondary analysis rather than new fieldwork. The authors pulled GBD 2023 modelled estimates of mortality, disability-adjusted life years (DALYs), years of life lost (YLLs) and years lived with disability (YLDs) for Kazakhstan from 2010 to 2023, split them into the early-neonatal (0–6 days) and late-neonatal (7–27 days) windows, stratified everything by sex, and fitted log-linear models to project the trends to 2030 [s1].

What the numbers show

In 2010 the combined early-neonatal mortality rate was 28,227 per 100,000 person-years within that six-day interval (95% uncertainty interval 25,303–31,856) [s1]. By 2023 it had fallen to 18,831 — an overall decline of about 33%, an estimated annual percent change of −3.96% (95% CI −5.39 to −2.51) [s1].

The late-neonatal period followed the same direction more gently. The combined rate eased from 4,397 per 100,000 person-years in 2010 (95% UI 3,913–4,927) to 3,482 in 2023, a reduction of roughly 21% and an annual change of −2.07% (95% CI −2.57 to −1.56) [s1]. In both windows, mortality was far heavier in the first six days than in the back three weeks of the neonatal month.

A sex gap that is only now closing

Across every year studied, boys carried the higher burden — higher mortality, higher YLLs and higher DALYs than girls in both neonatal windows [s1]. The early-neonatal gap was stark at the outset: in 2010 the male rate was 32,705 per 100,000 person-years against 23,486 for girls [s1]. By 2023 the male rate had fallen to 19,524, a 40.3% drop, while the female rate fell more modestly to 18,105, down 22.9% [s1].

Because boys started so much higher and fell faster, the two lines converged. The authors report that the male excess, though present in point estimates throughout, narrowed substantially and became statistically uncertain in the most recent years, with rate-ratio uncertainty intervals spanning or approaching unity [s1]. In plain terms: the data can no longer confidently say boys die at higher rates than girls in Kazakhstan's latest figures, even though the central estimates still point that way.

One measure moved differently. Years lived with disability were modestly higher among girls in the early-neonatal window and near-equal between the sexes later — but YLDs were tiny next to YLLs, which dominated the DALY totals [s1]. That tells you the neonatal burden in Kazakhstan is driven by deaths, not by surviving infants living with disability.

The caveats that matter

Two cautions sit around these figures. The first is technical but important: GBD age-specific rates are expressed per 100,000 person-years within the neonatal age interval and are not mortality probabilities per live birth [s1]. They are not directly comparable to the familiar "deaths per 1,000 live births" neonatal rate, and should not be read as such.

The second is the standard GBD limitation. These are modelled estimates, only as reliable as the vital-registration and survey inputs feeding them; the authors used national mortality statistics to check external consistency, and the work received no specific funding [s1]. Projections to 2030 from a log-linear model assume the past trend continues, and the authors note their out-of-sample checks did not show the model clearly beating simpler alternatives.

What to watch

The model projects early-neonatal mortality continuing down to roughly 12,054 per 100,000 person-years by 2030, with the male and female estimates still close together [s1]. The question the next few years of data will settle is whether the narrowing male–female gap holds or reopens — and whether the gains concentrate, as they should, on the first six days of life, where the deaths still cluster. For a middle-income health system, a one-third cut in early-neonatal mortality inside 13 years is real progress; the sex gap is the unfinished part.

Sources

  1. Sex differences in early and late neonatal all-cause mortality and disease burden in Kazakhstan, 2010–2023: A population-based analysis of Global Burden of Disease estimates — PLOS One , October 8, 2026
Related coverage