Abstract
BACKGROUND
Declining neonatal mortality has increased the number of surviving newborns; however, in health systems without validated neonatal morbidity indicators, population-level patterns of diagnoses recorded at hospital discharge remain poorly characterised.
OBJECTIVE
To characterise national trends in neonatal mortality, survival among infants with extremely low and very low birth weight, and International Classification of Diseases 10th revision (ICD-10) diagnoses recorded at discharge among surviving newborns in Kazakhstan from 2014 to 2024.
METHODS
We conducted a nationwide retrospective population-based study using de-identified administrative health data covering all live births in Kazakhstan from 2014 to 2024. Temporal trends in neonatal mortality, survival among infants with extremely low and very low birth weight and recorded primary ICD-10 diagnoses at discharge were assessed.
RESULTS
Among 4 437 357 live births, neonatal mortality decreased from 6.2 to 4.0 per 1000 live births, whereas the proportion of surviving newborns with a recorded primary diagnosis at discharge increased from 4.9% to 16.7% (relative risk 3.41; 95% CI 3.36 to 3.46; p<0.001). By 2024, P91, P59 and P22 accounted for 88.4% of recorded primary diagnoses. The increase in P91 diagnoses occurred without a corresponding increase in the proportion of newborns weighing <2500 g.
CONCLUSION
Declining neonatal mortality was accompanied by an increase and a changing distribution of recorded diagnoses among surviving newborns. These data cannot determine individual post-discharge follow-up needs but may provide an initial population-level basis for planning subsequent clinical risk stratification after discharge.