Child injury in the preschool years

Child Injury in the Preschool Years
01 Dec 2020


The aim of this research was to understand how social and physical factors in combination, and over time, affect a preschool child’s risk of experiencing injury.

The specific research objectives were:

  1. To investigate how combinations of situations and multiple events act across the life-course to either protect a child or, alternatively place them at risk of isolated/repeated injuries requiring medical attention.
  2. To determine how these life-course determinants of childhood injury vary between population subgroups in particular for Māori and Pacific children.


This study analysed data from the longitudinal GUiNZ cohort study. Access to external GUiNZ datasets described below was approved by the Data Access Committee in April 2019 (Reference: DA 18_1011). Data from the antenatal (AN), 6 week (6w), 9 month (9m), 2 year (2Y) and 4½ year (4.5Y) data collection waves (DCW) were included in this study as follows:

  • Mother questionnaire (AN, 9m, 2Y & 4.5Y)
  • Linkage to health data (6w)
  • Child proxy (mother-completed) questionnaire (9m, 2Y & 4.5Y)
  • Child observation (interviewer-completed) questionnaire (4.5Y)

The raw injury outcome data obtained at 2Y and 4.5Y were binary (at least one injury vs. no injury; at least one hospitalisation due to injury vs. no hospitalisation due to injury) and ordinal (number of injuries). Different combinations of derived injury variables were explored to determine the primary outcome variable for the study, see Injury outcomes (page 20). Analyses were carried out using SPSS and Stata. At both univariate and multivariable levels, significance was defined as p<0.05.

Key Results

We found no associations between injury and the presence of home safety features (including working smoke alarms, driveway, pool and boundary fencing, safe power outlets, safe hot water temperature, securely stored potential poisons).

Five factors were significantly associated with injury outcome (see Table for OR and 95% Confidence Intervals), after adjusting for covariates*.

  • Children in high nurturing environments were less likely to be in the high injury risk group than those in lower nurturing environments (significant across all maternal ethnicities except Asian).

These factors were associated with an increased likelihood of a child being in the high injury risk group:

  • Living in a high-need environment (also significant for those in areas of high deprivation, children with Māori, Pacific and European mothers).
  • High rate of household risk factors.
  • High rate of family risk factors (also significant for children with Pacific or
    European mothers, and girls).
  • Living in high stress households (also significant for those in areas of
    low/medium deprivation, and children with Pacific mothers).
Page last modified: 13 Oct 2023