The impact of childhood injury and injury severity on school performance and high school completion in Australia: a matched population-based retrospective cohort study

Abstract

Background Exploring the impact of injury and injury severity on academic outcomes could assist to identify characteristics of young people likely to require learning support services. This study aims to compare scholastic performance and high school completion of young people hospitalised for an injury compared to young people not hospitalised for an injury by injury severity; and to examine factors influencing scholastic performance and school completion. Method A population-based matched case-comparison cohort study of young people aged ≤18 years hospitalised for an injury during 2005–2018 in New South Wales, Australia using linked birth, health, education and mortality records. The comparison cohort was matched on age, gender and residential postcode. Generalised linear mixed modelling examined risk of performance below the national minimum standard (NMS) on the National Assessment Plan for Literacy and Numeracy (NAPLAN) and generalised linear regression examined risk of not completing high school for injured young people compared to matched peers. Results Injured young people had a higher risk of not achieving the NMS compared to their matched peers for numeracy (ARR: 1.12; 95%CI 1.06–1.17), reading (ARR: 1.09; 95%CI 1.04–1.13), spelling (ARR: 1.13; 95%CI 1.09–1.18), grammar (ARR: 1.11; 95%CI 1.06–1.15), and writing (ARR: 1.07; 95%CI 1.04–1.11). As injury severity increased from minor to serious, the risk of not achieving the NMS generally increased for injured young people compared to matched peers. Injured young people had almost twice the risk of not completing high school at year 10 (ARR: 2.17; 95%CI 1.73–2.72), year 11 (ARR: 1.95; 95%CI 1.78–2.14) or year 12 (ARR: 1.93; 95%CI 1.78–2.08) compared to matched peers. Conclusions The identification of characteristics of young people most likely to encounter problems in the academic environment after sustaining an injury is important to facilitate the potential need for learning support. Assessing learning needs and monitoring return-to-school progress post-injury may aid identification of any ongoing learning support requirements.

The impact of childhood burns on academic performance: A matched population-based cohort study

“…A previous analysis that examined all injuries found that young people hospitalised for any type of injury demonstrated a higher risk of not achieving the NMS on both numeracy and reading NAPLAN assessments compared with matched peers who had not been hospitalised for an injury (32). This is in contrast to the current analysis where only young females hospitalised for a burn were identi ed as having a higher risk of not achieving the NMS for reading compared with peers.…”

Risk factors of injury among urban adolescent students

“…This is possibly due to factors such as increased physical activity, sports participation, or other school-related activities. 16,17 On a similar note, younger aged students also showed a highly significant association with occurrence of injury. However, no significant association was found between the occurrence of injury and monthly income, the education level of the father, the occupation of the mother, or the number of family members.…”

Impact of childhood burns on academic performance: a matched population-based cohort study

“…Adjusted relative risks (ARRs) and 95% CIs were generated. Variables were included in the model that had previously been associated with school performance5 8 22–25 and forward selection was used to sequentially add covariates to the models, with significance assessed using p values (p<0.05). The final model included burn status, NAPLAN grade (ie, 3, 5, 7 or 9), sex, comorbidity status (Y/N), LBOTE, socioeconomic status of residential area, highest level of education for any parent/guardian (ie, bachelor or higher degree or other), log of hospital LOS and school sector (ie, government, Catholic, independent).…”

“…ARR and 95% CIs were calculated. The selection of variables for inclusion in the models was informed by previous studies,5 8 22–25 and forward selection was used to sequentially add covariates, with significance assessed at p<0.05. The final models included burn status, sex, comorbidity status (Y/N), LBOTE, socioeconomic status of residential area, highest level of education for any parent/guardian and geographical location of residence.…”